Introduction

Although randomized controlled trials have shown that artificial intelligence (AI) for polyp detection improve adenoma detection rate (ADR) in colonoscopy, it is unknown how implementation of AI changes endoscopists’ behavior on standard colonoscopy (colonoscopy without AI assistance). We aimed at answering this important clinical question.

Aims & Methods

We conducted a multicenter, observational study at four Polish endoscopy centers. These centers implemented AI tools for polyp detection in the end of 2021, after which colonoscopies were done either with or without AI assistance at endoscopists’ discretion. We evaluated the change in the quality of all the diagnostic colonoscopies between September of 2021 and February of 2022, by comparing two different phases: 3 months before and after the AI implementation in clinical practice. The primary outcome measure was ADR of standard colonoscopy. To investigate the independent factors that affected ADR, multivariable logistic regression analysis was also done.

Results

A total of 1,460 patients underwent standard colonoscopy before and after the AI implementation (795 and 665, respectively; mean age 57.1± 15.5 with female accounting for 58.6%). ADR of standard colonoscopy significantly decreased from 28.4% (226/795; 95% confidence interval [CI] 25.29- 31.56) to 22.1% (147/665; 95% CI 18.95 - 25.26; P=.006) after the AI implementation, corresponding to a 6.3% absolute reduction (95%CI 6.2 - 6.5) and a 22.2% relative reduction (95% CI 21.8 - 22.7). In multivariable regression analysis showed that AI implementation in clinical practice (risk ratio 0.774; 95% CI 0.64-0.93), sex (risk ratio 1.33; 95% CI 1.14-1.54) and age (risk ratio 1.535; 95% CI 1.29-1.79) were the only independent factors that affected ADR.

Conclusion

Implementation of AI in clinical practice negatively affected the ADR during standard colonoscopy, raising a question of deskilling effect brought about by an advanced AI technology.

Disclosure

Prateek Sharma: Consultant: Boston Scientific, Olympus Inc; Grant support: US Endoscopy,
Medtronic, Fujifilm, Ironwood, Cosmo pharmaceuticals, Erbe
Michał Filip Kamiński: Consultant: Olympus, Erbe speaker: Boston Scientific, Ipsen,
Recordati; Grant support: Olympus
Yuichi Mori - Olympus: consultation, speaking fees, device loan
Cessare Hassan is a consultant for Alpha-Sigma, Fujifilm, Medtronic, Norgine, Olympus and Pentax.
Rest of authors: nothing to declare.

Introduction

Pancreatic cancer remains a devastating disease with a poor prognosis, and liquid biopsy utilizing minimally invasive biofluids such as blood and duodenal juice has emerged as a promising approach for early detection. Our group has previously accumulated data on plasma- and duodenal fluid-based liquid biopsy, demonstrating potential clinical utility but also highlighting limitations in diagnostic sensitivity. This study aims to establish a robust cancer diagnostic and risk assessment method using plasma, duodenal juice, and pancreatic juice, while also charactering molecular abnormalities in resected tumors and surrounding tissues. To develop a novel clinical diagnostic framework, we comprehensively evaluated to cancer-associated genetic mutations in biofluids and compared these finding with tissue samples obtained via biopsy or surgical resection.

Aims & Methods

This multi-institutional study, initiated in 2021, focuses on developing a liquid biopsy platform for pancreatic cancer detection. Samples were collected as follows: plasma (n=224), duodenal juice (n=102), and pancreatic juice (n=85). Plasma was collected in 10 mL PAXgene Blood ccfDNA tubes (BD); cell-free DNA (cfDNA) was purified and concentrated to a 30 µL eluate. Duodenal fluid was collected through the endoscopic channel, and pancreatic juice was aspirated directly from the pancreatic duct using a catheter during ERP examination. Both fluid were each mixed with a preservation solution validated in a previous study. Digital PCR (dPCR) was performed to screen for KRAS G12/G13 and A59/Q61 mutations, followed by panel-based sequencing. For this study, we designed a Duplex UMI-based pancreatic cancer gene panel targeting 16 genes, including KRAS, TP53, CDKN2A, and SMAD4, to enable high sensitive detection of tumor-specific mutations.

Results

DPCR screening detected KRAS mutations in 75% of pancreatic juice samples and 50% of duodenal juice samples. Validation of the Duplex UMI-based target sequencing panel in a subset of samples (n=16) demonstrated that a minimum input of 20 ng of liquid-derived, non-genomic DNA was necessary to achieve concordance with mutations identified in matched tissue samples. Inputs below this threshold resulted in significantly reduced coverage and uniformity. Notably, duodenal and pancreatic juice exhibited superior sensitivity, reliably detecting tissue-confirmed mutations down to a minimum variant allele frequency (VAF) of 0.1%. Pancreatic juice not only captured a diverse spectrum of genetic alterations across multiple lesions but also identified mutations absent in corresponding tissue samples, suggesting its potential for detecting subclinical molecular changes. These findings underscore the complementary roles of different biofluids in capturing tumor heterogeneity.

Conclusion

Pancreatic juice-derived DNA shows exceptional promise for the accurate prediction of tumorous lesions, offering a powerful tool to prompt further diagnostic evaluation even in the absence of histological evidence. This approach could substantially enhance early detection and risk stratification in pancreatic cancer. Our ongoing study involves comprehensive analysis of all collected samples using the designed 16-gene panel. These efforts aim to establish liquid biopsy as a central component of pancreatic cancer diagnostics, supporting personalized treatment strategy and improving patient outcomes.

Disclosure

Y.O., K.T. and Y.M. receive funding form Hitachi High-Tech Corporation

Introduction

Endoscopic Submucosal Dissection (ESD) is the best endoscopic technique that enables en bloc resection of superficial colorectal neoplasms.Despite its effectiveness, ESD is not very widespread in Western countries due to technical complexity, long procedure times, and a steep learning curve. These barriers have created a significant gap in adoption and outcomes between Eastern and Western centers. Underwater ESD (U-ESD) has been introduced to simplify the procedure. This study aimed to evaluate the effectiveness of U-ESD versus conventional gas-based ESD (G-ESD) in a Western center, focusing on procedural efficiency and safety.

Aims & Methods

This is a single-center, prospective randomized controlled trial conducted at Fondazione Campus Bio-Medico in Rome between December 2024 and April 2025.Forty patients with non-pedunculated colorectal lesions were randomly assigned in a 1:1 ratio to undergo either Underwater ESD (U-ESD) or conventional ESD with insufflation of Carbon Dioxide (CO2) (G-ESD).The primary outcome was total procedure time. Secondary outcomes included dissection speed, en bloc and R0 resection rates, use of hemostatic forceps, and the incidence of adverse events such as delayed bleeding, perforation, and post-ESD coagulation syndrome (PECS).

Results

A total of 40 patients referred to our endoscopic unit for ESD between December 2024 and April 2025. The mean size of the colorectal lesions was 57.4 mm and 40.7 mm in the U-ESD and G-ESD groups. The location of the lesions was in the rectum in 15% and 55% of the U-ESD and C-ESD groups, respectively; in the right side of the colon in 35% and 20% of the U-ESD and G-ESD groups, respectively. The morphology of LST was classified as LST-nongranular in 20 % and 25% of the U-ESD and G-ESD groups, respectively; as LST granular in 60% and 35% of the U-ESD and G-ESD groups, respectively. Fibrosis (grade 1 and 2) was present in 45% and 30% of the U-ESD and G-ESD groups, respectively. The mean total procedure time was shorter in the U-ESD group compared with the G-ESD group (53.3 ±33 minutes vs 78.7 ± 25.6 minutes; mean difference,–25.4 minutes; P=0.009).Mean dissection speed was faster in the U-ESD group compared with the G-ESD group (42 ± 26.3 mm2/min vs 13.3 ± 8 mm2/min; mean difference, 28.7 mm2/min; P= 0.001). En bloc and R0 resection rates in both groups were 100%. The use of a hemostatic forceps was significantly lower in U-ESD compared G-ESD (40% vs 90%, P= 0.001). Delayed bleeding occurred in only one patient in G-ESD group and was managed successfully by endoscopic clipping. No perforations occurred in either group. PECS was noted in only one patient in the G-ESD group.

Conclusion

UESD has shown a significantly shorter procedure time and faster dissection speed compared with GESD. These both empower the possibility of considering U-ESD as the preferred colorectal ESD method for colorectal lesions.

Introduction

Metabolic dysfunction-associated steatohepatitis (MASH) increases the risk of cardiovascular (CV) events and liver fibrosis. Semaglutide, a glucagon-like peptide-1 (GLP-1) receptor agonist, reduces CV risk associated with overweight and obesity in the absence of diabetes and has a profound effect on weight reduction and MASH resolution. Management of MASH may benefit from a deeper understanding of the mechanistic basis of semaglutide-induced MASH resolution.

Aims & Methods

This study used a multi-omics approach to investigate the effects of semaglutide on the hepatic microenvironment and fibrosis development in the context of semaglutide-induced MASH resolution. GLP-1 receptor expression was assessed using RNA in situ hybridization and immunohistochemistry in human liver tissue and immunohistochemistry in liver tissue from DIO‑NASH and CDA-HFD mice. The effect of semaglutide treatment on liver fibrosis was also assessed using liver transcriptome analysis in both of these mouse models of MASH. Proteomic analyses of 4,979 proteins using the SomaScan Platform were conducted on serum samples from 293 patients with MASH who received semaglutide or placebo in a phase 2b study (NCT02970942), and an independent cohort of 141 patients with MASH and 89 healthy volunteers.

Results

GLP-1 receptor mRNA or protein expression were not detected in human or mouse liver tissue. However, semaglutide improved histological markers of fibrosis and expression of genes involved in collagen turnover in two distinct mouse models of MASH; one that phenocopies the metabolic-associated features of human disease and another that exhibits non-abdominal obesity-induced advanced steatohepatitis and rapid fibrosis. In semaglutide-treated patients with MASH, proteomics identified 72 proteins significantly associated with MASH resolution after 72 weeks of treatment. These 72 proteins were also identified as differentially expressed in patients with MASH who were not treated for MASH, relative to healthy individuals. In semaglutide-treated patients with MASH, the abundance of the 72 proteins was similarly altered; however, semaglutide reverted the altered disease profile to that seen in non-pharmacologically treated healthy individuals. In contrast, changes in the levels of the 72 proteins in patients who received placebo were negligible. Notably, changes in the abundance of 26 of the 72 proteins were significant even after correction for weight loss. Although most of these 26 proteins are associated with metabolic functions, several are implicated in fibrosis-related pathways and CV disease (e.g. ADAMTSL2 and ACY1), and hepatocellular carcinoma (e.g. SERPINC1).

Conclusion

Based on circulating proteomic profiling, semaglutide-induced MASH resolution is associated with changes in multiple pathogenic pathways even after correction for weight loss. Semaglutide appears to re-balance the expression levels of disease-associated proteins in MASH to levels found in healthy individuals.

Disclosure

MJ is an employee of Novo Nordisk A/S. JN is an employee of Novo Nordisk A/S. MSK is an employee of Novo Nordisk A/S. KA is an employee of Novo Nordisk A/S. KMB is an employee of Novo Nordisk A/S. EB served as a consultant or advisory board member for Boehringer Ingelheim, Gilead Sciences, Intercept, Merck, Novo Nordisk, Pfizer, ProSciento. Grants from Gilead Sciences for fatty liver research, Speaker for Gilead Sciences, Intercept, Merck, Novo Nordisk, and Pfizer. KC served as a consultant for Aligos Therapeutics, Arrowhead, AstraZeneca, 89Bio, BMS, Eli Lilly & Co., Merck, Novo Nordisk, Prosciento, Sagimet Biosciences, Siemens, and Terns Pharma. Received research support (University of Florida; Principal Investigator) from Boehringer Ingelheim, Echosens, Inventiva, Labcorp, and Perspectum. EDG is an employee of Novo Nordisk A/S. MG is an employee of Novo Nordisk A/S. LLG is an employee of Novo Nordisk A/S. LMH is an employee of Novo Nordisk A/S. RL is a consultant to Aardvark Therapeutics, Altimmune, Anylam/Regeneron, Amgen, Arrowhead Pharmaceuticals, AstraZeneca, Bristol-Myer Squibb, CohBar, Eli Lilly, Galmed, Gilead, Glympse bio, Hightide, Inipharma, Intercept, Inventiva, Ionis, Janssen Inc., Madrigal, Metacrine, Inc., NGM Biopharmaceuticals, Novartis, Novo Nordisk, Merck, Pfizer, Sagimet, Theratechnologies, 89 bio, Terns Pharmaceuticals and Viking Therapeutics. His institutions received research grants from Arrowhead Pharmaceuticals, Astrazeneca, Boehringer-Ingelheim, Bristol-Myers Squibb, Eli Lilly, Galectin Therapeutics, Galmed Pharmaceuticals, Gilead, Hanmi, Intercept, Inventiva, Ionis, Janssen, Madrigal Pharmaceuticals, Merck, NGM Biopharmaceuticals, Novo Nordisk, Pfizer, Sonic Incytes and Terns Pharmaceuticals. Co-founder of LipoNexus Inc. GM is an employee of Novo Nordisk A/S. PNN is a consultant for Novo Nordisk, Boehringer Ingelheim, Gilead, Intercept, Poxel Pharmaceuticals, Bristol Myers Squibb (BMS), Pfizer, Sun Pharma, Madrigal and GSK. He also reports honoraria as a speaker for Novo Nordisk and AiCME; support for attending meetings from for Novo Nordisk; and participation on an advisory board for Novo Nordisk, Boehringer Ingelheim, Gilead, Intercept, Poxel Pharmaceuticals, BMS, Pfizer, Sun Pharma, Madrigal, and GSK. LMN is an employee of Novo Nordisk A/S. MSP is an employee of Novo Nordisk A/S. VR received consulting fees from Novo-Nordisk, Sagimet, Madrigal, Enyo, Poxel, Northsea, Intercept Pharmaceuticals, Prosciento. Research grants (to institution) from Gilead Sciences and Intercept Pharmaceuticals. A-SS is an employee of Novo Nordisk A/S. VWW served as a consultant or advisory board member for AbbVie, Boehringer Ingelheim, Echosens, Gilead Sciences, Intercept, Inventiva, Novo Nordisk, Pfizer, Sagimet Biosciences, and TARGET PharmaSolutions. Has received a research grant from Gilead Sciences. Speaker for Abbott, AbbVie, Gilead Sciences, Novo Nordisk, and Unilab. Co-founder of Illuminatio Medical Technology Limited. Has received travel support from AbbVie and Gilead Sciences. QMA is coordinator of the IMI2 Liver Investigation: Testing Marker Utility in Steatohepatitis (LITMUS) consortium, which is funded by the European Union Horizon 2020 programme and the European Federation of Pharmaceutical Industries and Associations (EFPIA). This multi-stakeholder consortium includes industry partners. He reports research grant funding from AstraZeneca, Boehringer Ingelheim, and Intercept; consultancy on behalf of Newcastle University for Alimentiv, Akero, AstraZeneca, Axcella, 89Bio, Boehringer Ingelheim, Bristol Myers Squibb, Galmed, Genfit, Genentech, Gilead, GlaxoSmithKline, Hanmi, HistoIndex, Intercept, Inventiva, Ionis, IQVIA, Janssen, Madrigal, Medpace, Merck, NGMBio, Novartis, Novo Nordisk, PathAI, Pfizer, Prosciento, Poxel, Resolution Therapeutics, Roche, Ridgeline Therapeutics, RTI, Shionogi, and Terns; speaker fees/honoraria from Fishawack, Integritas Communications, Kenes, Novo Nordisk, Madrigal, Medscape, and Springer Healthcare; and royalties from Elsevier. LBK is an employee of Novo Nordisk A/S.