UEG's educational platform for gastroenterology education bringing recordings, courses, guidelines, and more together in one place.

UEG Week Berlin 2025 Recordings Gut Guide online

Rate for better recommendations!

Your feedback helps us show you more of what you like and less of what you dislike!

More of this!

We will show you more similar content.

Not for me!

We won't suggest this again.

Bookmark!

Save this content for later!

New to myUEG? Create an account! Log In with myUEG account

UEG Poster

Standard Poster

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

Free for all myUEG account holders. Your access level is set automatically based on your occupation. Medical professionals get full access to all content. If you are a non-medical user, you can only access UEG Week content from congresses you attended.

Not sure what you can access? Learn more about account types.

Introduction

Lumen-apposing metal stents (LAMSs) have significantly advanced endoscopic ultrasound (EUS)-guided treatment of pancreatic fluid collections (PFCs). However, their impact on clinical outcomes in real-world settings remains incompletely understood.

Aims & Methods

Using the Diagnosis Procedure Combination database—a nationwide inpatient database in Japan where LAMSs were approved in 2018—we identified 5,885 patients who underwent EUS-guided treatment for PFCs at 550 hospitals between 2010 and 2023. We examined the association between treatment period (before vs. after LAMS implementation at each hospital) and clinical outcomes. Multivariable logistic regression analyses were conducted to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for in-hospital mortality, bleeding, length of stay, and total healthcare costs, adjusting for potential confounders.

Results

Compared with patients treated in the pre-LAMS period, those treated in the post-LAMS period had a significantly higher risk of bleeding (adjusted OR, 1.56; 95% CI, 1.35–1.81). Additionally, the post-LAMS period was associated with an increase of approximately $3,300 (USD) in total healthcare costs, with an adjusted OR for being in the highest quartile of total costs of 1.55 (95% CI, 1.34–1.79). There was no significant difference in the risk of in-hospital mortality (adjusted OR, 1.21; 95% CI, 0.95–1.55) or prolonged hospitalisation (adjusted OR for top quartile, 0.98; 95% CI, 0.86–1.12).

Conclusion

The introduction of LAMSs was associated with an increased risk of bleeding and higher healthcare costs, without significant improvements in mortality or hospital stay duration. These findings suggest that routine use of LAMSs for EUS-guided PFC treatment should be reconsidered.

Disclosure

HI declares research funding from Boston Scientific Japan, Fujifilm, and Piolax Medical Devices; honoraria from Boston Scientific Japan, Century Medical, Create Medic, Fujifilm, Gadelius Medical, Hitachi Medical, Japan Lifeline, Kaneka, Kawasumi Laboratories, Olympus Medical, Piolax Medical Devices, Sumitomo Bakelite, UMIDAS, and Zeon Medical; and contributions from Boston Scientific Japan, Gadelius Medical, Japan Lifeline, and Zeon Medical. YN declares research funding from Boston Scientific Japan, Century Medical, Fujifilm, Gadelius Medical, HOYA Pentax Medical, Hitachi Medical, Kaneka, and Medico’s Hirata; and honoraria from Boston Scientific Japan, Fujifilm, Gadelius Medical, Hitachi Medical, J-MIT, Medico’s Hirata, and Olympus Medical. This work was not supported by any of those companies. No other conflicts of interest exist. The other authors declare that they have no conflicts of interest.

REAL-WORLD NATIONWIDE OUTCOMES OF LUMEN-APPOSING METAL STENTS FOR EUS-GUIDED MANAGEMENT OF PANCREATIC FLUID COLLECTIONS

Tomotaka Saito 1, Tsuyoshi Hamada 2, Nobuaki Michihata 3, Yuya Kimura 1, Takuji Iwashita 4, Hiroki Matsui 1, Hideyuki Shiomi 5, Mamoru Takenaka 6, Kiyohide Fushimi 7, Ichiro Yasuda 8, Hiroyuki Isayama 9, Hideo Yasunaga 1, Yousuke Nakai 10

1 The University of Tokyo, Tokyo, Japan

2 Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

3 Cancer Prevention Center, Chiba, Japan

4 Shiga University of Medical Science, Shiga, Japan

5 Hyogo Medical University, Hyogo, Japan

6 Kinki University, Osaka sayama, Japan

7 Tokyo Medical and Dental University, Tokyo, Japan

8 Gifu University 1st Dept. of Internal Medicine, Gifu, Japan

9 Juntendo University, Tokyo, Japan

10 Tokyo Women's Medical University, Tokyo, Japan

Conference

UEG Week Berlin 2025

Topics

Pancreas

Submission format

Abstract

Session

A scope, a needle and a stent: All you need in treating pancreatitis (Posters)

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

MACHINE LEARNING-DRIVEN PROGNOSTIC MODELING IN PANCREATIC DUCTAL ADENOCARCINOMA USING CLINICOPATHOLOGIC AND CAF EXPRESSION DATA

Not sure what you can access? Learn more about account types.

Introduction

Pancreatic Ductal Adenocarcinoma (PDAC) remains one of the deadliest solid tumors, largely due to its late-stage diagnosis, limited therapeutic responsiveness, and poor overall prognosis. Despite ongoing advances in imaging techniques and molecular characterization, effective tools for early prognostic stratification are still lacking. Recent research suggests that integrating machine learning (ML) with clinicopathologic variables and tumor microenvironment components, particularly cancer-associated fibroblasts (CAFs, may significantly enhance predictive accuracy. Leveraging these computational methods could facilitate earlier intervention and guide personalized treatment decisions, offering a potential paradigm shift in PDAC management.

Aims & Methods

This study aims to develop and validate machine learning models that incorporate clinicopathologic parameters and cancer-associated fibroblast (CAF) expression profiles to enhance prognostic stratification in patients with PDAC. A retrospective analysis was conducted on a cohort of 76 patients diagnosed with PDAC through endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB). Data collected included alcohol consumption, smoking status, type II diabetes, C-reactive protein (CRP), albumin, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), CRP/albumin ratio, carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA 19-9), CA 125, and expression levels of CAF markers S100A4 and alpha-smooth muscle actin (α-SMA). The primary endpoint was the classification of tumor stage into early (T1–T2) and advanced (T3–T4) categories. Four machine learning algorithms—Latent Dirichlet Allocation (LDA), Quadratic Discriminant Analysis (QDA), Logistic Regression (LR), and K-Nearest Neighbors (K-NN)—were applied and their performance evaluated using independent validation datasets.

Results

All four machine learning models demonstrated strong predictive performance, with accuracy scores ranging from 0.77 to 1.00. Quadratic Discriminant Analysis (QDA) emerged as the most accurate classifier. Latent Dirichlet Allocation (LDA) achieved 91% accuracy and identified the neutrophil-to-lymphocyte ratio (NLR) as a key feature associated with early-stage (T1–T2) disease. In contrast, CRP/albumin ratio and PLR were more predictive of advanced-stage (T3–T4) tumors. Logistic Regression (LR), with an 83% accuracy rate, also highlighted PLR, CRP/albumin, and CAF markers (S100A4 and α-SMA) as significant contributors to advanced staging. These findings underscore the potential of integrating clinicopathologic and stromal biomarkers into ML-driven prognostic models.

Conclusion

Integrating machine learning with clinicopathologic and cancer-associated fibroblast (CAF) expression data offers a promising strategy for improving prognostic assessment in PDAC. By accurately distinguishing early- from advanced-stage disease, these models can support more personalized risk stratification and treatment planning. The results underscore the value of AI-driven approaches in precision oncology, with potential to enhance clinical decision-making and ultimately improve outcomes for patients with pancreatic cancer.

MACHINE LEARNING-DRIVEN PROGNOSTIC MODELING IN PANCREATIC DUCTAL ADENOCARCINOMA USING CLINICOPATHOLOGIC AND CAF EXPRESSION DATA

Bogdan Ungureanu 1, Ioana Streata 2, Mirela Florescu 2, Cristian Lungulescu 2, ION ROGOVEANU 2, Daniel Pirici 2, Lucian Florescu 2, Adina Turcu-Stiolica 2

1 University of Medicine and Pharmacy of Craiova, Craiova, Romania

2 University of Medicine and Pharmacy of Craiova Romania, Craiova, Romania

Event

UEG Week Berlin 2025

Topics

Digestive Oncology Pancreas

Submission format

Abstract

Session

Prediction and prognosis of pre-malignant and malignant pancreatic lesions

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

Not sure what you can access? Learn more about account types.

Visceral thrombosis in pancreatic disorders: Incidence and treatment algorithms

Peter Hegyi 1

1 University of Pecs, Pecs, Hungary

Event

UEG Week Berlin 2025

Topics

Nurses Pancreas Radiology & Imaging

Session

Hot topics in European gastroenterology

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

Not sure what you can access? Learn more about account types.

Prediction of Clinical Events in Patients with MASLD Based on Baseline and Serial Changes in MRE

Takashi Kobayashi 1

1 Yokohama City University Graduate School of Medicine, Yokohama, Japan

Event

UEG Week Berlin 2025

Topics

Hepatobiliary

Session

Rising Stars from Europe and Japan: Metabolic liver diseases

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

Not sure what you can access? Learn more about account types.

Can immune cells cure autoimmune disease?

Andreas Mackensen 1

1 University Hospital of Erlangen, Erlangen, Germany

Event

UEG Week Berlin 2025

Topics

Immunology

Session

Opening Plenary

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

AN IMMUNE RESPONSIVE TUMOR MICROENVIRONMENT IMPRINTS INTO PBMCS AND ALLOWS TO PREDICT PATIENT OUTCOME USING MACHINE LEARNING IN ADVANCED PANCREATIC CANCER: LESSONS FROM THE PREDICT TRIAL

Not sure what you can access? Learn more about account types.

Introduction

Prognosis in advanced pancreatic ductal adenocarcinoma (aPDAC) is particularly poor. Few patients benefit from second-line chemotherapy, while most experience side effects, highlighting the need for biomarkers. The PREDICT trial examined in aPDAC patients whether first-line time-to-treatment failure (TTF1) predicts second-line treatment failure (TTF2) but found no association. In our translational study we hypothesized that the responsiveness of the tumor immune microenvironment (TiME) might determine second-line therapy outcome and examined whether treatment-naive tissue immune features are also reflected in the peripheral blood after one line of chemotherapy.

Aims & Methods

In the PREDICT trial, patients received 5-FU/LV plus nanoliposomal irinotecan second-line after gemcitabine/nab-paclitaxel (Gem-nabPac) first-line therapy. We stratified PREDICT patients by shortest vs. longest TTF2. Treatment-naive primary tumor tissues from 20 patients were analyzed by NanoString transcriptomic profiling and multiplexed immunohistochemistry to characterize differences in the TiME. Peripheral blood mononuclear cells (PBMCs) were obtained immediately before second-line therapy from 82 patients and analyzed using multiparameter flow cytometry with an automated gating procedure and gene expression profiling via RT-qPCR. A feature selection and machine learning pipeline integrated PBMC and clinical data to derive a minimal predictor for TTF2 outcome and prognosis that was validated across different treatment regimens on an independent external cohort of 30 patients with aPDAC.

Results

Treatment-naive tumors of patients in the long-TTF2 group exhibited an immune-active (“hot”) TiME with CXCR3⁺CD8⁺ cytotoxic T cell infiltration into the cancer cell compartment via CXCR3-dependant inflammatory chemokine signaling. Analysis of PBMCs obtained directly before second-line therapy showed that these immune features were sustained over one line of treatment and reflected in PBMCs. A PBMC-based machine learning model (“TTF2Pred”) identified a 7-feature immune signature that accurately distinguished long vs. short TTF2 patient groups outperforming multivariate clinical data or CA19-9 benchmark models. TTF2Pred robustness was confirmed by external validation and showed the best overall performance post-Gem-nabPac first-line treatment. Among the seven TTF2Pred features, long-TTF2 patients exhibited more circulating CXCR3⁺ activated T cells and plasmacytoid dendritic cells, while short-TTF2 patients had more circulating platelet–leukocyte (platelet-monocyte, platelet-neutrophil) aggregates and exhausted T cells.

Conclusion

PDAC patients with an immune-active, treatment-naïve TiME have a better outcome on second-line therapy, and these tissue characteristics are reflected in circulating PBMCs after one line of chemotherapy. A parsimonious circulating PBMC-based signature offers a minimally invasive predictor of second-line outcome that could inform personalized therapy.