Endoscopy
Prophylactic clip closure after endoscopic submucosal dissection of large flat and sessile colorectal polyps: the EPOC trial
Miyakawa A, Tamaru Y, Mizumoto T, et al. Prophylactic clip closure after endoscopic submucosal dissection of large flat and sessile colorectal polyps: a multicentre randomised controlled trial (EPOC trial). Gut 2025; 74: 1814-1820. doi: 10.1136/gutjnl-2024-334463.
Endoscopic submucosal dissection (ESD) has increasingly been utilised for resection of superficial neoplastic colorectal lesions larger than ≥20mm in size. However, ESD requires technical expertise and comes with risks such as bleeding and perforation. Prophylactic clip closure has shown reduction in delayed bleeding for endoscopic mucosal resection of large and proximal colon lesions but its efficacy in ESD has remained unclear. Miyakawa et al. are the first group to conduct a randomised controlled trial (RCT) investigating delayed bleeding rates as the primary outcome after colorectal ESD.
This was a prospective parallel two-arm, open-label RCT conducted across four Japanese centres. A total of 300 patients were electronically randomised in a 1:1 ratio to either closure post-ESD with Endoclips or no closure. The primary endpoint was 30-day delayed bleed rate with either mild delayed (haematochezia not requiring haemostasis or transfusion) or severe (requiring haemostasis or transfusion).
In the intention-to-treat analysis, 19/40 cases of delayed bleeding cases underwent emergency endoscopy with 15/19 cases requiring haemostasis. All cases requiring intervention were treated using clips and no blood transfusions were required. There were significantly lower delayed bleeding rates in the closure (6.7%, n=10/150) compared to the non-closure group (20.1%, n=30/149; OR (odds ratio) 0.28, p<0.001) as well as severe delayed bleeding rates (1.3%, n=2/150 vs 8.7% 13/149, OR 0.14, p=0.003). There were no differences between groups for post ESD coagulation syndrome nor perforation rates. Per protocol analysis showed similar findings. Multivariate regression analysis showed prophylactic clip closure was a significant preventative factor of delayed bleeding whilst hyperlipidaemia, rectal location, and specimen size >870mm2 were risk factors.
In summary, this is the first RCT that has set delayed bleeding rates as the primary outcome measure after colorectal ESD. They demonstrate that prophylactic clip closure decreases rates of delayed bleeding and this should be performed following colorectal ESD.
Prevention of rebleeding after primary haemostasis using haemostatic powder in non-variceal upper gastrointestinal bleeding
Shin J, Cha B, Hong J, et al. Prevention of rebleeding after primary haemostasis using haemostatic powder in non-variceal upper gastrointestinal bleeding: a multicentre randomised controlled trial. Gut 2025; 74: 1821-1827. doi: 10.1136/gutjnl-2024-332459
Non-variceal UGIB (upper gastrointestinal bleeding) still carries substantial early rebleed risk despite successful endoscopic haemostasis and PPIs (proton pump inhibitors). In this multicentre (3 Korean centres), single-blind RCT, Shin et al., randomised adults with high-risk NVUGIB (non-variceal upper gastrointestinal bleeding) lesions (Forrest Ia–IIb) who had achieved primary haemostasis to receive either adjunctive Nexpowder (muco-adhesive haemostatic powder) or no further endoscopic therapy. In the modified ITT (intention to treat) cohort (n=341), adjunctive powder significantly reduced 72-hour rebleeding (2.9% vs. 11.3%; p=0.005) and 30-day rebleeding (7.0% vs. 18.8%; p=0.002); effects were consistent in the peptic-ulcer subgroup. No device-related adverse events were reported. Benefit appeared independent of initial modality (thermal/clips/injection). Limitations include single-country setting, relatively few events and a fragility index of 5, so replication in broader populations is warranted. Nonetheless, this is the first trial since the PPI era to show that an additional endoscopic intervention after successful haemostasis can further reduce rebleeding.
There are implications for UK practice (versus current guidance): the British Society of Gastroenterology’s (BSG) acute UGIB care bundle emphasises early risk assessment, endoscopy within 24 hours, effective endoscopic therapy, and high-dose IV (intravenous) PPI for 72 hours; it does not recommend routine second-look endoscopy and has not previously endorsed prophylactic topical agents after haemostasis. This RCT data suggest that, where available, adding a muco-adhesive powder after standard therapy could be considered for selected high-risk peptic-ulcer bleeds to cut early rebleeding—aligning with the bundles’ aim to prevent deterioration without increasing procedure-related harm. Pending confirmatory trials, UK units might reasonably (1) consider adjunct powder for Forrest Ia–IIa ulcers after definitive therapy, particularly where rebleed risk is high (anti-thrombotics, large ulcers, fibrotic base) (2) maintain established care: early endoscopy, combination therapy when indicated, and high-dose IV PPI, and (3) audit outcomes to inform local protocols and any future BSG updates.
Pancreas
Histone lactylation-driven feedback loop modulates cholesterol-linked immunosuppression in pancreatic cancer
Yang J, Yu X, Xiao M, et al. Histone lactylation-driven feedback loop modulates cholesterol-linked immunosuppression in pancreatic cancer. Gut 2025; 74: 1859-1872. doi: 10.1136/gutjnl-2024-334361.
Pancreatic cancer is notoriously resistant to immunotherapy. One reason may be its unique metabolic environment — packed with lactate from cancer cell glycolysis.
This study by Yang et al. explored how lactate might do more than acidify the tumour microenvironment: it can reprogramme gene expression via histone lactylation, directly shaping immune suppression.
Yang et al. investigated whether histone lactylation — specifically at histone H3K18 — links tumour metabolism to immune evasion. Using patient samples, cell lines and mouse models, they combined metabolomic, epigenetic, and immunological analyses to dissect this pathway.
They found that pancreatic tumours had elevated histone H3K18 lactylation (H3K18la), tightly associated with increased expression of ACAT2 (Acetyl-CoA acetyltransferase 2), an enzyme in the acetoacetyl-CoA and cholesterol synthesis pathway. This created a positive feedback loop: lactate boosted H3K18la, which upregulated ACAT2, further enhancing lactate and cholesterol metabolism. Tumour cells with high ACAT2/H3K18la levels secreted factors that blunted dendritic- and T-cell activation. Knocking down ACAT2 reversed this, restoring immune infiltration and slowing tumour growth in mice.
The study uncovers a metabolic-epigenetic-immune circuit where lactate acts as a signalling molecule that locks pancreatic cancer into an immunosuppressive state.
This work reframes lactate — long seen as metabolic waste — as an epigenetic messenger with immune consequences. Targeting ACAT2 or histone lactylation could represent a novel way to “re-energise” the immune system in one of the most therapy-resistant cancers.
Hepatology
Hepatitis B surface antigen level identifies patients with inactive chronic hepatitis B from Asia with HCC risk below surveillance threshold
Tseng T-C, Huang S-C, Pan M-H et al. Hepatitis B surface antigen level identifies patients with inactive chronic hepatitis B from Asia with HCC risk below surveillance threshold. Gut 2025: 74: 1896–1904. doi:10.1136/gutjnl-2025-334911.
Chronic hepatitis B (CHB) infection is a well-recognised risk factor for the development of hepatocellular carcinoma (HCC). However, given the significant resources required to provide HCC surveillance to the heterogenous population of patients with CHB, there is a clinical need to identify population subgroups for whom surveillance may not be warranted, especially in resource-constrained healthcare systems.
Tseng et al., proposed that hepatitis B surface antigen (HBsAg) levels, which have been previously associated with varying clinical outcomes, could be used to more effectively risk-stratify patients for HCC surveillance. Tseng et al., studied the serological results of a total of 2674 patients from the ERADICATE-B and REVEAL-HBV cohorts from Taiwan, who had inactive HBV, were not cirrhotic and were predominantly (99%) not treated. Their primary study outcome development was HCC acquisition, with a median follow up period of 26.3 years.
A total of 76 patients developed an HCC within the study period. Among patients with a HBsAg <100 IU/mL, annual HCC incidence was 0.08% (95% CI 0.05-0.13%), compared with 0.14% (95% CI 0.10-0.21%) and 0.17% (95% CI 0.12-0.24%) among patients with HBsAg of 100-999 IU/mL and ≥1000 IU/mL respectively. The results remained significant when the cohort was adjusted for age, sex and HBV DNA levels. Further analysis of the patient group with HBsAg <100 IU/mL showed a similar HCC risk when compared to patients with a HBsAg <10 IU/mL and 10-99 IU/mL.
EASL (European Association for the Study of the Liver) guidelines recommend HCC surveillance for all HBsAg-positive, non-cirrhotic patients at intermediate or high risk of HCC. The results from this study would suggest that patients with CHB with a HBsAg <100 IU/mL have a comparable HCC risk to patients who do not have chronic viral hepatitis. These findings could support a more targeted and resource efficient HCC surveillance approach.
Gut Microbiota
Food for thought: cognitive decline in obesity may be ameliorated by fasting-related changes in microbiota
Mela V, Heras V, Iesmantaite M, et al. Microbiota fasting-related changes ameliorate cognitive decline in obesity and boost ex vivo microglial function through the gut-brain axis. Gut 2025; 74: 1828-1846. doi: 10.1136/gutjnl-2025-335353
People with obesity have chronic low-level inflammation driven in part by activation of macrophage populations in white adipose tissue. In the brain, macrophage-like resident immune cells, called microglia, can become dysfunctional in the presence of low-level inflammation. Despite its relative distance, the gut microbiome is increasingly linked with brain function via the microbiota-gut-brain axis. Changes to dietary composition and timing can modify the composition and function of gut microbiota, associated with markers of health and inflammation.
In this study, Mela et al. sought to examine the impact of dietary intervention on cognitive function in people with obesity. Ninety-six participants with obesity were randomised to receive Mediterranean diet, alternative-day fasting (ADF) diet or ketogenic (keto) diet. After three months of diet, there were subtle improvements in cognitive performance (independent of weight change), most consistently in the ADF group. The ADF group also had a lower inflammatory burden and more functional peripheral blood microglia-like cells.
Mela et al. looked towards the microbiota as a mediator of diet on cognitive function. Of interest, there was no improvement in the richness and diversity of microbiota following dietary intervention. However, pre-diet faecal exosomes (packages of protein, RNA and metabolites secreted from microbiota and host cells) impaired function of human microglia-like cells. When transplanted into mice, faecal exosomes from the ADF group improved short- and long-term cognitive function, associated with decreased neuroinflammation and a healthier microglial phenotype, compared to pre-diet faecal exosomes. Compared with the pre-diet group, faeces from the ADF group had an increase in metabolites known to have a role in modulating neuroinflammation, such as riboflavin.
These data raise the possibility that diet-related changes to the microbiota can modulate neuroinflammation, linked with cognitive function. Microbiota modulation remains a tractable therapeutic target that requires further research.
Reviewers
Dr Michael Yulong Wu, Gastroenterology Advanced Trainee, Port Macquarie Base Hospital and Lismore Base Hospital, New South Wales Health, New South Wales, Australia
Dr Mohamed Elseragy, ST6 Gastroenterology Registrar, St James’s Hospital, Leeds Teaching Hospitals NHS Trust, Leeds, UK
Dr Muhammad Waleed Khan, IMT-3, Ysbyty Gwynedd Hospital, Betsi Cadwaladr University Health Board, Bangor, UK
Dr Rachel Perry, Gastroenterology and Hepatology ST5, Nottingham University Hospitals NHS Trust, Nottingham, UK
Dr Tom Butler, NIHR Academic Clinical Lecturer, University of Manchester and Salford Royal Hospital, Salford, UK
BSG GUT Highlights is edited by Dr Philip J Smith, Honorary Consultant Luminal Gastroenterologist, Royal Liverpool Hospital, Liverpool University Hospitals NHS Foundation Trust, Mount Vernon, Liverpool, UK. Dr Smith is the Digital and Education Editor of GUT, Editor in chief of Frontline Gastroenterology and Associate Editor of BMJ Open Gastroenterology.
Funding
The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Provenance and peer review: Not commissioned; internally peer reviewed.