UEG Week 2020 Oral Presentations

doi:10.1177/2050640620927344

. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344

UEG Week 2020 Oral Presentations

PMCID: PMC8939486 PMID: 33043828

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.1

OP001 Identification of Emp2 Gene as a New Therapeutic Target to Limit Aiec Colonization in Cd Patients

M Chervy ^1,^✉, N Barnich ¹, J Denizot ¹

Introduction

A particular pathotype of Escherichia coli abnormally colonizes the intestinal mucosa of Chron's disease (CD) patients, the adherent-invasive Escherichia coli (AIEC) and most of the studies performed on AIEC support the fact that AIEC play an important role in the etiology of CD, besides genetic and environmental factors.

However, others factors such as epigenetic modifications seem involved in the pathogenesis of CD. Various epigenetic alterations have been observed in patients such as a global lower expression of enzymes involved in the deacetylation of histones, the histone deacetylases (HDAC), compared to healthy controls. Based on these data, we investigated whether AIEC bacteria are able to modulate the epigenomic landscape of the host cell, favoring their own invasion, in order to identify new therapeutic targets to prevent AIEC colonization in CD patients.

Aims & Methods

Our aims were:

(1)
to investigate the consequences of AIEC infection on histones post-translational modifications in intestinal epithelial cells (IECs) in culture using western blot analysis;
(2)
to assess the role of HDAC in the AIEC invasion process by studying their activity during the course of infection and by evaluating the AIEC invasion ability in Caco-2 cells treated with a specific pan-HDAC inhibitor (SAHA) and in cells transfected with a siRNA targeting HDAC1. Finally, we aimed at identifying:
(3)
additional human proteins used by AIEC to favor invasion of IECs by studying the HDAC regulated genes through a RNA-seq analysis in cells silenced for HDAC1.

The genes potentially involved in the invasion process have then been silenced with specific siRNA in order to propose new therapeutic target for CD patients.

Results

AIEC infection significantly increased H3K9 acetylation level, which was associated to a global decrease in HDAC activity in Caco-2 cells. Chemical inhibition of HDAC as well as specific silencing of HDAC1 resulted in increased invasion of cells by AIEC demonstrating the crucial role of HDAC1 in the control of bacterial invasion. We identified HDAC1 regulated genes by performing a RNA-seq analysis in cells silenced for HDAC1. Particularly, 149 genes were up-regulated while 199 genes were downregulated upon HDAC1 silencing.

Among the up-regulated genes, we focused on 3 genes which could be involved in the better ability of AIEC to invade IECs in a HDAC1-deficient context: RDX and INPPL1, involved in ruffle formation, and EMP2 (epithelial membrane protein), coding for a protein regulating the cell membrane composition. Specific silencing of these genes in Caco-2 cells revealed that only EMP2 silencing decreased the invasion ability of diverse AIEC strains while a better ability of AIEC to invade Caco-2 cells has been observed in cells over-expressing this gene, demonstrating the involvement of EMP2 in the AIEC invasion process.

Conclusion

This work demonstrates that AIEC infection modulate the epigenetic landscape of host cell through the decrease of HDAC1 activity, leading to gene expression changes favoring the AIEC infection process. Targeting HDAC1-regulated genes, such as EMP2, could represent new interesting therapeutic approach to limit AIEC colonization in CD patients carrying these bacteria.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.2

OP002 Host-Genetics, Dysbiosis and Clinical History Explains Fecal Metabolic Alterations in Patients with Inflammatory Bowel Disease

Vila A Vich ^1,^2,^✉, S Hu ^1,², L Bolte ^1,², V Collij ^1,², BH Jansen ¹, RAA Ruigrok ^1,², G Abu-Ali ³, C Giallourakis ³, J Schneider ³, J Parkinson ³, A Al Garawi ³, A Kurilshikov ², R Gacesa ^1,², J Fu ^2,⁴, A Zhernakova ², RK Weersma ¹

Introduction

In recent years gastrointestinal dysbiosis has been suggested as an important factor in the pathogenesis of inflammatory bowel diseases (IBD). The decrease in bacterial richness as well as changes in the abundance of several bacterial species have been described in patients with IBD, however, the functional and clinical implications of these alterations are still not clear. The study of fecal metabolites is essential for understanding the relations between host and gut microbes, and can potentially lead to the discovery of new diagnostics and microbiome directed therapies.

Aims & Methods

In this study we characterized alterations in the gut microbial metabolism of patients with IBD. We aimed to identify potential new biomarkers for the disease as well as to improve our knowledge on the host-microbiota interaction, considering factors like life-style and clinical history. in short, we have explored the relation between host genetics, gut microbiota and fecal metabolites composition of 500 patients with IBD and 255 population controls. We used Metagenomic shotgun sequencing to profile the microbiota composition, and untargeted metabolomics platform (Metabolon INC., USA) was used to identify the levels of 1684 fecal metabolites including eight short chain fatty acids. Whole exome sequencing and GSA Illumina chip were used to characterize host's genetics. Associations between multiple data layers were performed using multivariate linear regression correcting for age, sex and technical covariates such as sample storage time and day of the experiment. Significant associations were considered after Bonferroni multiple testing adjustment.

Results

We found that the levels of 597 fecal metabolites were altered in patients with Crohn's disease (CD) and 209 in patients with ulcerative colitis (UC) as compared to the population controls (adjusted p-value< 0.05). The level of metabolites related to the sphingolipid synthesis was increased in both CD and UC samples, while fatty acid metabolites were decreased. Interestingly, metabolic profiles could discriminate IBD samples reasonably well from non-IBD samples (Random Forest classifier, AUC= 0.81-0.89). Within IBD patients, the resection of the ileocecal valve explained a significant variation on the levels of 83 metabolites. Moreover, we observe a high correlation between the microbial composition and the metabolic profile. 968 metabolites were related to at least the abundance of 1 bacterial species. For example, tryptamine was correlated to the abundance of the tryptophan converter Ruminococcus gnavus (p-value= 1.17e-21) and riboflavin to Faecalibacterium prausnitzii abundance (p-value=0.0004). Dietary habits explained variation in the levels of 104 metabolites significantly. Furthermore, we found that 3 host genetic signals associated to metabolites, including the association between NAT2 gene and coffee metabolites (p-value= 0.008).

Conclusion

Our study illustrates the complex relationship between host, gut microbiota and metabolisms. in patients with IBD, gut dysbiosis is also translated into the alteration of the fecal metabolic profile. We demonstrate that metabolite levels can be used to distinguish IBD from non-IBD individuals, as a potential novel biomarker for disease diagnosis. Moreover, we observed a strong correlation between various metabolites and bacterial species which, after further validation, could lead to new therapeutic approaches.

Disclosure

RKW received unrestricted grants from Takeda, Johnson and Johnson, Tramedico, Ferring.and acted as consultant for Takeda. This work is partially supported by Takeda Pharmaceutical Company Limited.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.3

OP003 Dose-Dependent Differential Effects of Vedolizumab Therapy on Adhesion of Regulatory and Effector T Cells

E Becker ^1,^✉, M Wiendl ¹, M Dedden ¹, A Schulz-Kuhnt ¹, I Atreya ¹, R Atreya ¹, MF Neurath ¹, S Zundler ¹

Introduction

Vedolizumab has emerged as an important pillar of treatment in inflammatory bowel disease (IBD). However, for unknown reasons, not all patients respond to therapy. Clinical studies suggested decreased response rates with highest compared to medium exposure^1-3. Interestingly, vedolizumab has been shown to inhibit the homing of both regulatory (Treg) and effector T (Teff) cells and previous data from our group suggested different effect sizes in both populations⁴.

Aims & Methods

We hypothesized that the non-linear exposure-efficacy correlation might be explained by dose-dependent differential effects of vedolizumab on Treg and Teff homing. Therefore, we studied functional effects of different vedolizumab exposure levels on Treg and Teff cell trafficking.

The α4β7 expression on different human T cell subsets as well as the binding characteristics of vedolizumab to these cells at different exposure levels were analysed via flow cytometry. Functional effects of different vedolizumab concentrations on the adhesion of Treg and Teff to mucosal addressin cell adhesion molecule 1 (MAdCAM-1) were examined using dynamic adhesion and transmigration assays in vitro as well as in vivo homing assays in a humanized mouse model. The in vivo binding of vedolizumab to Treg and Teff in patients receiving therapy was quantified and correlated with vedolizumab serum levels. To investigate the mechanisms underlying differential vedolizumab binding to Treg and Teff, single cell RNA sequencing was performed.

Results

With an exposure of 10 μg/mL, vedolizumab preferentially bound to Teff compared to Treg. At an exposure level of 50 μg/mL we observed equal binding to both cell types. Consistently, at 10 μg/mL, dynamic adhesion of Treg to MAdCAM-1 was increased compared to Teff, but no difference was noted at 50 μg/mL. Additionally, a higher number of Treg compared to Teff were able to transmigrate in a MAdCAM-1-dependent manner at an exposure of 10 μg/mL vedolizumab. In vivo data from homing experiments in a humanized mouse model and IBD patients treated with vedolizumab support these observations. Single cell RNA sequencing of vedolizumab-binding compared with non-binding a4β7-expressing Treg and Teff revealed subsets of Treg with profound differences in integrin and chemokine receptor pathways consistent with “resistance” to medium ve-dolizumab concentrations.

Conclusion

Our findings support a dose-dependent differential binding of vedolizumab to different T cell subpopulations and suggest an optimal “therapeutic window” of exposure, in which effects on Teff predominate over Treg. This differential binding seems to be due to a4β7-expressing Treg subsets with particular trafficking properties impairing vedolizumab binding. While offering a potential explanation for earlier findings in dose-ranging studies, our data might lay the basis for the establishment of individualized dose optimization in IBD patients.

Disclosure

Sebastian Zundler has received honoraria from Takeda and Roche. Markus F. Neurath has served as an advisor for Pentax, Giuliani, MSD, Abbvie, Janssen, Takeda and Boehringer. Emily Becker has no disclosures.

References

1.Feagan B. G. et al. Treatment of ulcerative colitis with a humanized antibody to the a4β7 integrin. N. Engl. J. Med. 352, 2499–2507 (2005). [DOI] [PubMed] [Google Scholar]
2.Parikh A. et al. Vedolizumab for the treatment of active ulcerative colitis: A randomized controlled phase 2 dose-ranging study. Inflamm. Bowel Dis. 18, 1470–1479 (2012). [DOI] [PubMed] [Google Scholar]
3.Rosario M. et al. Exposure-efficacy Relationships for Vedolizumab Induction Therapy in Patients with Ulcerative Colitis or Crohn's Disease. J. Crohns. Colitis 11, 921–929 (2017). [DOI] [PubMed] [Google Scholar]
4.Fischer A. et al. Differential effects of a4β7 and GPR15 on homing of effector and regulatory T cells from patients with UC to the inflamed gut in vivo. Gut 65, 1642–1664 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.4

OP004 Systemic Iron Deficiency and Intestinal Inflammation Control Gut Mucosal Hif-1α and Tfrc Activation in Inflammatory Bowel Disease

AR Bourgonje ^1,^✉, Rosa R Fagundes ¹, S Hu ¹, R Barbieri ¹, BH Jansen ¹, T Blokzijl ¹, CT Taylor ², KN Faber ¹, G Dijkstra ¹

Introduction

Many patients with inflammatory bowel disease (IBD) exhibit extra-intestinal symptoms and anemia is the most common hematological manifestation affecting up to two thirds of patients. Anemia in IBD is considered to be the combined result of iron deficiency anemia (IDA) and anemia of chronic disease (ACD). IBD patients who experience IDA often receive iron supplementation, but treatment response is difficult to predict. Iron homeostasis is intimately associated with oxygen metabolism in the intestinal mucosa, which is orchestrated by the hypoxia-inducible factor-1α (HIF-1α) pathway. Here, iron (Fe²+) is a critical cofactor to the hydroxylase reaction that targets the HIF-1α protein to degradation in the presence of oxygen. Following this, the interplay between systemic iron availability and the HIF-1α pathway regulates oxygen sensing in the intestinal epithelium and may be postulated to cause IDA in patients with IBD. in this study, we aimed to determine the association between systemic iron status and mucosal activation of the HIF-1α pathway in patients with IBD.

Aims & Methods

Intestinal biopsies from patients with IBD (n=110) were collected from both ileum and colon and evaluated for inflammatory status. Serum laboratory parameters of iron metabolism (hemoglobin, MCV, free iron, ferritin, TYBC, transferrin, and transferrin saturation) close to the date of biopsy were used to subdivide patients into ‘healthy iron status’ (n=81) and ‘systemic IDA’ (n=29). Expression data of 20 different HIF-1α pathway genes (including HIF1A, EPAS1), HIF-hydroxylases (EGLN1, EGLN2, VHL and HIF1AN), and HIF-1α target genes (EGNL3, CA9, PDK1, SLC2A1, MUC3, TFF3, A2BAR, SLC29A1, ADK, HAMP, SLC11A2, HMOX1, TFRC and TF), some of which are involved in iron metabolism, was extracted from bulk RNA sequencing data of intestinal biopsies (n=165) and were correlated to iron status parameters using multivariable linear mixed models. Finally, IBD patient-derived ileal and colonic organoids were used to further validate findings on expression levels of HIF-1α-related genes.

Results

RNA sequencing profiling revealed significant differential expression of HIF-1α pathway genes for intestinal location (ileum vs. colon) and inflammatory status (inflamed vs. non-inflamed). More specifically, expression levels of TFRC, a HIF-1α target gene involved in iron absorption, were significantly increased in inflamed biopsies compared to controls (P < 0.001). Moreover, systemic IDA further enhanced TFRC and HIF1A expression levels. TFRC expression levels were significantly inversely associated with systemic free iron levels in inflamed tissue (P < 0.05), while the direction of this association shifted in the absence of inflammation. Finally, colonic organoids showed higher sensitivity to DMOG-induced activation of the HIF-1α pathway compared to ileal organoids.

Conclusion

Our findings reveal a cumulative induction of mucosal TFRC expression levels depending on both systemic iron levels and mucosal inflammatory status. Furthermore, our data indicate a location-specific response to hypoxia in the human gut. This may implicate that patients with IBD and systemic IDA may benefit from HIF-1α targeted therapy to improve their responsiveness to iron supplementation.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.5

OP005 Anti-Inflammatory Actions of Butyrate on Macrophages and Epithelial Cells in Intestinal Mucosa of Inflammatory Bowel Disease Patients

Venegas D Parada ^1,^2,^✉, MK De la Fuente ^1,³, T Blokzijl ², Y Cui ², D Simian ³, HJM Harmsen ⁴, R Quera ⁵, G Dijkstra ², KN Faber ², MA Hermoso ¹

Introduction

Butyrate is a short-chain fatty acid produced by colonic bacterial fermentation that regulates gut mucosal homeostasis. Butyrate acts either intracellularly after internalization by the monocarboxylate transporter 1 (MCT1), or through transmembrane signalling by activating the G-coupled protein receptor 109A (GPR109A). Inflammatory Bowel Diseases (IBD), e.g. ulcerative colitis (UC) and Crohn's disease (CD), are characterized by a dysregulated gut microbiome (dysbiosis), impaired butyrate production and an exacerbated immune response (with excessive monocyte/ M1 macrophage infiltration and activation) and epithelial damage. To date, the cellular distribution of MCT1 and GPR109A, and their regulation by butyrate and downstream effects on human monocytes, macrophages and the gut epithelium have not been clarified.

Aims & Methods

To determine the cellular distribution of MCT1 and GPR109A in intestinal mucosa of IBD patients and analysing the effect of butyrate on MCT1 and GPR109A expression, and mediators restoring mucosal homeostasis in inflammatory conditions. MCT1 and GPR109A expression in inflamed and non-inflamed mucosa of IBD patients (CD and UC) was analysed and quantified by immunohistochemistry and compared to healthy controls (HC) (n=10 each group). Intestinal biopsies (CD n=11; UC n=9; HC n=10) were treated ex vivo with butyrate and cytokine production was analysed by a cytometric bead array. Human peripheral blood monocytes, THP-1 macrophages and human colonic organoids were exposed to butyrate and analysed for expression of MCT1 (SLC16A1), GPR109A (HCAR2), and markers of inflammation (TNF-α, CD40, CD80) and epithelial barrier (Cldn3, Ocln) by qPCR, ELISA and/or flow cytometry. Statistical analysis was performed using GraphPad Prism software (significant differences p < 0.05).

Results

MCT1 and GPR109A expression was reduced in lamina propria macrophages of IBD patients, when compared to HC. Similarly, MCT1 levels are reduced in UC epithelia, when compared to HC and CD. in contrast, GPR109A levels were enhanced in UC epithelia. In vitro cytokine-exposure reduced SLC16A1 (MCT1) levels in monocytes/macrophages, which was reversed by butyrate and was accompanied suppression inflammatory markers (TNF-α, CD40, and CD80). Moreover, butyrate suppressed pro-inflammatory cytokine secretion in ex vivo-treated IBD mucosa and induced the expression of the tight junction component Cldn3 in colonic organoids.

Conclusion

In an inflammatory intestinal environment, butyrate restores SLC16A1 (MCT1) expression and exerts anti-inflammatory actions in human intestinal tissue and cultured monocytes/macrophages and colonic epithelial cells. This provides mechanistic support for the notion that IBD patients may benefit from promoting butyrate production in the gut.

Disclosure

Nothing to disclose

Acknowledgments

CONICYT Scholarship 21150517, FONDECYT 1170648

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.6

OP006 Synergizing Effect of Alcohol Consumption and Smoking on Severity and Complications in Acute Pancreatitis

A Szentesi ^1,^2,^✉, N Gede ¹, Z Gyömbér ², A Vincze ³, S Gódi ³, J Bajor ³, B Nemeth ², T Takács ², L Czakó ², F Izbéki ⁴, J Hamvas ⁵, M Varga ⁶, V Sallinen ⁷, M Macarie ⁸, I Torok ⁸, A Mickevicius ^10,⁹, Maldonado E Ramirez ¹¹, A Miseta ¹², T Nagy ¹², N Faluhelyi ¹³, P Kanizsai ¹⁴, D Pécsi ¹, P Varjú ¹, N Zádori ¹, PJ Hegyi ¹, A Parniczky ^1,⁵, P Hegyi, on behalf of the Hungarian Pancreatic Study Group^1,²

Introduction

Alcohol consumption and smoking have been found to be harmful to the pancreas and these addictions often go together. Our aim is to evaluate the independent and joint clinical effects of smoking and alcohol consumption habits in acute pancreatitis (AP).

Aims & Methods

Our aim is to evaluate the independent and joint clinical effects of smoking and alcohol consumption habits in acute pancreatitis (AP). 2536 adult AP patients from 30 centers were enrolled by the Hungarian Pancreatic Study Group. Four groups of patients were retrospectively formed: non-smoker-non-drinker (NS-ND), smoker-non-drinker (S-ND), non-smoker-drinker (NS-D) and smoker-drinker (S-D).

Results

1094 (43.1%) of the patients were NS-ND, 206 (8.1%) S-ND, 653 (25.7%) NS-D and 561 (22.1%) were S-D. The average age at the first episode of AP was lower in the S-ND, NS-D and S-D groups as compared to the NS-ND group (50.6±14.9, 58.1±15.6 and 47.6±12.7 years respectively vs. 62.0±17.9, p< 0.001). The male ratio was 34% in the ND-NS group, 51% in the S-ND, 74% in the NS-D and 85% in the S-D groups. Smoking in addition to drinking was associated with lower BMI (NS-D: 28.8±5.5 and S-D: 25.6±5.4, p< 0.001).

There are lower rate of mild and higher rate of moderately severe cases and local complications in the NS-D and S-D groups as compared to the NS-ND group of patients (moderately severe AP: NS-D: 23.1%, S-D: 27.8% vs. ND-NS: 21.1%; local complications: NS-D: 29.1%, S-D: 32.6% vs. ND-NS: 24.4%).

Drinking and smoking together elevate the rate of recurrent AP (NS-ND: 17.0%, NS-D: 22.9%, S-D: 27.1%).

Smoking together with drinking has a synergic moderating effect on on-admission amylase (S-D: 711 vs. NS-ND: 1275, p< 0.05) and C-reactive protein levels but have no effect on white blood cell count. Heavy drinking alone is associated with higher rate of moderately severe AP, fluid collection, necrosis and pseudocyst as compared to no drinking (28% vs. 21%, 28% vs. 21%, 10% vs. 8% and 10% vs. 7% respectively).

Conclusion

Drinking and smoking together is associated with the first episode of pancreatitis 15 years earlier and elevate the risk of recurrent AP. Drinking and smoking together and heavy drinking alone increase the rate of moderately severe AP and local complications. Education of patients on drinking and smoking cessation is extremely important.

Disclosure

Nothing to disclose

References

Apte M., Pirola R., and Wilson J. (2009). New insights into alcoholic pancreatitis and pancreatic cancer. J Gastroenterol Hepatol 24 Suppl 3, S51–56. doi: 10.1111/j.1440-1746.2009.06071.x. [DOI] [PubMed] [Google Scholar]
Banks P.A., Bollen T.L., Dervenis C., Gooszen H.G., Johnson C.D., Forsmark C.E., Vege S.S., and Wilcox C.M. (2016). Acute Pancreatitis. N Engl J Med 375(20), 1972–1981. doi: 10.1056/NEJMra1505202. [DOI] [PubMed] [Google Scholar]
Kaner E.F., Dickinson H.O., Lugea A., Gerloff A., Su H.Y., Xu Z., Go A., Hu C. et al. (2017). The Combination of Alcohol and Cigarette Smoke Induces Endoplasmic Reticulum Stress and Cell Death in Pancreatic Acinar Cells. Gastroenterology 153(6), 1674–1686. doi: 10.1053/j.gastro.2017.08.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
Maleth J., Balazs A., Pallagi P., Balla Z., Kui B., Katona M. et al. (2015). Alcohol disrupts levels and function of the cystic fibrosis transmembrane conductance regulator to promote development of pancreatitis. Gastroenterology 148(2), 427–439e416. doi: 10.1053/j.gastro.2014.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
Nordback I., Pelli H., Lappalainen-Lehto R., Jarvinen S., Raty S., and Sand J. (2009). The recurrence of acute alcohol-associated pancreatitis can be reduced: a randomized controlled trial. Gastroenterology 136(3), 848–855. doi: 10.1053/j.gastro.2008.11.044. [DOI] [PubMed] [Google Scholar]
Sahin-Toth M., and Hegyi P. (2017). Smoking and Drinking Synergize in Pancreatitis: Multiple Hits on Multiple Targets. Gastroenterology 153(6), 1479–1481. doi: 10.1053/j.gastro.2017.10.031. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.7

OP007 Postponed Or Immediate Drainage of Infected Necrotizing Pancreatitis (Pointer): A Multicenter Randomized Trial

L Boxhoorn ^1,^2,^✉, SM van Dijk ^2,³, J van Grinsven ^2,³, RC Verdonk ⁴, MA Boermeester ³, TL Bollen ⁵, SAW Bouwense ⁶, MJ Bruno ⁷, V Cappendijk ⁸, CHC Dejong ⁶, P van Duijvendijk ⁹, CHJ van Eijck ¹⁰, P Fockens ¹, MFG Francken ³, H van Goor ¹¹, M Hadithi ¹², ND Hallensleben ², JW Haveman ¹³, MAJM Jacobs ¹, JM Jansen ¹⁴, KP van Lienden ^5,¹⁵, ER Manusama ¹⁶, JSD Mieog ¹⁷, IQ Molenaar ¹⁸, VB Nieuwenhuijs ¹⁹, AC Poen ²⁰, JW Poley ⁷, M van de Poll ⁶, R Quispel ²¹, TE Römkens ²², MP Schwartz ²³, TC Seerden ²⁴, M Stommel ¹¹, JWA Straathof ²⁵, NG Venneman ²⁶, W van de Vrie ²⁷, BJM Witteman ²⁸, MGW Dijkgraaf ²⁹, HC van antvoort ^18,³⁰, MGH Besselink ³; Dutch Pancreatitis Study Group

¹Amsterdam UMC, Department of Gastroenterology and Hepatology, Amsterdam, Netherlands

²St. Antonius Hospital, Department of Research and Development, Nieuwegein, Netherlands

³Amsterdam UMC, Department of Surgery, Amsterdam, Netherlands

⁴St. Antonius Ziekenhuis, Department of Gastroenterology and Hepatology, Nieuwegein, Netherlands

⁵St. Antonius Hospital, Department of Radiology, Nieuwegein, Netherlands

⁶Maastricht University Medical Center, Department of Surgery, Maastricht, Netherlands

⁷Erasmus University Medical Center, Department of Gastroenterology and Hepatology, Rotterdam, Netherlands

⁸Jeroen Bosch Hospital, Department of Radiology, ‘s Hertogenbosch, Netherlands

⁹Gelre Ziekenhuizen, Department of Surgery, Apeldoorn, Netherlands

¹⁰Erasmus University Medical Center, Department of Surgery, Rotterdam, Netherlands

¹¹Radboud University Medical Center, Department of Surgery, Nijmegen, Netherlands

¹²Maasstad Hospital, Department of Gastroenterology and Hepatology, Rotterdam, Netherlands

¹³University Medical Center Groningen, Department of Surgery, Groningen, Netherlands

¹⁴Onze Lieve Vrouwe Gasthuis, Department of Gastroenterology and Hepatology, Amsterdam, Netherlands

¹⁵Amsterdam UMC, Department of Radiology, Amsterdam, Netherlands

¹⁶Medical Center Leeuwarden, Department of Surgery, Leeuwarden, Netherlands

¹⁷Leiden University Medical Center, Department of Surgery, Leiden, Netherlands

¹⁸University Medical Center Utrecht, Department of Surgery, Utrecht, Netherlands

¹⁹Isala Clinics, Department of Surgery, Zwolle, Netherlands

²⁰Isala Clinics, Department of Gastroenterology and Hepatology, Zwolle, Netherlands

²¹Reinier de Graaf Gasthuis, Department of Gastroenterology and Hepatology, Delft, Netherlands

²²Jeroen Bosch Hospital, Department of Gastroenterology and Hepatology, ‘s Hertogenbosch, Netherlands

²³Meander Medical Center, Department of Gastroenterology and Hepatology, Amersfoort, Netherlands

²⁴Amphia Hospital, Department of Gastroenterology and Hepatology, Breda, Netherlands

²⁵Máxima Medisch Centrum, Department of Gastroenterology and Hepatology, Veldhoven, Netherlands

²⁶Medisch Spectrum Twente, Department of Gastroenterology and Hepatology, Enschede, Netherlands

²⁷Albert Schweitzer Hospital, Department of Gastroenterology and Hepatology, Dordrecht, Netherlands

²⁸Gelderse Vallei Hospital, Department of Gastroenterology and Hepatology, Ede, Netherlands

²⁹Amsterdam UMC, Department of Epidemiology and Data Science, Amsterdam, Netherlands

³⁰St. Antonius Hospital, Department of Surgery, Nieuwegein, Netherlands

^✉

Contact E-Mail Address: l.boxhoorn@amsterdamumc.nl

Introduction

Necrotizing pancreatitis with secondary infection is a potentially lethal disease that generally requires invasive intervention. Although current treatment guidelines advise to postpone drainage of infected (peri-)pancreatic collections for several weeks to await full encapsulation, the optimal timing of intervention is still debated. Whereas encapsulation may facilitate safe image-guided percutaneous and endoscopic transluminal drainage, it may also cause unnecessary delay leading to clinical deterioration and prolonged hospital stay. Therefore, we conducted a multicenter randomized trial to determine whether immediate catheter drainage in patients with infected necrotizing pancreatitis is superior to postponed catheter drainage.

Aims & Methods

Patients with necrotizing pancreatitis were monitored in 22 Dutch hospitals and were randomized once secondary infection was diagnosed to immediate catheter drainage, within 24 hours after diagnosing secondary infection, or postponed catheter drainage. Postponed catheter drainage included the use of antibiotics and to await drainage until collections reached the stage of walled-off necrosis. in the study, both image-guided percutaneous as well as endoscopic transluminal drainage were permitted. The primary end point was the Comprehensive Complication Index (CCI), covering all complications during 6 months follow-up. Secondary end points included mortality, major complications, total number of interventions, and total length of intensive care and hospital stay during 6 months of follow-up. Subgroup analysis was performed for patients with organ failure at time of randomization. Outcomes were assessed by a blinded adjudication committee.

Results

We randomly assigned 104 patients to immediate catheter drainage (n = 55) or postponed catheter drainage (n = 49). Immediate catheter drainage was performed after a median of 22 days (IQR 19 - 30 days), and postponed catheter drainage after a median of 29 days (IQR 23 - 40 days) after onset of acute pancreatitis.

The primary end point CCI did not differ between the immediate and postponed drainage groups (median CCI 56.46 vs. 48.22, p = 0.92). No differences in secondary end points were found, including mortality (11% vs. 10%; RR 1.07, 95%CI 0.35 - 3.29, p = 1.00), new-onset organ failure (25% vs. 22%; RR 1.13, 95%CI 0.57 - 2.26, p = 0.82) or other major complications. The length of intensive care stay was equal in both groups (mean 12 days vs. 12 days, p = 0.76) and total hospital stay during 6 months follow-up did not differ significantly (mean 59 days vs. 51 days, p = 0.07). The median number of interventions in the immediate drainage group was significantly higher (median of 4 vs. 1 intervention, p < 0.001), as compared to the postponed drainage group. in the postponed drainage group, 19 patients (39%) were successfully treated with antibiotics alone, without the need for drainage or necrosectomy during 6 months of follow-up. in the predefined subgroup of patients with organ failure at randomization, there were also no differences in the primary end point (median CCI 79.77 vs. 90.57, p = 0.51).

Conclusion

Immediate catheter drainage in patients with infected nec-rotizing pancreatitis is not superior to the current practice of postponed catheter drainage. with a postponed catheter drainage strategy including antibiotic treatment, more than one-third of patients may be treated conservatively.

Disclosure

The POINTER trial is an investigator-initiated trial. Financial support was provided by Fonds NutsOhra (grant number 1404-004), the Netherlands, and the Amsterdam UMC, University of Amsterdam, the Netherlands. The sponsors have no influence on the design of the study, data collection, results or publication.

References

Van Grinsven J, Van Dijk S.M., Dijkgraaf M.G. et al. Postponed or immediate drainage of infected necrotizing pancreatitis (POINTER trial): Study protocol for a randomized controlled trial. Trials 2019; 20: 239. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.8

OP009 Prospective Trial Evaluating the Safety and Effectiveness of the Interscope Endorotor® Resection System for Direct Endoscopic Necrosectomy of Walled-of Pancreatic Necrosis (Endorotor Den Trial)

PMC Stassen ^1,^✉, PJF de Jonge ¹, AD Koch ¹, MJ Bruno ¹, A Trindade ², P Benias ², DV Sejpal ², UD Siddiqui ³, C Chapman ³, E Villa ³, B Tharian ⁴, S Inamdar ⁴, JH Hwang ⁵, M Barakat ⁵, I Andalib ⁶, M Gaidhane ⁶, A Sarkar ⁶, H Shahid ⁶, A Tyberg ⁶, KF Binmoeller ⁷, R Watson ⁷, A Nett ⁷, C Schlag ⁸, M Abdelhafez ⁸, M Friederich-Rust ⁹, A Schlachterman ¹⁰, A Chiang ¹⁰, D Loren ¹⁰, T Kowalski ¹⁰, M Kahaleh ⁶

Introduction

Debridement of infected walled-off pancreatic necrosis (WOPN) is indicated to treat ongoing infection and sepsis-related multiorgan failure. The lack of dedicated and effective accessories results in the need for time-consuming repetitive procedures. in the published literature, a mean of 4 endoscopic procedures is typically needed to clear the necrotic cavity using conventional accessories. The aim of this study was to evaluate the use of a new 3.1 mm flexible microdebrider catheter (the EndoRotor® powered endoscopic debridement system) to remove solid debris under direct endoscopic visualization.

Aims & Methods

10 international sites enrolled patients requiring direct endoscopic necrosectomy (DEN) in a prospective study. Cases with WOPN size of ≥6 mm and <22 mm, with >30% solid component based on computed tomography (CT), were included. Outcomes such as successful clearance of necrosis (≥70% removal of necrotic debris at 21 days), procedural time, number of procedures until resolution, adverse events, length of hospital stay and quality of life were included in analyses.

Results

Between November 2018 and August 2019, 30 patients (mean age 55 years, 60% male) underwent necrosectomy using the EndoRotor resection system. 15/30 (50%) achieved complete debridement in one session, and 21/30 (73%) achieved complete debridement after 2 sessions. No En-doRotor-associated adverse events were reported. One patient died during the follow up period due to shock and multi-organ failure, unrelated to the treatment with the EndoRotor. Mean time between LAMS or SEMS placement and debridement was 14 days. A median of 1.5 interventions (range 1-7) was required. Mean EndoRotor procedure time was 71 minutes (SD 37 minutes). Mean overall endoscopic procedure time was 117 (SD 50 minutes). Baseline necrotic debris was 69% (SD 20%) and the mean reduction of solid necrosis of 68% (SD 29%), 54% (SD 34%), 58% (SD 36%) and 34% (SD 29%) was achieved after the first, second, third and fourth procedure, respectively. At the 21-day follow-up, the mean reduction in necrosis volume from baseline was 90% (SD 19%). Average duration from the start of necrosectomy until discharge was 16 days (SD 27 days).

Conclusion

Direct endoscopic necrosectomy using the EndoRotor resection system seems to be a safe and effective treatment for patients with WOPN. Necrosectomy using the EndoRotor resection is associated with a lower number of endoscopies and shorter procedural time when compared to published literature.

Disclosure

This is an investigator initiated study financed by Interscope Inc.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.9

OP010 Protocolized Endoscopic Ultrasound-Guided Drainage of Walled-Off Pancreatic Necrosis with Lumen-Apposing Metal Stents: A Prospective Multicenter Trial

Dayyeh BK Abu ^1,^✉, V Chandrasekhara ¹, JJ Easler ², RJ Shah ³, AC Storm ¹, MJ Levy ¹, JA Martin ¹, BT Petersen ¹, S Wani ³, JA Peetermans ⁴, O Gjata ⁴, E McMullen ⁴, FF Willingham ⁵

Introduction

Walled-off necrosis (WON) is a severe complication of nec-rotizing pancreatitis. Endoscopic ultrasound (EUS)-guided drainage using double-pigtail plastic stents (DPPS) or lumen-apposing metal stents (LAMS) is now widely regarded as first-line treatment for WON patients with severe symptoms or infection. Prior clinical trials of endoscopic vs. surgical management of necrotizing pancreatitis reported high rates of serious adverse events (SAEs) including death.^1-3

Protocolized WON treatment using LAMS warrants further prospective study because LAMS’ wider lumen allows more effective through-the-stent direct endoscopic necrosectomy and potentially improved drainage of solid necrosis.

Aims & Methods

This Ethics Committee and FDA-approved study was a prospective, single arm, multicenter trial to demonstrate the safety and effectiveness of a LAMS (AXIOS™, Boston Scientific Corporation, Marlborough, MA; NCT03525808) to treat WON with ≥6cm diameter and >30% necrotic material, as estimated by EUS or CT/MRI imaging. Four academic US centers recruited patients with infected and/or symptomatic WON per the 2012 Revised Atlanta Classification. Patients had protocolized serial assessments 1-2 weeks apart, with endoscopic necrosectomy if WON size reduction was insufficient. Upon radiographic WON resolution to ≤ 3cm on CT or MRI, or by 60 days of LAMS indwell, the stent was removed and the patient was followed for 6 months. Primary endpoints were probability of radiographic WON resolution to ≤3cm by 60 days after LAMS placement and SAEs related to the LAMS or WON drainage procedure. Secondary end-points included technical success, mean time to radiographic WON resolution, recurrence of fluid collection by 6 months after LAMS removal, and SF-12 quality of life score.

Results

Forty patients were enrolled September 2018-March 2020. Mean age was 54.5 years, 26 (65%) were male, and the principal WON-related symptoms were abdominal pain in 25 (62.5%) and weight loss, lethargy or gastric outlet obstruction in 9 (22.5%) patients. LAMS diameter was 15mm in 15 (37.5%) and 20mm in 25 (62.5%) cases. Mean WON size was 15.0±5.6 cm with mean 53%±17% solid necrosis [range 30 to 95%]. Fourteen (35.0%) patients required parenteral or tube feeding and 27 (67.5%) were inpatients. Mean baseline SF-12 score was 35.1±19.5 (n=38).

All 40 patients had successful LAMS placement and removal. The probability of radiographic WON resolution by 60 days post-LAMS removal was 97.5% (95% CI, 86.8%, 99.9%) among all patients, without recurrence of fluid-collection among 20 patients having 6-month follow-up to date. 100% percent of the cohort achieved WON resolution with only 1 patient requiring adjunct percutaneous drainage beyond 60 days. Mean time to radiographic WON resolution was 34.1±16.8 days (range 4-100). 90% of the cohort required direct necrosectomy. The mean number of necrosec-tomies was 2.2±1.8. Mean SF-12 score at LAMS removal was improved to 59.2±16.8 (n=38). Three LAMS-related SAEs occurred in 3 patients (7.5%), including vancomycin-resistant enterococcal bacteremia, sepsis, and upper gastrointestinal bleeding. No surgical intervention or related mortality was reported in the cohort.

Conclusion

Expert endoscopists used LAMS for endoscopic treatment of WON with > 30% necrotic material utilizing a protocolized endoscopic necrosectomy approach achieved 100% resolution with a low rate of stent-related SAEs, no mortality or new-onset organ failure, and improved quality-of-life scores maintained to 6 months after LAMS removal.

Disclosure

BA: Research grant and consultation fee from Boston Scientific Corporation, research support from Medtronic, and education/lecture fees from Olympus. VC: Consultant, Interpace Diagnostics, Shareholder Nevakar Corporation. AS: Research support from Boston Scientific and consulting fees from ERBE. JE: Research grant and consulting fee from Boston Scientific. SW: Consultant for BSC, Medtronic, Interpace, and Cer-nostics. RS: Consultant and advisor board member BSC, Cook, Olympus. BP: Consultant Boston Scientific, Olympus, Pentax, GI Medical. JP, EM, OG: full-time employees of Boston Scientific Corporation. FW: reports research funding to the institution from Cancer Prevention Pharmaceuticals, CSA Medical, Boston Scientific, and Cook Medical. Others have no relevant disclosures.

References

1.Bakker O.J., van Santvoort H.C., van Brunschot S. et al. Endoscopic transgastric vs surgical necrosectomy for infected necrotizing pancreatitis: a randomized trial. JAMA 2012; 307: 1053–61. [DOI] [PubMed] [Google Scholar]
2.van Brunschot S., van Grinsven J., van Santvoort H.C. et al. Endoscopic or surgical step-up approach for infected necrotising pancreatitis: a multicentre randomised trial. Lancet 2018; 391: 51–58. [DOI] [PubMed] [Google Scholar]
3.Bang J.Y., Arnoletti J.P., Holt B.A. et al. An Endoscopic Transluminal Approach, Compared with Minimally Invasive Surgery, Reduces Complications and Costs for Patients with Necrotizing Pancreatitis. Gastroenterology 2019; 156: 1027–1040 e3. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.10

OP011 Risk of Serious Infections Associated with Vedolizumab Versus Tumor Necrosis Factor Antagonists in Patients with Inflammatory Bowel Disease: A Multinational, Population-Based Cohort Study

J Kirchgesner ^1,^2,^✉, RJ Desai ¹, L Beaugerie ², S Schneeweiss ¹, SC Kim ¹

Introduction

While the risk of serious infections associated with anti-TNF in inflammatory bowel disease (IBD) has been extensively assessed, the risk associated with vedolizumab remains uncertain. Its presumed gut-selectivity may decrease the risk of extra-intestinal immunosuppression-related adverse events, notably serious infections; however, concerns have been raised on the risk of gastrointestinal infections associated with vedolizumab.

Aims & Methods

The aim of this study was to assess the risk of serious infections associated with the use of vedolizumab versus anti-TNF in IBD patients, according to IBD subtype. Eligible patients were adult patients with Crohn's disease (CD) or ulcerative colitis (UC) in two nationwide US commercial insurance databases (MarketScan [2010-2017] and Optum [2010-2019]) and the French nationwide health insurance database (2010-2018). We identified initiators of vedolizumab or anti-TNF agents among anti-TNF naïve and experienced patients. Anti-TNF naïve and experienced patients were assessed separately to reduce confounding by disease duration and severity.

All infections requiring hospitalization were identified and classified according to infection sites: ([1] (upper and lower) respiratory tract; [2] gastrointestinal; [3] any other serious infections).

Up to 4 anti-TNF users were propensity score (PS) matched to vedolizumab users in each cohort, in order to control for confounders including demographics, infection risk factors, other comorbidities, and baseline IBD disease activity. Hazard ratios (HR) of serious infections were estimated using Cox regression in each cohort, which were then combined with a fixed-effects meta-analysis.

Results

We included a total of 10,421 vedolizumab (CD, 5434; UC, 4987) and 75,928 anti-TNF users (CD, 54,978; UC, 20,950) in the 3 data sources, 8768 vedolizumab initiators PS-matched to 35,424 anti-TNF users were analyzed. A total of 893 serious infections occurred during 37,725 person-years (PY) of follow-up (23.7 per 1000 PY). The risk of serious infections was not different between vedolizumab and anti-TNF in the overall IBD cohort (HR 0.95; 95%CI 0.79-1.13), while the risk was decreased with vedolizumab compared to anti-TNF in patients with UC (HR 0.68; 95%CI 0.50-0.93), but not CD (HR 1.10; 95%CI 0.87-1.38). in UC patients, a similar trend in the magnitude of risk reduction was observed for the risk of respiratory infections (HR 0.55; 95%CI 0.25-1.20) and infections excluding respiratory and gastrointestinal infections (HR 0.65; 95%CI 0.39-1.08). Findings were consistent after exclusion of gastrointestinal and mycobacterial infections (Table 1).

Table 1.

[Hazard ratios (95% CI) of serious infections in patients treated with vedolizumab versus anti-TNF,according to infection site]

	Hazard-ratio (95% CI)
	Serious infections, overall	Respiratory tract infections	Gastrointestinal infections	Other infections^*	Serious infections excluding gastrointestinal infections	Serious infections excluding mycobacterial infections
Inflammatory bowel disease	0.92 (0.63-1.34)	1.20 (0.91-1.57)	0.79 (0.59-1.05)	0.84 (0.67-1.05)	0.95 (0.79-1.13)
Crohn's disease	1.10 (0.87-1.38)	0.93 (0.58-1.51)	1.37 (0.96-1.97)	1.01 (0.70-1.44)	0.99 (0.74-1.32)	1.10 (0.87-1.38)
Ulcerative colitis	0.68 (0.50-0.93)	0.55 (0.25-1.20)	0.87 (0.55-1.36)	0.65 (0.39-1.08)	0.59 (0.39-0.90)	0.68 (0.50-0.92)

Open in a new tab

^*

Other infections include all non-respiratory and non-gastrointestinal infections.

In patients with UC, vedolizumab was associated with lower risk of serious infections after exclusion of gastrointestinal infections (HR, 0.59; 95%CI, 0.39-0.90).

Conclusion

Based on three large population-based cohorts of IBD patients including more than 10,000 patients exposed to vedolizumab, the risk of serious infections associated with vedolizumab versus anti-TNF varies according to IBD subtype. This risk was decreased in patients with UC, but not CD. These findings may help to clarify the optimal position of vedolizumab in the therapeutic management of IBD.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.11

OP012 90-Day Specific Readmission for Clostridium Difficile Infection After Hospitalization with an Inflammatory Bowel Disease Flare: Outcomes and Predictors

P Palacios ^1,^✉, M Salazar ², CR Simons-Linares ³, B Shen ⁴

Introduction

Patients with inflammatory bowel disease (IBD) have increased Clostridium difficile infection (CDI) rates and significantly higher morbidity and mortality compared to those without IBD. Specific readmissions rates for CDI and predictors after an IBD flare index admission have not yet been studied; and may worsen IBD outcomes and impose a significant health care burden. We aim to investigate specific readmissions for a CDI episode after a recent index hospitalization for an IBD flare.

Aims & Methods

The Nationwide Readmission Database (NRD) was queried for the year 2016. We collected data on hospital readmissions of 50,799 adults who were hospitalized for urgent IBD flare and discharged. Primary outcome was specific readmission rate for CDI within 90-days of discharge. Secondary outcomes were readmission rate of colonoscopic procedures, morbidity (mechanical ventilation, shock), and hospital economic burden. Independent risk factors for readmission were identified using Cox regression analysis.

Results

The 90-day specific readmission rate for CDI was 0.1% (n=477). Patients readmitted with CDI infection are more likely to have Parkinson's disease (1.1% vs. 0.002% P=0.01), BMI above 30 (19.0% vs. 9.9%; P< 0.01), to have malnutrition (23.6% vs. 11.1%; P< 0.01), to have Medicare (37.9% vs. 26.1; P< 0.01) as primary payer. They are less likely to have previous CDI (0.03% vs 2.8%; P=0.01), to undergo colonoscopy with intervention (12.8% vs. 22.1%; P< 0.01), colonoscopy with and without intervention (15.0% vs. 26.5%; P< 0.01). A total of 3,005 days were associated with readmission and the total health care in-hospital economic burden of readmission was $19.1 million (in charges) and $4.79 million (in costs). Independent predictors of readmission were mechanical ventilation required for over 24 hours (aOR 18.2;P=0.02), history of previous CDI (aOR 5.48;P< 0.01), HIV positive status (aOR4.60; P=0.04), alcohol abuse disorder (aOR2.07; P=0.01), Parkinson's disease (aOR 4.68;P< 0.01), index admission for non-complicated ulcerative colitis (aOR 4.72;P< 0.01), complicated ulcerative colitis (aOR 4.49; P=0.01), non-complicated Crohn's disease (aOR 2.54; P< 0.01) and hospital length of stay (aOR 1.01; P=0.04).

Conclusion

In patients recently admitted with IBD flare there is a 0.1% ninety-day specific readmission rate for CDI. We found multiple independent predictors for this CDI readmissions such as history of Parkinson's disease, HIV positive status, alcohol abuse disorder. The use of these predictors can be incorporated in the tailored care of IBD patients to prevent hospital readmissions. Finally, our study also revealed a high health care cost, charges and burden. This data can be used to improve care of IBD patients and health care policies.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.12

OP013 Acute Infusion Reactions Under Vedolizumab in Inflammatory Bowel Disease Patients: Results of a Multicenter Retrospective Observational Cohort Study

C Venturin ¹, S Nancey ², X Roblin ³, L Peyrin-Biroulet ⁴, N Mathieu ⁵, C Gay ⁶, P Danion ⁷, R Duclaux-Loras ⁸, C Meunier ⁸, B Flourie ⁸, G Boschetti ^10,^✉

Introduction

Vedolizumab is a fully humanized monoclonal IgG1 antibody directed toward a4β7-integrin approved for Crohn's Disease and Ulcerative Colitis treatment. Until now, a systematic follow-up after all vedolizumab infusions is recommended. Clinical trials and post marketing studies have reported infusion reactions ranged from 0.1 to 2.3%, but specific symptoms, circumstances and severity are not always detailed.

Aims & Methods

To report the frequency, the features, risk factors and severity of acute infusion reactions to vedolizumab in inflammatory bowel disease (IBD) patients. We performed a multicenter retrospective systematic review of all IBD patients treated with vedolizumab in 4 French university hospitals (Lyon-Sud, Saint-Etienne, Nancy and Grenoble). We collected patient's characteristics, symptoms, duration of treatment, concomitant drugs, history of previous infusion reaction to previous other biologics, anti-drug antibodies and outcomes of drug-induced acute infusion reaction.

Results

From May 2014 to February 2019, 550 patients received a total of 6459 vedolizumab infusions. in our cohort, 7 acute infusion reactions (0.1 %) could be identified, all occurring during drug infusion and none within 2 hours after infusion. No severe reaction was reported and vedolizumab was definitely discontinued due to acute infusion reaction in only

two cases. We failed to identify associated risk factors, especially history of infliximab infusion reaction, concomitant immunosuppressant or anti-vedolizumab antibodies.

Conclusion

We confirm in this multicenter study the excellent very short term safety profile of vedolizumab especially the absence of acute infusion reaction occurring within 2 hours of infusion. These data support the use-lessness of systematic follow-up of patients after vedolizumab infusion.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.13

OP014 Vedolizumab Levels in Mothers and Newborns, Clearance in Neonates, Susceptibility to Infection, and Achievement of Developmental Milestones

M Julsgaard ^1,^2,^✉, DC Baumgart ³, SMD Jørgensen ¹, MM Hansen ¹, A Grosen ¹, J Kjeldsen ⁴, HG Sørensen ⁵, L Larsen ⁶, S Wildt ⁷, P Weimers ⁸, KV Haderslev ⁹, I Vind ¹⁰, L Svenningsen ², J Brynskov ¹¹, S Lyhne ¹², T Vestergaard ¹, CL Hvas ¹, J Kelsen ¹

Introduction

Cord blood levels of anti-TNF agents in infants exposed in utero are higher than maternal levels.¹ Little is known about in utero exposure to and clearance of vedolizumab (VDZ) in neonates. Further, effects of in utero exposure to VDZ on risk of infant infections and development are not known.

Aims & Methods

In 12 hospitals in Denmark and Canada, we did a prospective study of pregnant women with inflammatory bowel disease who received VDZ during pregnancy. We investigated neonatal VDZ levels after exposure in utero and correlated these with maternal levels, VDZ duration, time to clearance, and infection risk during the first year of life. Infant developmental milestones were measured by the validated Ages and Stages Questionnaire (ASQ-3®). Levels of VDZ were measured in blood samples from women at delivery and in umbilical cords, and in infants for every 3 months until VDZ was no longer detected. Drug levels were measured by ELISA (IDKmonitor®). Multivariate and logistic regression analysis determined factors correlated with drug levels at birth in mothers and newborns, time to clearance, and factors influencing clearance in newborns.

Results

Of 44 women, 7 (16%) miscarried during the 1^st trimester. of the 37 evaluable women (18 CD, 18 UC and 1 IBDU), 3 (7%) remain pregnant. Eight (22%) received concomitant thiopurine therapy. Birth outcomes were: 1 (3%) with APGAR < 7 at 5 minutes, 2 (5%) preterm births, 3 (8%) small for gestational age, and no congenital malformations. We observed a positive correlation between gestational week for final VDZ infusion and both cord drug levels (r=0.59, p< 0.0001) and maternal levels (r=0.60, p< 0.0001). Median maternal and cord VDZ levels were 3.3 (range 0-36.1) μg/ ml and 2.9 (0.0-13.3) μg/ml with a median infant:maternal ratio of 0.52 (95% CI 0.38-0.67). The maternal and cord levels were significantly lower when VDZ was discontinued prior to the 3^rd trimester (p< 0.0001). None of the women (n=11, 30%) who discontinued VDZ prior to gestational week 29 experienced a relapse in the 3^rd trimester. of 30 infants, 23 (77%) had a detectable VDZ level at birth (median 2.9 (0-13.3) μg/ml), 3 (10%) at 3 months (0.3 (0.17-0.42) μg/ml), and all infants had cleared VDZ at 6 months. Normal or above average developmental milestone scores were observed in all infants at 1 year. Viral or bacterial infection occurred in 12 (32%) infants, with no sequela. Infection risk in the infant during the first year of life was not correlated with VDZ drug level at birth (median VDZ level: infection 2.9 (0-11.4) μg/ml and no infection 2.0 (0-13.3) μg/ml; p=0.6). Maternal combination therapy of VDZ and thiopurine compared with VDZ monotherapy was not associated with increased risk for infant infection (RR 1.6, 95%CI 0.83-3.09; p=0.24).

Conclusion

Intrauterine VDZ exposure does not appear to increase the risk of adverse pregnancy outcomes. Cord and maternal VDZ levels correlate with gestational week for final infusion in pregnancy. in women in remission, discontinuation of VDZ prior to the 3^rd trimester is an option for reduction of foetal exposure, but timely restart after delivery is advisable to maintain remission. Infant VDZ levels are lower than maternal levels at the time of delivery. A rapid post partum clearance of VDZ is observed, supporting that live vaccines may be administrated at 6 months of age.

Disclosure

MJ has received unrestricted grants for this investigator-initiated study from the Health Research Fund of the Central Denmark Region, Colitis-Crohn Denmark, the A.P. Moeller Foundation of the Advancement of Medical Science, the Department of Hepatology & Gastroenterology, Aarhus University Hospital (Aarhus, Denmark) and Takeda Pharma. The external funders had no involvement in any aspect of the study or writing of the abstract.

References

1.Julsgaard M., Christensen L.A., Gibson P.R. et al. Concentrations of Adalimumab and Infliximab in Mothers and Newborns, and Effects on Infection. Gastroenterology. 2016; 151(1): 110–119. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.14

OP015 Pregnancy in Women with Inflammatory Bowel Disease: A French Nationwide Study 2010-2018

A Meyer ^1,^2,^✉, J Drouin ¹, A Weill ¹, F Carbonnel ², R Dray-Spira ¹

Introduction

Inflammatory bowel diseases (IBD) most commonly occur in young adults, including women of childbearing age.

Aims & Methods

This study was designed to describe pregnancy in IBD women in the age of widespread use of biologics. Using French national health data system (SNDS) data, we identified all pregnancies ending between April 1, 2010 and December 31, 2018 in IBD and non-IBD patients in France. Pregnancy and IBD characteristics were described. Pregnancy outcomes were compared between IBD and non-IBD pregnancies using multivariable logistic regression models.

Results

We included 36,654 IBD and 8,608,142 non-IBD pregnancies. Among IBD pregnancies, 75.6% ended in live births, 0.4% in stillbirths, 21.2% in abortions and 1.3% were ectopic pregnancies. Pregnancies in women with IBD vs. without IBD more frequently resulted in caesarean section (26.1% vs 20.0%, aOR 1.39; 95%CI: 1.35 - 1.42), premature birth (9.3% vs 6.7%, aOR 1.44; 95%CI: 1.38 - 1.50) and small for gestational age birth (11.1% vs 9.8%, aOR 1.13; 95%CI: 1.11 - 1.21). Active IBD during pregnancy was associated with particularly marked increases in the rates of prematurity (aOR 1.99; 95%CI: 1.86 - 2.14), low birth weight (aOR 1.86;

95%CI: 1.71 - 2.03), pregnancy-related hospitalization (aOR 1.92; 95%CI: 1.81 - 2.04), and an increased rate of stillbirths (aOR 1.43; 95%CI: 1.09 -1.86). CD activity decreased during pregnancy, while UC activity did not change.

Conclusion

Pregnancies in women with IBD are associated with moderately increased risks of caesarean section, prematurity and low birth weight, especially among women with active IBD. Disease activity decreased during pregnancy in women with CD, but remained unchanged in women with UC.

Disclosure

Franck Carbonnel received board or lecture fees from En-terome, MSD, BMS, Janssen, Pfizer, Abbvie, Mayoly Spindler, Takeda, Pile-je. Other authors: no disclosure.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.15

OP016 Structural Variant Events Dominate a Molecular Continuum in the Progression of Barrett's Oesophagus

A Katz-Summercorn ^1,^✉, S Jammula ², A Frangou ³, I Peneva ³, M O'Donovan ¹, M Tripathi ¹, S Malhotra ¹, M di Pietro ¹, S Abbas ¹, G Devonshire ², A Blasko ¹, W Januszewicz ¹, K Nowicki-Osuch ¹, S Macrae ¹, A Northrop ¹, D Wedge ³, RC Fitzgerald ¹

Introduction

Barrett's oesophagus (BE) is the main risk factor for the development of oesophageal adenocarcinoma (OAC) yet fewer than 1% per year of non-dysplastic BE cases progress to cancer ¹. Genomic diversity has been shown to be a factor for progression ²³.

Aims & Methods

We present the first large, multi-omics study comparing non-progressors with those who progressed to a range of dysplasia grades, with the aim of providing a comprehensive insight into the diversity and molecular changes driving the disease to cancer. Whole genome sequencing (50X) was performed on 35 biopsies from ten cases, representing both indolent and progressing, dysplastic disease in order to analyse the intralesional, spatial heterogeneity. Next, to delineate what determines an individual's clinical trajectory, we integrated the genomic, transcriptomic and epigenomic (850K EPIC array) landscapes of patients with distinct progression profiles (n=147 patients: 27 “indolent” (non-dysplastic non-progressors); 12 non-dysplastic pre-progressors; 61 progressed (15 low-grade, 25 high-grade dysplasia, 21 intramucosal carcinoma) and 47 cases of BE taken adjacent to OAC). All biopsies were well-annotated with clinical information and reviewed independently by three pathologists.

Results

Spatial analysis within a BE segment reveals the molecular features to be quantitatively similar with respect to the mutation burden, copy number and total structural variant (SV) count, irrespective of the histopathological grade of the biopsy. Clonality analysis highlights the complexity. Some cases arise from a clear ancestral clone while others share minimal variants, but with no difference in the total number of clones according to progression status.

Genomic analysis across the entire 147 patient cohort emphasises the in-ter-lesional heterogeneity. Mutational signatures (tri-nucleotide) are laid down early and persist. Hence, Cosmic signature 17, the hallmark of OAC, is enriched throughout the cohort. As expected, specific driver events do discriminate indolent from progressed cases to some extent including APC, GATA6 and ERBB2 amplification. However, the frequency of these events is low making them poorly predictive aside from TP53 which affects 65% progressed cases.

By contrast, structural variants (SVs), which are more readily characterised from WGS compared with exomes or mutation panels, dominate the BE landscape and are the feature that best explains the variance of the cohort. The SVs increase gradually with the grade and correlate with TP-53mutation, forming a continuum along which other molecular features align. Further SV analysis reveals fragile sites to be recurrently altered with a dominance of unclustered translocations from the indolent stage. Clustered translocations, which correlate with retrotransposon activity, are more prominent in progressed samples, with complex patterns e.g. breakage-fusion bridges and chromothripsis. We observe whole genome doubling to be a late event, only in progressed samples (23%, 14/61). Transcriptomic analysis correlates with driver amplifications and reveals a clear but gradual differential gene expression between indolent and progressed biopsies: mainly fatty acid metabolism downregulation and early signs of DNA damage and immune infiltration.

Conclusion

These data suggest that BE is a disease continuum in which the accumulation of events, dominated by SVs and usually occurring in the context of TP53 mutation, tips the balance to progression. We suggest that it is the genomic instability, rather than the heterogeneity that is the driving force in this progression.

Disclosure

R. Fitzgerald and M. O'Donovan are named on patents related to Cytosponge and related assays which have been licensed by the Medical Research Council to Covidien GI Solutions (now Medtronic) and are a co-founders of CYTED Ltd. M. O'Donovan has a patent on a method for detecting benign conditions in upper intestinal tract. These are not directly involved in the topic of this paper.

References

1.Bhat S. et al. Risk of malignant progression in Barrett's esophagus patients: results from a large population-based study. J Natl Cancer Inst 103, 1049–1057 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Maley C. C. et al. Genetic clonal diversity predicts progression to esophageal adenocarcinoma. Nat. Genet. 38, 468–473 (2006). [DOI] [PubMed] [Google Scholar]
3.Martinez P. et al. Dynamic clonal equilibrium and predetermined cancer risk in Barrett's oesophagus. Nat. Commun. 7, 12158 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.16

OP017 Independent Validation of Tissuecypher to Predict Progression in Community-Based Barrett's Esophagus with Low-Grade Dysplasia: Analysis in a Completed Prospective Study

N Frei ^1,^✉, A Khoshiwal ¹, K Konté ¹, EA Bossart ², K Stebbins ², Z Yi ², RE Pouw ³, FJWT Kate ⁴, CA Seldenrijk ⁵, SL Meijer ⁶, RJ Critchley-Thorne ², JJGHM Bergman ¹

Introduction

Low-grade dysplasia (LGD) is the best predictor for malignant progression in Barrett's esophagus (BE). The majority of community-based LGD cases are down-staged to non-dysplastic (ND)BE when reviewed by gastrointestinal (GI) expert pathologists, whereas patients with confirmed LGD have a 10% annual risk of progression to high-grade dysplasia (HGD) or esophageal adenocarcinoma (EAC). Although recommended

by current guidelines, expert review is poorly defined, subjective, prone to inter-observer variation and not widely available. Recent studies indicate that an objective, automated risk prediction assay (TissueCypher Barrett's Esophagus Assay) accurately stratifies NDBE into low-, intermediate- and high-risk for progression to HGD/EAC.

Aims & Methods

We aimed to investigate if this risk prediction assay objectively predicts progression to HGD and/or EAC in community-based BE with LGD, comparable to three GI expert pathologists. A blinded, retrospective cohort study was derived from the screening cohort of a randomized controlled trial of Surveillance vs RadioFrequency ablation (SURF) for BE patients with community-based LGD. Hematoxylin and Eosin and p53 immunohistochemistry slides from all random biopsy levels of the first endoscopy with LGD were independently reviewed by 3 expert patholo-gists with an international reputation. All random biopsy levels were additionally tested by the risk prediction assay, which classifies patients into low-risk (LR), intermediate-risk (IR) and high-risk (HR) for progression to HGD/EAC within 5 years. A histological revision diagnosis of NDBE was considered as low- (LR), while indefinite for dysplasia (IND) and LGD were considered as high-risk (HR) for progression.

Results

155 BE patients (123 male), mean age 61±10 years, were studied. 34 patients progressed (progressors) and 121 did not progress (non-pro-gressors) to HGD and/or Ca. Median time to progression was 2.4 years, and median surveillance time in non-progressors was 7.9 years. Sensitivity of the risk prediction assay (HR + IR combined vs. LR) was 65%, and ranged from 68%-71% for the 3 experts, and specificity was 81% compared to 63%-78% for the experts. Patients scoring HR by the assay had a 9.5% annual progression risk compared to an annual progression risk of 7.6%-7.7% in BE patients with expert-confirmed LGD.

Conclusion

The risk prediction assay stratified BE patients with a community-based diagnosis of LGD with diagnostic accuracy comparable to three renowned expert pathologists. The predictive performance of the assay outperformed all expert pathologists. The assay allows for automated and objective risk stratification, which may be a practical and effective solution to the lack of standardization of expert pathology review of LGD as advocated by all guidelines.

Disclosure

E.A. Bossart, and K. Stebbins have ownership interest (stock options) in Cernostics, Inc., R.J. Critchley-Thorne has ownership interest (stock, stock options and patents) in Cernostics, Inc., and Y. Zhang is a consultant to Cernostics, Inc. J.J. Bergman has received financial support for clinical trials from Medtronic, Pentax Medical, C2 Therapeutics, Aqua Medical, Boston Scientific, Erbe Medical, Cernostics, Ninepoint Medical, Fujifilm, and Olympus. He is a recipient of speaker's fees from Fujifilm and is a consultant for Olympus and Fractyl. The other authors declare no conflict of interest.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.17

OP018 Long-Term Outcomes after Endoscopic Resection without Subsequent Ablation Therapy for Early Barrett's Neoplasia

S van Munster ^1,^✉, E Nieuwenhuis ², BLAM Weusten ³, Herrero L Alvarez ⁴, A Bogte ⁵, A Alkhalaf ⁶, BE Schenk ⁷, EJ Schoon ⁸, WL Curvers ⁹, AD Koch ¹⁰, SEM van de Ven ¹¹, PJ De Jonge ¹⁰, TJ Tang ¹², WB Nagengast ¹³, FTM Peters ¹⁴, J Westerhof ¹⁵, M Houben ¹⁶, JJ Bergman ¹⁷, RE Pouw ¹

Introduction

After endoscopic resection (ER) of neoplastic lesions in Barrett's esophagus (BE), it is generally recommended to ablate the remaining flat BE to minimize the risk for metachronous disease. However, the majority of patients will not develop metachronous disease, and if it occurs, it is generally detected at an early stage allowing re-ER. Ablation is still accompanied by complications and requires multiple hospital visits. We report the long-term outcomes for all patients treated in the Netherlands (NL) between 2008 and 2018 who did not undergo ablation after ER for early BE neoplasia.

Aims & Methods

Endoscopic therapy for BE neoplasia in NL is centralized in 9 expert centers with specifically trained endoscopists and patholo-gists. Uniformity is further ensured by a joint protocol and regular group meetings. Prospectively collected treatment and FU data are registered in a uniform database. We report all patients who underwent ER for a neoplastic lesion and in whom subsequently no ablation therapy was applied. We aimed to report progression rates during endoscopic FU and mortality after endoscopic FU was stopped. Data on date and cause of death were extracted from Statistics Netherlands (CBS).

Results

Of the 2,098 BE patients with early neoplasia; 1,305 underwent ER for a visible neoplastic lesion and 95 (7%) entered endoscopic surveillance without additional ablation. 78% was male, mean age 74(±10)yrs, BMI 27(±6)kg/cm2, ASA classification II (64%) or ≥III (29%). Median BE was C4M6. ER was performed for LGD (10%), HGD (25%) or EAC (65%). Reasons for not proceeding to ablation therapy were: a combination of age, comorbidity and extent of residual BE (88%); anticipated poor response upon ablation therapy (e.g. BE regeneration of the ER-scar) (14%); other treatment protocols by that time (13%); patient preference (7%); and/or complications after ER (4%). After ER, median BE was C2M5 with IM (52%), LGD (31%), or HGD (6%) (no histology obtained: 12%). During median 21 months of endoscopic surveillance (IQR 11-51) with a median of 4 endoscopies

per patient, no patient progressed to advanced cancer. 17 pts (18%) progressed to HGD/EAC after median 29 mo, all underwent curative endoscopic treatment for HGD (n=10) or early EAC (n=7). in 62 patients (64%), endoscopic surveillance was stopped median 20 months after ER (IQR 5-59) because of comorbidity and anticipated limited life expectancy. During median 44 months (IQR 28-77) after endoscopic FU was stopped, no patient developed symptomatic disease or had a tumor-related deaths, whilst 40% died of unrelated causes.

Conclusion

In selected patients, ER monotherapy with endoscopic surveillance of the residual BE, is a valid alternative to prophylactic ablation therapy.

Disclosure

Nothing to disclose

[Follow-up after ER monotherapy]

	Follow-up duration	Follow-up, worst histology
Remaining histology in flat BE after ER	Median FU in months (IQR)	Median number of endoscopies (IQR)	NDBE	LGD	HGD	Early EAC	No follow-up	Annual progression risk [95% confidence interval]
NDBE	20 (10-48)	3 (1-5)	37 (77%)	3 (6%)	3 (6%)	5 (10%)	-	6.4% [3.3-12.1]
LGD	22 (14-60)	3 (1-5)	13 (46%)	19 (32%)	5 (18%)	1 (4%)	1	6.7% [3.1-13.9]
HGD	41 (14-62)	7 (2-10)	-	-	3 (50%)	3 (50%)	-	14.5% [5.0-34.6]
No biopsies obtained	17 (8-65)	3 (1-6)	5 (83%)	1 (17%)	-	-	5	-

Open in a new tab

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.18

OP019 P53 Immunohistochemistry Improves the Inter-Observer Agreement Among Histopathologists and Reduces the Rate of Indefinite for Dysplasia Diagnosis in Barrett's Esophagus

W Januszewicz ^1,^2,^3,^✉, ND Pilonis ³, R Philips ³, M O'Donovan ³, T Sawas ⁴, DA Katzka ⁴, A Miremadi ³, S Malhotra ³, M Tripathi ³, RC Fitzgerald ⁵, M di Pietro ⁶

Introduction

A diagnosis of indefinite for dysplasia (IND) in Barrett's esophagus (BE) is often made in the presence of cytological changes, which cannot be unequivocally classified and generally triggers early en-doscopic follow up. IND carries a slightly higher cancer risk, but it is often downgraded at follow up after high dose PPI. Inter-observer agreement among histopathologists for IND is poor. Assessment of expression of p53 by immunohistochemistry can aid in the histopathological diagnosis of dysplasia, but its role in the cases with IND has not been fully characterized.

The aim of the study was to assess the utility of the p53 staining in improving inter-observer agreement (IOA) in patients with diagnosis of BE-IND.

Aims & Methods

In this multi-center retrospective study, we have reviewed cases with a previous diagnosis of BE-IND. Slides were reviewed by 4 expert gastrointestinal pathologists divided into 2 groups (Group 1 and 2), so that each individual slide was assessed independently by two different pathologists. in the first stage, hematoxylin and eosin stained slides were assessed and IOA for each BE diagnosis (H&E diagnosis) was generated using Cohen unweighted Kappa correlation coefficients (K). After a wash out period of at least eight weeks the cases were reviewed by the same pathologists with p53 immunochemistry available (H&E+p53 diagnosis). Aberrant p53 expression was defined as either overexpression or absent expression. The IOA and the proportion of down-graded, confirmed, and upgraded IND diagnoses after H&E=p53 diagnosis was calculated. The K values were compared using the McNemar's test.

Results

In total, 120 pathology slides of BE-IND cases were analyzed. in the initial H&E assessment, the diagnosis of BE-IND was confirmed in approximately a third of the cases [41 cases (34.2%) and 32 cases (26.6%) in Groups 1 and 2,respectively]. Most of the diagnoses were downgraded into a non-dysplastic BE (NDBE) [Group 1: 72 cases (60.0%),Group 2: 80 cases (66.7%)]. An up-graded diagnosis was made in 7 cases (5.8%) by Group 1 [4 low-grade dysplasia (LGD),2 high-grade dysplasia (HGD),1 in-tramucosal carcinoma (IMC)], and in 8 cases (6.7%) by Group 2 (all into LGD). The overall IOA was poor with K=0.33 (95%CI 0.18-0.47). The K values for the IOA for NDBE, IND, LGD and HGD/IMC were 0.43 (95%CI 0.26-0.59), 0.32 (95%CI 0.14-0.49), -0.05 (95%CI 0.08- -0.01) and 0.0, respectively. in the second assessment, the p53 expression was found aberrant in 14 cases (11.6%), equivocal in 14 cases (11.6%) and normal in 92 cases (76.8%). The adjunct of p53 expression led to a 2-fold reduction of IND diagnoses. The IOA for IND remained fair (K=0.30; 95%CI 0.06-0.53) when compared to pre-p53 assessment. For NDBE the IOA increased to moderate (K=0.53; 95%CI 0.36-0.7 and K=0.52; 95%CI 0.34-0.70, respectively. The highest increase in IOA was observed for LGD and HGD/IMC diagnoses, where the agreement was substantial (K =0.80; 95%CI 0.61-0.99) and perfect (K=1.0; 95%CI 1.0-1.0), respectively (P< .001 for both).

Conclusion

P53 immunochemistry staining significantly improves the IOA among histopathologists, particularly with regards to dysplasia diagnosis and reduced the rate of IND diagnosis. P53 immunostaining should be routinely used as adjunct to the histological assessments of BE samples.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.19

OP021 Laryngeal and Pharyngeal Squamous Cell Carcinoma after Antireflux Surgery in the Five Nordic Countries

M Yanes ^1,^✉, G Santoni ², J Maret-Ouda ^2,³, E Ness-Jensen ^2,⁴, MA Färkkila ⁵, E Lynge ⁶, E Pukkala ^7,⁸, P Romundstad ⁸, L Tryggvadóttir ¹⁰, M von Euler-Chelpin ⁶, J Lagergren ^2,¹²

Introduction

Gastroesophageal reflux disease (GERD) seems to increase the risk of laryngeal and pharyngeal squamous cell carcinoma.

Aims & Methods

This population-based cohort study aimed to clarify whether antireflux surgery prevents laryngeal and pharyngeal squamous cell carcinoma. All adults with documented GERD were identified in the five Nordic countries (Denmark, Finland, Iceland, Norway and Sweden) in 1980-2014. First, cancer risk after antireflux surgery was compared to the expected risk in the corresponding background population by calculating standardized incidence ratios (SIR) with 95% confidence intervals (CI). Second, cancer risk among antireflux surgery patients was compared to non-operated GERD-patients using multivariable Cox regression, providing hazard ratios (HR) with 95% CIs, adjusted for sex, age, calendar period, and diagnoses related to tobacco smoking, obesity, and alcohol overcon-sumption.

Results

Among 814,230 GERD-patients, 47,016 (5.8%) underwent antireflux surgery. The overall SIRs and HRs of the combined outcome laryngeal or pharyngeal squamous cell carcinoma (n=39) were decreased after antireflux surgery (SIR=0.62 [95% CI 0.44-0.85]) and HR=0.55 [95% CI 0.38-0.80]). The point estimates were further decreased >10 years after antireflux surgery (SIR=0.48 [95% CI 0.26-0.80] and HR=0.47 [95% CI 0.26-0.85]). The risk estimates of laryngeal squamous cell carcinoma were particularly decreased >10 years after antireflux surgery (SIR=0.28 [95% CI 0.08-0.72] and HR=0.23 [95% CI 0.08-0.69]), while no such decrease over time after surgery was found for pharyngeal squamous cell carcinoma. Analyses of patients with severe GERD (reflux esophagitis or Barrett's esophagus) showed similar results.

Conclusion

Antireflux surgery may decrease the risk of laryngeal squa-mous cell carcinoma and possibly also of pharyngeal squamous cell carcinoma.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.20

OP023 Hybrid Argon-Plasma-Coagulation Versus Radiofrequency Ablation in Barrett'S Esophagus After Endoscopic Resection of Neoplastic Lesions. A Randomized Trial at a Tertiary Center

M Knabe ^1,^✉, J Wetzka ², J Richl ³, H Kronsbein ⁴, L Welsch ¹, A May ²

Introduction

Neoplastic Barrett esophagus (BE) is usually treated by a combination of endoscopic resection for (visible) neoplasia followed by ablation of the remaining BE mucosa with the aim to eradicate the entire BE. Different ablation techniques have been established in clinical routine for this therapy approach. Radiofrequency ablation (RFA) and argon-plasma -coagulation (APC) are most frequently uses. There is some evidence that prior injection to APC (Hybrid-APC) lowers stricture rates in these patients. Beside this there is only one study comparing APC and RFA in a randomised trial setting.

Aims & Methods

Following resection (EMR), patients were randomized to Hybrid-APC or RFA. Hybrid-APC was used with 60 W after injection with saline. RFA was used with 90° focal catheter and simplified protocol of 12j/cm² x 3 or HALO 360° Balloon with 10j/cm²-clean 10j/cm². Complications, time for ablation and postinterventional pain (visual scale 1-10) and pain duration were recorded. Patients underwent follow-up protocol after 3,6,12 and 24 months. Primary therapeutic success was defined as follow-up endoscopy (6 months) showing normal endoscopic neo-Z-line with negative biopsies. The Aim of this study is to compare the two major ablation techniques in a large group of patients. The study was planned as a non-inferior trial. All patients who have been treated for neoplastic lesions in Barrett's esophagus and were planned for subsequent ablation were randomized to RFA or H-APC. Time of deviation, complications and postinterventional pain (1-10 on scale, duration) were measured. All patients were planned for follow-up gastroscopy with chromoendoscopy and biopsies at the Zline.

Results

103 patients (60 males, 4 females, mean age 64,9 years) have been included to the study with N=48 randomized to RFA and N=55 to H-APC group. N=58 patients have completed ablation and short-term follow-up (3 or 6 months). 199 ablative interventions have been performed so far. in mean 1,95 ablations (min 1 max 5) were needed for successful Barrett's eradication. There was no significant difference in Barrett's length between the two groups. The mean length of BE before ablation was 4,3cm (H-APC) and 4.7cm (RFA).

Barrett's eradication rate after 6 months follow-up was 87% in the RFA group and 91 % in the hybrid- APC group. Except of non-healers (10.1 %) all Barrett remnants were localized at the neo-Zline. However, the post interventional pain was significantly higher in the RFA group: mean 4.1 /10 and duration of 5.7 days, whereas patients in H-APC group complained about pain for 3.3 days in mean and severity of 2.1/10. Stenosis with need for intervention were registered in 2% in the H-APC arm and 13% in the RFA Arm.

Conclusion

Both ablation techniques have good results in eradication rate. The pain severity and duration were significantly higher in the RFA group. Due to the substantial difference in the complication rate (stenosis 13% vs 2 %) we decided to stop further study enrolments.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.21

OP024 Over-The-Scope Clips (Otsc) Versus Transarterial Angiographic Embolisation (Tae) As Salvage Therapy For Refractory Peptic Ulcer Bleeding - A Propensity Score Matched Analysis

A Kuellmer ^1,^✉, T Mangold ¹, D Bettinger ^1,², L Maruschke ³, A Wannhoff ⁴, K Caca ⁴, E Wedi ⁵, C Jung ⁶, T Kleemann ⁷, T Schulz ⁸, R Thimme ⁹, AR Schmidt ¹

Introduction

Refractory bleeding from peptic ulcer is difficult to treat and associated with substantial morbidity and mortality. Over-The-Scope clips (OTSC) have shown superiority to standard endoscopic treatment for recurrent bleeding, but whether it can compete with other salvage therapeutic modalities such as transarterial angiographic embolisation (TAE) has not been investigated yet.

Aims & Methods

In this retrospective, multicenter study, we report on one-hundred twenty-eight patients treated with OTSC (n=66) or TAE (n=62) as salvage therapy for peptic ulcer bleeding refractory to standard endoscopic therapy between 2009-2019. Primary endpoint was clinical success defined as successful hemostasis and absence of rebleeding within 7 days. Main secondary endpoints were overall and procedure-related adverse events, length of hospital and intensive care unit stay, number of transfusions and mortality. Propensity score matching using the nearest neighborhood method was performed to adjust for differences in baseline characteristics of the included patients.

Results

For baseline characteristics, there were no significant differences in both groups regarding age, Charlson comorbidity index, Rockall score, Helicobacter pylori status, ongoing anticoagulation, NSAID intake, primary hemostasis rate in first line therapy and number of endoscopic treatment attempts before salvage therapy. The majority of ulcers in both groups were located in the duodenal bulb (65% in OTSC group; 85.5% in the TAE group; p=0.014). The TAE group included significantly more Forrest Ia bleedings compared to the OTSC group (38.7% vs. 19.7%, p=0.02). The proportion of FIb bleedings was significantly higher in the OTSC group (63.6% vs. 43.5%;p=0.03). in both groups, the proportion of patients with ulcer size > 20mm was similar (27.3% vs. 33.9%, p=0.48). Clinical success was by trend higher in the OTSC group (74.2%) compared to the TAE group (59.7%) (p=0.092). Rebleeding during hospital stay occurred in 27.4% of cases in the TAE group as opposed to 18.2% of cases in the OTSC group (Odds ratio: 1.7 {0.74-9.93}; p=0.29). In-hospital mortality was higher in the TAE group compared to the OTSC group (9.1 vs. 22.6%, OR 2,92 {1.04-8.16}; p=0.05). TAE patients stayed significantly longer in ICU than OTSC patients (mean 10.45 ± 13.7 vs. 7.01 ± 9.9 days; p=0.02) and received significantly more blood transfusions (mean 10.3 ± 14.0 vs. 9.4 ± 9.0; p=0.04). Overall adverse events were similar in both groups. After propensity score matching, there were no significant differences between the two treatment groups regarding clinical success and rebleeding. Moreover, there was a statistically significant higher In-hospital mortality in patients treated with TAE compared to OTSC (OR 5.52 {1.11-27.43}; p=0.048).

Conclusion

In this propensity score matched analysis, OTSC treatment for refractory peptic ulcer bleeding showed comparable efficacy compared to TAE with respect to a durable hemostasis. However, patients undergoing OTSC therapy required less transfusions, had a shorter ICU stay and showed lower in-hospital mortality.

Disclosure

PD Dr. Arthur Schmidt and Prof. Dr. Karel Caca received lecture fees and study grants from Ovesco Endoscopy GmbH, Tübingen, Germany.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.22

OP025 Revising The Paradigm of Endoscopic Treatment For Gastric Hyperplastic Polyps: Analysis of Recurrence Rate and Malignant Progression

M Turkot ^1,^✉, W Januszewicz ^1,², J Regula ^1,², MF Kaminski ^1,^2,³

Introduction

Gastric hyperplastic polyps (GHPs) remain the second most common type of epithelial polyps in the stomach after fundic gland polyps. The rate of dysplasia within GHPs is highly variable among reports and most of the guidelines recommend to resect GHPs ≥1cm due to their potential risk of malignant progression. Unlike neoplastic polyps, GHP may arise as a response to mucosal injury, therefore, complete removal of these lesions do not guarantee a definite cure and recurrent GHPs have been widely reported in the literature.

Aims & Methods

In this single-center retrospective study we aimed to assess the rate of GHPs’ malignant progression and the recurrence rate after GHP removal. We reviewed the hospital endoscopic database to identify adult individuals with a diagnosis of GHPs. We then used the National Cancer Registry to follow-up those patients until 12.2017 and identify those who developed a subsequent gastric cancer (GC). We then analyzed a subset of GHPs ≥1cm that underwent endoscopic resection to determine the rate of local recurrence and rate of cancerous component in removed lesions. We defined recurrence as the presence of a polyp or number of polyps during the follow-up (FU) which was exceeding the number of polyps at the time of initial polypectomy; or the presence of any remnant tissue within a post-polypectomy scar with biopsies demonstrating GHP.

Results

Overall, we identified 675 patients [464 (68.7%) females], of a mean age 63.0(±13.3) with GHPs between 2005 and 2017. of these, 368 patients (54.5%) had a single lesion, 221 (32.7%) had several GHPs (≤5) and 86 (23.3%) had multiple GHPs (>5). in 89 cases (13.2%), the polyps were present in multiple locations throughout the stomach, and in 202 (29.9%), 258 (38.2%), 61 (9.0%) and 65 (9.7%) they were confined to the antrum, body, fundus and the cardia, respectively. The initial median polyp size was 4mm (IQR 3-8). The background mucosa was assessed in 221 individuals (32.7%) with atrophic gastritis and/or intestinal metaplasia being present in most of the patients (179/221; 81.0%). H.pylori status had been obtained in 304 individuals (45.0%), within those nearly a third were found to be positive (108/304; 35.5%). Among patients with GHP who did not undergo endoscopic treatment (n=197; 29.2%), only two patients (1.0%) were diagnosed with a subsequent GC after a median 5.9 years

(range 0.9-11.9 years). We identified 105 (15.6%) patients with GHPs ≥1cm who underwent endoscopic resection. of these, 15 patients had a local recurrence (14.3%). The majority of the recurrent polyps were initially localized in the antrum (11/15, 73.3%) and most had a sessile morphology (Paris 0-Is; 66.7%). of all resected GHPs ≥1cm, 3 harboured cancer (2.9%). All of the cancerous polyps had a pedunculated morphology (0-Ip) and a size of ≥2cm

Conclusion

GHPs carry a low risk of malignant progression. Given a significant recurrence rate (>14%), endoscopic resection of GHPs ≥1cm, especially when sessile and localized in the antrum, remains debatable. in contrast, bigger (≥2cm) and pedunculated polyps do require complete endoscopic resection as they may harbour cancer.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.23

OP027 Risk of A Post-Colonoscopy Colorectal Cancer Diagnosis in Patients with Inflammatory Bowel Disease: A Danish Population-Based Cohort Study

FS Troelsen ^1,^✉, HT Sørensen ¹, L Pedersen ¹, R Erichsen ¹

Introduction

Post-colonoscopy colorectal cancers (PCCRCs), defined by the World Endoscopy Organization (WEO) as colorectal cancers (CRCs) diagnosed within 6-36 months following a colonoscopy, account for up to 8% of all CRCs. Previous studies suggest that PCCRCs account for up to 50% of all CRCs diagnosed in patients with inflammatory bowel disease (IBD); however, few studies have investigated the absolute risk of PCCRC in IBD-patients undergoing colonoscopy. Better understanding of IBD-relat-ed PCCRC risk would improve patient guidance clinical decision making, and help foster the understanding of PCCRC development.

Aims & Methods

Using data from Danish health and administrative registries, we conducted a population-based cohort study of all IBD and non-IBD patients undergoing colonoscopy in Denmark during 1995-2015. We defined PCCRC as recommended by the WEO. We followed IBD and non-IBD patients from their first-time colonoscopy plus six months until diagnosis of PCCRC, death, colectomy, emigration, or 36 months, whichever occurred first. We calculated 6-36 month cumulative incidence proportions (CIPs) of PCCRC among IBD and non-IBD patients. We also computed hazard ratios (HRs) of PCCRC, comparing IBD with non-IBD patients, utilizing Cox proportional-hazards regression analyses. We applied the same approach for subsequently performed colonoscopies. in line with the WEO, we also calculated PCCRC-3 year rates by dividing the number of false negative colonoscopies with the total number of true positive and false negative colonoscopies to estimate the proportion of IBD and non-IBD CRC patients experiencing PCCRC. Finally, we calculated PCCRC-3 year rates in IBD and non-IBD patients stratified by the total number of colonoscopies performed during the study period.

Results

We observed 138 PCCRCs in 34,688 IBD patients who underwent colonoscopy, and 1,909 PCCRCs among 358,217 non-IBD patients. The 6-36 month CIP of PCCRC after first-time colonoscopy was 0.21% [95% confidence interval (CI): 0.17-0.27] for IBD patients and 0.37% (95% CI: 0.35-0.39) for non-IBD patients. The absolute risks were comparably low after subsequent colonoscopies for both IBD and non-IBD patients. Comparing IBD with non-IBD patients, the HR of PCCRC after the first colonoscopy was 0.96 (95% CI: 0.75-1.22). The HRs after subsequent colonosco-pies were also close to 1.0. The PCCRC-3 year rate was 24.3% for IBD patients (19.2% for patients with Crohn's Disease and 26.4% for patients with ulcerative colitis) and 7.5% for non-IBD patients. The PCCRC-3 year rate increased with the total number of colonoscopies performed for both IBD as well as non-IBD patients.

Conclusion

Although PCCRCs accounted for a substantial proportion of all IBD-related CRCs, IBD patients had a low absolute risk of PCCRC after colonoscopy. The high PCCRC-3 year rates may stem from increased colonoscopy frequency in IBD patients thereby increasing the potential of CRC to be categorized as PCCRC.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.24

OP028 Trends in Colorectal Cancer Incidence and Stage Distribution in Europe in The Colorectal Cancer Screening Era: An International Population-Based Study

R Cardoso ^1,^2,^✉, M Hackl ³, P Ihle ³, H De Schutter ⁴, N Van Damme ⁴, Z Valerianova ⁵, T Atanasov ⁵, O Májek ^6,⁷, J Mužík ^6,⁷, MC Nilbert ⁸, AJ Tybjerg ⁸, K Innos ⁹, M Mägi ¹⁰, A-M Bouvier ¹¹, V Bouvier ¹², A-S Woronoff ¹³, M Cariou ¹⁴, M Robaszkiewicz ¹⁴, P Delafosse ¹⁵, F Poncet ¹⁵, C Senore ¹⁶, S Rosso ¹⁷, I Vincerzevskiene ¹⁸, VEPP Lemmens ^19,²⁰, MAG Elferink ¹⁹, TB Johannesen ²¹, MJ Bento ^22,^23,²⁴, F Alves da Costa ²⁵, A Miranda ²⁶, V Zadnik ²⁷, T Zagar ²⁷, A Lopez de Munain Marques ²⁸, R Marcos-Gragera ^29,^30,³¹, M Puigdemont ³², J Galceran ³³, M Carulla ³³, López MD Chirlaque ³⁴, K Sundquist ^35,^36,³⁷, M Weber ³⁸, A Jordan ³⁸, C Herrmann ³⁹, M Mousavi ⁴⁰, A Ryzhov ^41,⁴², H Brenner ^43,^44,⁴⁵

¹German Cancer Research Center (DKFZ) and National Center for Tumor Diseases (NCT), Division of Preventive Oncology, Heidelberg, Germany

²Medical Faculty Heidelberg, Heidelberg, Germany

³Austrian National Cancer Registry, Statistics Austria, Vienna, Austria

⁴Belgian Cancer Registry, Brussels, Belgium

⁵Bulgarian National Cancer Registry, University Hospital of Oncology, Sofia, Bulgaria

⁶Institute of Health Information and Statistics of the Czech Republic, Prague, Czech Republic

⁷Institute of Biostatistics and Analyses, Faculty of Medicine, Masaryk University, Brno, Czech Republic

⁸Danish Cancer Society Research Center, Copenhagen, Denmark

⁹National Institute for Health Development, Department of Epidemiology and Biostatistics, Tallinn, Estonia

¹⁰National Institute for Health Development, Estonian Cancer Registry, Tallinn, Estonia

¹¹Digestive cancer registry of Burgundy, Dijon, France

¹²Digestive Tumors Registry of Calvados, U1086INSERM UCN - ANTICIPE, Caen, France

¹³Cancer Registry of Doubs, Besançon, France

¹⁴Digestive Tumors Registry of Finistère, CHRU Morvan, Brest, France

¹⁵Cancer Registry of Isère, Grenoble, France

¹⁶University Hospital ‘Città della Salute e della Scienza’, SSD Epidemiologia e screening - CPO, Turin, Italy

¹⁷Piedmont Cancer Registry, Turin, Italy

¹⁸Lithuanian Cancer Registry, National Cancer Institute, Vilnius, Lithuania

¹⁹Netherlands Comprehensive Cancer Organisation (IKNL), Department of Research and Development, Utrecht, Netherlands

²⁰Erasmus MC, University Medical Center Rotterdam, Department of Public Health, Rotterdam, Netherlands

²¹Cancer Registry of Norway, Oslo, Norway

²²North Region Cancer Registry of Portugal (RORENO), Portuguese Oncology Institute of Porto (IPO Porto), Department of Epidemiology, Porto, Portugal

²³IPO Porto Research Center (CI-IPOP), Portuguese Oncology Institute of Porto (IPO Porto), Porto, Portugal

²⁴Institute of Biomedical Sciences Abel Salazar, University of Porto, Department of Population Studies, Porto, Portugal

²⁵South Region Cancer Registry of Portugal (ROR-Sul), Portuguese Oncology Institute of Lisbon, Lisbon, Portugal

²⁶South Regional Cancer Registry (ROR-Sul), Portuguese Oncology Institute of Lisbon, Lisbon, Portugal

²⁷Slovenian Cancer Registry, Institute of Oncology, Ljubljana, Slovenia

²⁸Basque Country Cancer Registry, Vitoria-Gasteiz, Spain

²⁹Epidemiology Unit and Girona Cancer Registry, Oncology Coordination Plan, Department of Health, Autonomous Government of Catalonia, Catalan Institute of Oncology, Girona, Spain

³⁰Descriptive Epidemiology, Genetics and Cancer Prevention Group, Biomedical Research Institute (IDIBGI), Salt, Spain

³¹Consortium for Biomedical Research in Epidemiology and Public Health (CIBER Epidemiología y Salud Pública, CIBERESP), Madrid, Spain

³²Epidemiology Unit and Girona Cancer Registry, Descriptive Epidemiology, Genetics and Cancer Prevention Group, IdIbGi, Catalan Institute of Oncology, Girona, Spain

³³Tarragona Cancer Registry, Epidemiology and Prevention Cancer Service, Hospital Universitari Sant Joan de Reus, Pere Virgili Health Research Institute, Reus, Spain

³⁴Murcia Regional Health Council, Department of Epidemiology, Murcia, Spain

³⁵Center for Primary Health Care, Lund University, Research Department of Clinical Sciences, Malmö, Sweden

³⁶Icahn School of Medicine at Mount Sinai, Department of Family Medicine and Community Health, Department of Population Health Science and Policy, New York, United States

³⁷Center for Community-based Healthcare Research and Education (CoHRE), School of Medicine, Shimane University, Department of Functional Pathology, Shimane, Japan

³⁸Cancer Registry Bern-Solothurn, Bern, Switzerland

³⁹Cancer Registry of Eastern Switzerland and Liechtenstein, St. Gallen, France

⁴⁰Cancer Registry of Eastern Switzerland and Liechtenstein, St. Gallen, Switzerland

⁴¹National Cancer Registry of Ukraine, National Institute of Cancer, Kiev, Ukraine

⁴²Taras Shevchenko National University of Kiev, Kiev, Ukraine

⁴³German Cancer Research Center (DKFZ), Division of Clinical Epidemiology and Aging Research, Heidelberg, Germany

⁴⁴German Cancer Research Center (DKFZ), Division of Preventive Oncology, Heidelberg, Germany

⁴⁵German Cancer Research Center (DKFZ), German Cancer Consortium (DKTK), Heidelberg, Germany

^✉

Contact E-Mail Address: rafael.cardoso@nct-heidelberg.de

Introduction

Colorectal cancer (CRC) screening with faecal occult blood test (FOBT), flexible sigmoidoscopy or colonoscopy has been shown to considerably reduce CRC mortality. Therefore, in recent years, screening programmes have been extensively implemented in European countries, however, with large differences in timing of enrolment, screening modalities offered and uptake rates. Since the effects of screening on CRC mortality are expected to be fully disclosed in the long run only, the assessment of intermediate endpoints such as CRC incidence and stage distribution are essential to timely monitor potential screening effects.

Aims & Methods

We aimed to analyse trends in CRC incidence and stage distribution with regards to the various CRC screening programmes that have meanwhile been introduced in European countries. We used population-based cancer registry data (CRC cases, ICD 10 codes: C18-C20, diagnosed from 2000 to 2016 or up to the most recent year with available data) from 16 European countries. We calculated age-standardised CRC incidence rates and analysed trends using Joinpoint regression. We investigated trends in stage distribution based on the UICC TNM classification in place at diagnosis. Furthermore, stratified analyses by tumour subsite (proximal colon, distal colon, and rectum) for both incidence and stage distribution were performed.

Results

In countries with earlier implementation of screening (e.g. Austria and Germany with colonoscopy programmes since 2005 and 2002, respectively, and with FOBT programmes for more than 30 years), CRC incidence has substantially decreased in recent years. Conversely, in countries with more recent implementation (e.g. Belgium in 2013 in all regions; the Netherlands in 2014; FOBT-based programmes) or where no large-scale screening programmes have been implemented yet (Bulgaria, Estonia, Finland, Norway, Portugal (North region), and Ukraine), CRC incidence either remained stable or increased over time. For the former, an apparent increase in CRC incidence immediately after introduction of screening was also observed. A similar pattern was seen for example for England and Slovenia; however, the slightly earlier screening implementation in these countries (2006 and 2009, respectively), allowed for investigation of trends for a longer time period with screening in place and, particularly, to observe a decline in incidence through the most recent years. For most countries offering screening, we also found an overall increase in the proportion of stage I CRCs in the years immediately after its implementation.

The described trends in incidence and stage distribution were essentially observed for distal colon cancer and rectal cancer and were much less pronounced for proximal colon cancer.

Conclusion

Trends in CRC incidence and stage distribution varied considerably across European countries and reflect, to a large extent, different levels of CRC screening implementation and participation rates. The downward trend in CRC incidence observed for countries with earlier screening implementation and the substantial increase in the proportion of early stage CRCs for countries which have recently introduced screening suggest that notable progress in the prevention of new cases and early detection by screening is already being achieved. with such trends, the first two necessary conditions for CRC mortality reduction are met. On the other hand, the lack of progress observed for countries where no large-scale screening programme is available call for developments towards its implementation.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.25

OP029 Participation in Competing Strategies For Colorectal Cancer Screening - A Randomized Health Services Study Within The National Screening Program in Poland (Piccolino Study)

N Pilonis ^1,^2,^3,^✉, M Bugajski ¹, P Wieszczy ⁴, M Pisera ⁵, M Rupinski ¹, E Pawlak ⁶, J Regula ⁷, MF Kaminski ⁶

Introduction

Primary colonoscopy and fecal immunochemical testing (FIT) are considered first tier tests for colorectal cancer (CRC) screening. Although colonoscopy is considered the most efficacious test, FIT may achieve a higher participation rate. It is uncertain what strategy is the best

for offering population wide CRC screening, however, an optimum strategy would ideally combine the high diagnostic yield of colonoscopy with the high participation rate of FIT.

Aims & Methods

This was a multicenter randomized health services study performed within the framework of the Polish Colonoscopy Screening Program (PCSP) between January 2019 and March 2020. Within the study, screening-naïve individuals between 55 and 64 years of age were drawn from the Population Registry. The candidates were randomly assigned in a 1:1:1 ratio to participate in one of three competing screening invitation strategies: (I. Control, practice as usual) postal invitation to screening colonoscopy only and a re-invitation to colonoscopy after 6 weeks for initial non-responders; (II. Sequential) postal invitation to primary colonoscopy, and then a postal invitation for screening using FIT (mailed FIT kit) after 6 weeks for non-responders to initial invitation; or (III. Active choice) postal invitation offering a choice between FIT (mailed FIT kit) and colonoscopy, and then a re-invitation after 6 weeks for initial non-responders offering the same choice again. The primary outcome was participation in CRC screening within 18 weeks after enrollment, defined as completion of colonoscopy, or completion of FIT (with subsequent colonoscopy in cases with positive FIT results). Secondary outcome was diagnostic yield for advanced neoplasia (CRC and advanced adenomas). FIT tests were analyzed in a central laboratory, with 10μg Hb/g feces positivity threshold. Active call-recall systems after positive FIT test result were established. Colonos-copy was performed as part of usual screening program practice. Study registration number: NCT03790475.

Results

Out of the 12,298 individuals, 514 (4.18%) were excluded due to previous diagnosis of CRC, death or unavailable postal address, leaving 11,785 for randomization and invitation to CRC screening. Participation rate in the control group (16.6%) was significantly lower as compared with the sequential group (24.8%) and active choice (26.4%) (p< 0.001 for both comparisons). in the control group, screening colonoscopy yielded participation comparable to the colonoscopy participation after the first invitation in sequential group (16.6% and 14.6%, respectively). in the active choice group, the percentage of individuals who chose colonoscopy was significantly lower (8.5%, p=0.001). FIT positivity rates were 9.3% and 9.4% in the sequential and active choice groups, respectively, and the diagnostic work-up colonoscopy rates for positive FITs were 55.5% and 73.3% respectively, despite active call-recall efforts. in the intention to screen analysis, advanced neoplasia detection rates per invitee were comparable between control (1.1%), sequential (1.0%) and active choice groups (1.1%), respectively.

Conclusion

Offering a combination of FIT and primary colonoscopy as a sequential or active choice strategy results in increased participation in CRC screening. Due to suboptimal compliance with diagnostic work-up colonoscopy after a positive FIT test, increased participation rates did not translate into higher diagnostic yield.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.26

OP030 The Comparison of Colorectal Adenomas, Intramucosal Cancers, and Invasive Cancers Detected Between Direct Screening Colonoscopy and Fit-Positive Colonoscopy

S Kimura ^1,^✉, S Fukuda ², D Nishiya ¹, F Ishibasi ¹, S Hayamizu ², T Arai ²

Introduction

There were few studies regarding the comparison of all colorectal neoplasms detected between direct screening colonoscopies and FIT-positive colonoscopies. We tried to compare the prevalences and the clinicopathological characteristics of colorectal adenomas, intramu-cosal cancers, and invasive cancers detected between asymptomatic subjects undergoing direct screening colonoscpy and subjects undergoing colonoscopy for the evaluation of a positive FIT.

Aims & Methods

The study included 1793 patients (mean age 66.4+-11.9 yr, M:F = 1090:703) underwenting colonoscopy for asymptomatic direct screening (group A), and 2227 (65.4+-11.6 yr, 1306:921) underwenting colonoscopy for screening as the evaluation of a positive FIT (group B). The prevalences of colorectal adenomas and intramucosal cancers more than 5 mm in size detected between group A and B, and clinicopathologi-cal characters as to the location, size, configulation, and histology of them were compared with each other. Subsequently the prevalences of colorectal invasive cancers detected between group A and B, and clinicopatholog-ical characters as to the location, size, depth, histology, and lymph node metastasis of them were compared with each other.

Results

A total of 573 adenomas and/or intramucosal cancers in 338 patients (18.8%) in group A and 1071 in 650 (29.2%) in group B were detected, respectively. The locations of the lesions between group A and B were 47.6%, 37.5% in the right colon, 52.4%, 62.5% in the left colon. The sizes of the lesions were < 10 mm 64.2%, 64.7% and > 10 mm 35.8%, 35.3%. The configulations of the lesions were polypoid 85.0%, 88.8%, and non-polypoid 15.0%, 11.2%. The histologies of the lesions were low-grade adenoma 82.4%, 77.6% and high-grade adenoma or cancer 17.6%, 22.4%. The statistical comparisons disclosed that adenomas and intramucosal cancers detected in group A were characterized by their right-sided location (p< 0.001), non-polypoid appearance (p< 0.05), and less-advanced histology (p< 0.05). in contrast the lesions detected in group B were characterized by their left-sided location (p< 0.001), polypoid appearance (p< 0.05), and more-advanced histology (p< 0.05). A total of 28 invasive cancers in 28 patients (1.6%) in group A and 115 in 115 (5.2%) in group B were detected, respectively. The locations of the lesions between group A and B were 53.6%, 34.8% in the right colon, and 46.4%, 65.2% in the left colon. The sizes of the lesions were < 30 mm 50.0%, 52.2%, and > 30 mm 50.0%, 47.8%. The depths of the lesions were T1 46.4%, 38.3%, and T2 or T3 53.6%, 61.7%. The histologies of the lesions were well differentiated adenocarcinoma 64.3%, 58.3%, and poorly or moderately differentiated 35.7%, 41.7%. The prevalences of lymph node involvements were 21.4%, 26.9%, respectively. There were no statistical differences in clinicopatho-logical factors of detected invasive cancers between group A and B except the locations of the lesions; Invasive cancers in group A in the right colon and those in group B in the left colon (p= 0.07).

Conclusion

FIT-positive colonoscopy detected a significantly higher amount of precancerous lesions when compared with direct screening colonoscopy. Adenomas and intramucosal cancers detected by FIT were characterized by their left-sided location, polypoid appearance, and more-advanced histology. FIT-positive colonoscopy and direct screening

colonoscopy, both of them dominantly detected invasive cancers of clinically early stage and FIT-positive colonoscopscopy detected them more effectively, although the locations of the lesions were diferrent.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.27

OP031 R.Intestinalis Ameliorates Gut Inflammation in Inflammatory Bowel Disease Via Toll Like Receptor 5 Expressing On Iecs and Mediated Tslp Production and Tregs Differentiation

W Luo ^1,^2,^✉, Z Shen ¹, X Wang ^1,², B Tan ¹, M Deng ¹, M Xiao ^1,², X Wu ^1,², K Nie ^1,², X Meng ^1,², S Wu ^1,², T Tong ^1,²

Introduction

Inflammatory bowel disease (IBD), including Ulcerative colitis (UC) and Crohn's disease (CD), is an immune-related chronic recurrent intestinal inflammatory disease. The onset of IBD is closely related to intestinal microbial disorders. Our previous research showed that in patients with newly diagnosed CD, the abundance and diversity of intestinal microbiota was decreased compared to healthy donors, among which Roseburia. intestinalis (R. intestinalis) was significantly reduced. We found that R. intestinalis is able to inhibit TNBS-induced colitis and enhance Treg cells differentiation. Here, we further investigated the anti-inflammatory mechanism of R. intestinalis. Intestinal epithelial cells (IECs) are intermediary bridges connecting intestinal bacteria and gut homeostasis and disease. TLRs are the most representative pattern recognition receptors on the surface of IECs in which TLR2, TLR4 and TLR5 are related to bacterial recognition. TLR-expressing IECs produce factors that alter the phenotypes of DCs and the differentiation of T cells. Moreover, intestinal mucosal DCs, producing IL-10 and TGFβ, promote the development and differentiation of Treg cells.

Aims & Methods

In order to explore whether TLR-expressing IECs and are necessary for R. intestinalis to induce DCs and Treg cell differentiation, we used LPS-stimulated human intestinal epithelial cell line, Caco-2, as colon inflammatory cell model with or without TLR5-siRNA intervention in the presence of R. intestinalis and we incubated human monocyte-derived DCs using supernatant of LPS- treated Caco-2 cells; then we cocultured the human CD4+ T cells with above DCs to determine the development of Tregs using flow cytometry in vitro. Meanwhile, we sought to detect the effects of R.intestinalis on colitis indications, the expressions of TSLP, TGFβ and the differentiation of Treg cells using murine TNBS colitis of bone marrow chimera mice (generating with wild-type (WT) and Tlr5^-/- mice). Besides, to investigate whether TSLP and TGFβ expressed by IECs are the key factors for R. intestinalis to induce DCs and Treg cell differentiation, Caco-2 cells were transfected with TSLP siRNA/TGFβ siRNA with or without recombinant TSLP/TGFβ in the presence of LPS and R. intestinalis and then DCs were incubated with the supernatant of Caco-2 cells in vitro, and then we co-cultured Th0 cells with washed DCs described above. At the same time, we neutralized endogenous TSLP/TGFβ using a neutralizing anti-TSLP/anti-TGFβ mAb in C57BL/6 WT mice treated with TNBS and R.intestinalis.

Results

We found that TLR5 expressing intestinal epithelial cells (IECs) play an important role for R. intestinalis to suppress inflammation, and radiation bone marrow (BM) chimeric mice generated with TLR5-deficient (Tlr5^-/-) mice and wild-type (WT) mice were used to confirm that R. intestinalis promotes intestinal epithelial cells to secrete anti-inflammatory factors through TLR5, of which TSLP is an important mediator to promote DC-induced Treg cell differentiation.

Conclusion

Collectively, we revealed that R. intestinalis induced TSLP secretion from TLR5-expressing IECs, and then promoted DCs-induced Treg cell differentiation. Our research provides more theoretical basis to verify R. intestinalis as symbiotic bacterium that inhibits intestinal inflammation and lays a foundation for its application in future clinical research.

Disclosure

Nothing to disclose

References

1.de Souza H.S.P., Fiocchi C., Iliopoulos D. The IBD interactome: an integrated view of aetiology, pathogenesis and therapy. Nat Rev Gas-troenterol Hepatol. 2017; 14(12): 739–49. [DOI] [PubMed] [Google Scholar]
2.Caruso R., Lo B.C., Núñez G. Host-microbiota interactions in inflammatory bowel disease. Nature Reviews Immunology. 2020; 10.1038/s41577-019-0268-7. [DOI] [PubMed] [Google Scholar]
3.Mishima Y., Sartor R.B. Manipulating resident microbiota to enhance regulatory immune function to treat inflammatory bowel diseases. J Gastroenterol. 2020; 55(1): 4–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Belkaid Y., Harrison O.J. Homeostatic Immunity and the Microbiota. Immunity. 2017; 46(4): 562–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Burgueno J.F., Abreu M.T. Epithelial Toll-like receptors and their role in gut homeostasis and disease. Nat Rev Gastroenterol Hepatol. 2020. [DOI] [PubMed] [Google Scholar]
6.Gay N.S. M.F., Gangloff M., Bryant C.E. Assembly and localization of Toll-like receptor signalling complexes. Nat Rev Immunol. 2014. 14(8): 546–58. [DOI] [PubMed] [Google Scholar]
7.Price A.E., Shamardani K., Lugo K.A. et al. A Map of Tolllike Receptor Expression in the Intestinal Epithelium Reveals Distinct Spatial, Cell Type-Specific, and Temporal Patterns. Immunity. 2018; 49(3): 560–75.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Shen Z., Zhu C., Quan Y. et al. Insights into Roseburia intestinalis which alleviates experimental colitis pathology by inducing anti-inflammatory responses. J Gastroenterol Hepatol. 2018; 33(10): 1751–60. [DOI] [PubMed] [Google Scholar]
9.Duncan S.H. Roseburia intestinalis sp. nov., a novel saccharolytic, butyrate-producing bacterium from human faeces. International Journal of Systematic and Evolutionary Microbiology. 2002; 52(5): 1615–20. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.28

OP032 Critical Role of Ormdl Proteins in The Regulation of Autophagy in Intestinal Homeostasis

S Stengel ¹, F Tran ^1,^2,^✉, M Jentzsch ¹, M Falk-Paulsen ¹, J Kuiper ¹, D Bordoni ¹, B Messner ¹, S Imm ¹, K Aden ^1,², P Rosenstiel ¹

Introduction

ORM1-like protein 1, 2 and 3 (ORMDL1, ORMDL2, ORMDL3) belong to a family of endoplasmic reticulum (ER)-resident proteins which have been implicated in ER homeostasis and sphingolipid biosynthesis. Genome-wide association studies identified susceptibility loci in or adjacent to ORMDL genes for inflammatory diseases such as asthma and inflammatory bowel disease (IBD). Preliminary studies suggested a role of ORMDL proteins in orchestrating the unfolded protein response (UPR) upon ER stress. However, the exact function of these proteins in IBD remains enigmatic.

Aims & Methods

Here, we set out to investigate the role of ORMDL proteins in intestinal inflammation by generating mice deficient for one or more ORMDL proteins. We studied the effect of such knockouts in vivo and in vitro (by generating intestinal organoids) on autophagy function and their functional consequences in intestinal epithelial cells, such as quantification of secretory cells like Paneth cells and Goblet cells.

Results

Mice deficient for two ORMDL proteins display a strongly reduced body weight. Mice with triple knockout of Ormdl1/2/3 are embryonically lethal. Histological assessment of the small intestine revealed a reduced

number of Paneth cells and Goblet cells in Ormdl1- and Ormdl3-knockout animals, while this reduced was strongly reduced in mice with Ormdl1/3-double deficiency. This correlated with reduced production of antimicrobial peptides and reduced bacterial killing. Mechanistically, loss of ORMDL proteins lead to accumulation of autophagy proteins, suggesting that ORMDL proteins are involved in the autophagy machinery. ORMDL3 colo-calises and coprecipitates with LC3, indicating a direct protein interaction with autophagy proteins.

Conclusion

Our data suggest a critical regulatory role of ORMDL proteins in the autophagy machinery in the intestinal epithelium. Loss of ORDML proteins lead to autophagy defects, which result in inflammatory pheno-type due to secretory defects of epithelial cells.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.29

OP033 Setdb1 Maintains Intestinal Epithelial Genome Stability and Prevents Intestinal Inflammation

L Južnic ^1,^2,^✉, K Peuker ^1,², A Strigli ^1,², M Brosch ^1,², A Herrmann ^1,², R Häsler ³, M Koch ^1,², L Matthiesen ^1,², Y Zeissig ⁴, B-S Löscher ³, A Nuber ⁵, G Schotta ⁵, V Neumeister ⁶, T Chavakis ⁶, T Kurth ⁷, M Lesche ⁸, A Dahl ⁸, A von Mässenhausen ^10,⁹, A Linkermann ^10,⁹, S Schreiber ^1,³, K Aden ^1,³, P Rosenstiel ³, A Franke ³, J Hampe ^1,², S Zeissig ^1,²

Introduction

The intestinal epithelium is a complex, constitutively renewing tissue composed of functionally distinct intestinal epithelial cells (IECs), whose specific cell fates are established and maintained through cell-specific activity of transcription factors, as well as precise gene silencing by chromatin compaction. By facilitating chromatin condensation through histone modification, SETDB1, a histone-lysine N-methyltransfer-ase, is one of the central factors involved in regulation of epigenetic transcriptional repression. Therefore, we set out to study the potential roles of SETDB1 in intestinal epithelial homeostasis and inflammatory bowel disease (IBD).

Aims & Methods

To elucidate the role of SETDB1 in intestinal homeostasis and pathogenesis of IBD we studied mice with constitutive (Villin-Cre) or tamoxifen (TAM)-inducible (Villin-Cre^ERT2) IEC-specific deletion of Setdb1, and further evaluated the expression of Setdb1 in mouse models of IBD, specifically the dextran sulfate sodium (DSS) model of chemically induced colitis and the TNF^MRE model of ileitis. Additionally, we evaluated the expression of SETDB1 in IBD patients and investigated the abundance of SETDB1 variants in healthy individuals and IBD patients.

Results

Constitutive loss of Setdb1 in IECs was associated with embryonal lethality, however surviving mice partially escaped Cre-mediated Setdb1 deletion and showed altered epithelial morphology and proliferation in regions without Setdb1 expression. TAM-induced IEC-specific deletion of Setdb1 in adult mice was associated with altered intestinal morphology, defects in IEC differentiation, barrier dysfunction, intestinal inflammation and mortality within 10 days after TAM injection start. Mechanistically, loss of Setdb1 was associated with de-repression of endogenous retroviruses and consequent genomic instability, DNA damage and p53 accumulation, activation of type I interferon signaling and IEC cell death. Intestinal epithelial SETDB1 expression was largely unaltered in mouse models of IBD and in IBD patients. Nevertheless, rare missense and predicted loss-of-function variants of SETDB1 are over-represented in IBD patients compared to healthy individuals.

Conclusion

Our study unveils a critical role of SETDB1 in maintenance of genome stability in IECs, and consequently intestinal epithelial homeostasis and prevention of intestinal inflammation. As certain rare variants in SETDB1 are over-represented in IBD patients, we continue to investigate if these variants can contribute to IBD pathogenesis, potentially in a monogenic manner.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.30

OP034 Microbiome Dysbiosis, Intestinal Epithelial Cell Dysfunction and Immune System Deregulation in Patients with Crohn'S Disease

Tabib NS Seyed ^1,^✉, C Caenepeel ¹, K Machiels ¹, S Verstockt ¹, B Verstockt ², Davani N Ardeshir ¹, J Sabino ², M Ferrante ², S Vermeire ²

Introduction

The perturbation of composition, function, and structure of the gut microbiota known as dysbiosis is a key factor in inflammatory bowel disease (IBD) pathogenesis. There is a crosstalk between the micro-biota, intestinal epithelial cells, and the gut immunological niche, resulting in a heterogeneous clinical phenotype.

Aims & Methods

To better understand this interaction, we characterized the degree of dysbiosis and dysregulation of the immune proteome in Crohn's disease (CD) patients to see if subtypes of patients could be identified. We collected faecal, ileum biopsies, and serum samples of 146 CD patients and 63 healthy controls (HC),and studied microbiota phylogenetic (16S rRNA gene sequencing), intestinal epithelial cell(RNA sequencing), and serum proteomic (91 inflammatory proteins OLINK). Microbial dysbi-otic index (MDI)¹, defined as the logarithm of the sum of [abundance in organisms increased in CD] over the [abundance of organisms decreased in CD] was calculated and patients were ranked from Q1(the least dysbiotic state) to Q4(the most dysbiotic state). For the proteomic score, 32 proteins that correlated (adj. p < 0.01) with faecal calprotectin (FC) were selected. A penalized logistic regression model was trained on these proteins, to distinguish HC from super active (defined as FC > 1800 μg/g). We next developed an inflammatory proteomic score (IPS) defined as the weighted

sum of serum level of inflammatory proteins, using the coefficient value of the regression model as the protein's weight. Using the IPS score, patients were clustered from Q1(the least inflammatory state) to Q4(the most inflammatory state). Similar to the proteomic score, a penalized logistic regression model was trained to differentiate CD patients from HC, on a pre-selected intestinal barrier related gene. We next developed a barrier integrity score (BIS) defined as the weighted sum of serum level of barrier genes, using the coefficient value of the regression model as the gene's weight. Using the BIS score, patients were clustered from Q1(the least dysfunction state) to Q4(the most dysfunction state). Statistical analyses were performed in R 3.5.2.

Results

The MDI did not correlate with standard phenotypic subgroups based on the Montreal classification but did positively correlate with C-re-active protein (CRP) and FC level (p< 0.001). The regression model identified 14 proteins distinguishing super active CD patients from HC with accuracy of 95.6%, respectively. IPS positively correlated with CRP and FC level (p< 0.001). Likewise, MDI and IPS-based clusters were significantly different in CRP and FC levels. Different components of the microbiome correlated with the proteome in a subset of samples. For example, fibroblast growth factor 19 (FGF-19) positively correlated with Faecalibacterium and negatively with Fusobacterium. of note, we observed a significant positive correlation between MDI and IPS (r=0.33, p< 0.001). BIS is significantly different in CD patients compare to HC (p< 0.001), BIS did not correlate with Montreal classification, CRP and FC levels.

Conclusion

We were able to define clusters of patients based on molecular characterization of different players in IBD pathogenesis such as microbiota, transcriptome, and proteome. This molecular clustering in a given patient could be considered as a novel therapeutic and personalized approach to IBD. Further validation in larger cohorts is required.

Disclosure

S. Vermeire: research grant: MSD, AbbVie, Takeda, Pfizer, J&J; lecture fee: MSD, AbbVie, Takeda, Ferring, Centocor, Hospira, Pfizer, J&J, Genentech/Roche; consultancy: MSD, AbbVie, Takeda, Ferring, Centocor, Hospira, Pfizer, J&J, Genentech/ Roche, Celgene, Mundipharma, Cell-trion, SecondGenome, Prometheus, Shire, Prodigest, Gilead, Galapagos. J. Sabino: Speaker's fee: Abbvie and Nestle Health Sciences. M. Ferrante: Research grant: Amgen, Biogen, Janssen, Pfizer, Takeda; Consultancy: Abbvie, Boehringer-Ingelheim, Janssen, MSD, Pfizer, Sandoz, Takeda; Speaker's fee: Abbvie, Amgen, Biogen, Boehringer-Ingelheim, Falk, Fer-ring, Janssen, Lamepro, MSD, Mylan, Pfizer, Takeda. B. Verstockt: research grant: Pfizer; lecture fees: Abbvie, Ferring Pharmaceuticals, Janssen, R-Biopharm and Takeda; consultancy fees: Janssenand Sandoz

References

1.Gevers D. et al. The treatment-naive microbiome in new-onset Crohn's disease. Cell Host Microbe 15, 382–392 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.31

OP035 Colibactin-Producing E. Coli Induces A Metabolic Reprogramming of Colonic Epithelial Cells Toward A Pro-Carcinogenic Metabolism

R Villager ^1,^✉, B Diémé ², M Lagrée ², J Delmas ³, G Roche ¹, N Barnich ¹, C Jousse ², M Bonnet ¹

Introduction

Colonization of the colonic mucosa by pro-carcinogenic E. coli could be involved in the development of colorectal cancer (CRC), especially through the production of genotoxins such as colibactin. Colibac-tin is produced by an enzymatic machinery encoded by the pks genomic island. Intracellular multiplication of colibactin-producing E. coli (CoPEC) likely requires the exploitation of host resources to replicate and survive which could alter the metabolic states of intestinal epithelial cells leading to transformation.

Aims & Methods

The aim of this study is to investigate the effect of CoPEC infection on the metabolic state of human colonic epithelial cells. For this purpose, human colorectal carcinoma T84 cells were infected with the 11G5 strain, a CoPEC strain isolated from a human CRC sample, or its mutant (11G5-zlpks). Expression of genes coding for colibactin enzymes synthesis (clb) was evaluated by RT-qPCR after 3h, 24h and 48h following infection. Intracellular metabolomes were investigated by untargeted metabolomics approach using ¹H-NMR and LC-MS, in positive and negative modes (PFEM). Extracellular metabolomes were analyzed by targeted metabolomics analysis using LC-MS (Biocrates). Untargeted metabolomic data were analyzed using XCMS package on Galaxy-W4M platform.

Results

CoPEC 11G5 strain was found intracellularly 3h post-infection, then persisted and multiplied in cells during the following 24h and 48h. A higher expression of clb genes was observed after 24h of infection when compared to 3h and 48h. Principal component analysis revealed strong discrimination between T84 non-infected cells and cells infected with colibactin-positive E. coli, but also between cells infected with colibactin-negative E. coli versus colibactin-positive E. coli. Our preliminary results showed an increase of lactate and pyruvate concentrations in 11G5 and 11G5-zlpks-infected cells when compared to control cells. This result revealed a disruption of glycolytic metabolism, suggesting the induction of a Warburg-like metabolism in infected cells. Analysis of the extracellular metabolome supported this result. To highlight the specific role of colibactin in the metabolic reprogramming induced by CoPEC, we then compared the intracellular metabolome of cells infected with 11G5 CoPEC strain or the colibactin-negative mutant 11G5-zlpks. Discriminant metabolites between the colibactin-positive and the colibactin-negative groups suggest the specific effect of colibactin on the disturbance of colonic cell metabolome. Preliminary results showed us that colibactin affects mostly metabolites involved in host defense against DNA-damages and oxidative stress, with increased intracellular concentrations of glutathione, UDP-GlcNAc and pantothenic acid, and decreased intracellular and extracellular concentrations of serine, which could in turn contribute to CoPEC pro-carcinogenic properties.

Conclusion

Altogether, our work supports the fact that CoPEC-induced metabolic reprogramming in colonic epithelial cells could participate to its pro-carcinogenic activity and provides new molecular targets for CRC treatment.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.32

OP036 A Randomized Trial of Fluid Hydration To Prevent Post-Ercp Pancreatitis (Fluyt)

Weiland CJ Sperna ^1,^2,^✉, XJNM Smeets ^1,², W Kievit ¹, RC Verdonk ³, AC Poen ⁴, A Bhalla ⁵, NG Venneman ⁶, BJM Witteman ⁷, DW da Costa ⁸, BC van Eijck ⁹, MP Schwartz ¹⁰, TEH Römkens ¹¹, JM Vrolijk ¹², M Hadithi ¹³, AMCJ Voorburg ¹⁴, LC Baak ¹⁵, WJ Thijs ¹⁶, RL van Wanrooij ¹⁷, ACITL Tan ¹⁸, TCJ Seerden ¹⁹, YCA Keulemans ²⁰, TR de Wijkerslooth ²¹, W van de Vrie ²², P van der Schaar ³, SM van Dijk ^2,²³, NDL Hallensleben ^2,²⁴, Weiland RL Sperna ²⁵, HC Timmerhuis ², DS Umans ^2,¹⁷, JE van Hooft ^26,²⁷, H van Goor ²⁸, HC van Santvoort ^29,³⁰, MG Besselink ²³, MJ Bruno ³¹, P Fockens ²⁶, JPH Drenth ¹, EJM Van Geenen ¹; Dutch Pancreatitis Study Group

¹Radboudumc, Gastroenterology and Hepatology, Nijmegen, Netherlands

²St Antonius Hospital, Research and Development, Nieuwegein, Netherlands

³St Antonius Hospital, Gastroenterology and Hepatology, Nieuwegein, Netherlands

⁴Isala Clinics, Gastroenterology and Hepatology, Zwolle, Netherlands

⁵Hagaziekenhuis, Gastroenterology and Hepatology, The Hague, Netherlands

⁶Medisch Spectrum Twente, Gastroenterology and Hepatology, Enschede, Netherlands

⁷Gelderse Vallei Hospital, Gastroenterology and Hepatology, Ede, Netherlands

⁸St Antonius Hospital, Radiology, Nieuwegein, Netherlands

⁹Spaarne Gasthuis, Gastroenterology and Hepatology, Hoofddorp, Netherlands

¹⁰Meander Medical Center, Gastroenterology and Hepatology, Amersfoort, Netherlands

¹¹Jeroen Bosch Hospital, Gastroenterology and Hepatology, Den Bosch, Netherlands

¹²Rijnstate Hospital, Gastroenterology and Hepatology, Arnhem, Netherlands

¹³Maasstad Hospital, Gastroenterology and Hepatology, Rotterdam, Netherlands

¹⁴Diakonessenhuis, Gastroenterology and Hepatology, Utrecht, Netherlands

¹⁵Onze Lieve Vrouwe Gasthuis, Gastroenterology and Hepatology, Amsterdam, Netherlands

¹⁶Martini ziekenhuis, Gastroenterology and Hepatology, Groningen, Netherlands

¹⁷Amsterdam UMC, Gastroenterology and Hepatology, Amsterdam, Netherlands

¹⁸Canisius Wilhelmina ziekenhuis, Gastroenterology and Hepatology, Nijmegen, Netherlands

¹⁹Amphia Hospital, Gastroenterology and Hepatology, Breda, Netherlands

²⁰Zuyderland Hospital, Gastroenterology and Hepatology, Heerlen, Netherlands

²¹Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Gastroenterology and Hepatology, Amsterdam, Netherlands

²²Albert Schweitzer Hospital, Gastroenterology and Hepatology, Dordrecht, Netherlands

²³Amsterdam UMC, Surgery, Amsterdam Gastroenterology Endocrinology Metabolism, Amsterdam, Netherlands

²⁴Erasmus Medical Center, Anesthesiology, Rotterdam, Netherlands

²⁵University of Amsterdam, Amsterdam, Netherlands

²⁶Amsterdam UMC, Gastroenterology and Hepatology, Amsterdam Gastroenterology Endocrinology Metabolism, Amsterdam, Netherlands

²⁷Leiden University Medical Center, Gastroenterology and Hepatology, Leiden, Netherlands

²⁸Radboudumc, Surgery, Nijmegen, Netherlands

²⁹St Antonius Hospital, Surgery, Nieuwegein, Netherlands

³⁰University Medical Center Utrecht, Surgery, Utrecht, Netherlands

³¹Erasmus Medical Center, Gastroenterology and Hepatology, Rotterdam, Netherlands

^✉

Contact E-Mail Address: christa.spernaweiland@radboudumc.nl

Introduction

Pancreatitis is the most common complication of endoscopic retrograde cholangiopancreatography (ERCP). Rectal non-steroidal anti-inflammatory drugs (NSAID) reduce post-ERCP pancreatitis rates and are current standard-of-care as per European and American guidelines. A number of studies suggest that intensive periprocedural hydration is effective and safe in reducing post-ERCP pancreatitis. However, patients in these studies did not receive rectal NSAID. It is therefore unclear whether periprocedural hydration affords additional protection against post-ERCP pancreatitis in a population treated with rectal NSAIDs. Thus, we performed a randomized trial to investigate whether periprocedural hydration with Lactated Ringer's solution can prevent post-ERCP pancreatitis in average- to high-risk patients undergoing ERCP who already receive prophylactic rectal NSAID.

Aims & Methods

In this multicenter, randomized controlled superiority trial we assigned patients with an average- to high-risk of post-ERCP pancreatitis to receive a combination of intensive hydration and rectal NSAID (hydration group) or rectal NSAID monotherapy (control group). Intensive hydration comprised 20ml/kg Lactated Ringer's intravenously within 60 minutes from the start of ERCP, followed by 3ml/kg/h for 8 hours thereafter. The control group received normal saline with a maximum of 1.5ml/ kg/h and 3L/24h. The primary endpoint was incidence of post-ERCP pancreatitis. Secondary endpoints included post-ERCP pancreatitis severity and ERCP- and hydration-related complications.

Results

A total of 826 patients were randomized. Patient baseline and ERCP characteristics were similar in both groups. Post-ERCP pancreatitis developed in 30 of 388 patients (7.7%) in the hydration group and in 39 of 425 patients (9.2%) in the control group (relative risk, 0.84; 95% confidence interval [0.53-1.33]; P=0.53). Moderate-to-severe pancreatitis developed in 21 patients (5%) in the hydration group and in 32 patients (8%) in the control group (P=0.39). ERCP- and hydration-related complications did not differ between both groups (P=0.6 and P=1.0, respectively). The 30-day mortality rate was comparable between groups. None of the deaths were related to the study intervention.

[Primary and secondary outcomes]

	Hydration group (N=388)	Control group (N=425)	P Value
Post-ERCP pancreatitis - no. (%)	30 (7.7)	39 (9.2)	0.53
Post-ERCP pancreatitis severity Cotton - no. (%)			0.39
Mild	9 (2)	7 (2)
Moderate + severe	21 (5)	32 (8)
Post-ERCP pancreatitis severity Atlanta - no. (%)			0.26
Mild	27 (7)	29 (7)
Moderate + severe	3 (1)	10 (2)
Exocrine insufficiency - no.	0	2	0.25
Endocrine insufficiency - no.	4	3	0.49
ERCP-related complications - no. (%)	16 (4)	21 (5)	0.6
Infection (cholangitis)	2	6	0.44
Bleeding	10	12	0.53
Perforation	5	4	0.49
Hydration-related complication - no. (%)	8 (2)	9 (2)	1
Pulmonary edema	3	5	0.8
Peripheral edema	6	3	0.22
Cardiac insufficiency	0	3	0.25
Hypernatremia	1	1	1
30 day mortality - no. (%)	5 (1)	6 (1)	1
Mortality during 180 days of follow-up - no. (%)	11 (3)	9 (2)	0.65

Open in a new tab

Conclusion

The combination of rectal NSAIDs and intensive peripro-cedural hydration does not reduce the incidence of post-ERCP pancreatitis, as compared to rectal NSAID monotherapy in patients with average- to high- risk of post-ERCP pancreatitis (ISRCTN registry, Identifier ISRCTN13659155).

Disclosure

Funded by the Netherlands Organization for Health Research and Development (ZonMw) and Radboud university medical center

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.33

OP037 Endotherapy of Chronic Pancreatitis Related Benign Biliary Strictures, Plastic Or Metal? An International Randomized Controlled Trial

M Ramchandani ^1,^✉, S Lakhtakia ¹, G Costamagna ², A Tringali ², A Püspök ³, B Tribl ³, W Dolak ³, J Deviere ⁴, M Arvanitakis ⁴, S van der Merwe ⁵, W Laleman ⁵, T Ponchon ⁶, A Gabbrielli ⁷, MJ Bruno ⁸, U Arnelo ⁹, JYW Lau ¹⁰, A Roy ¹¹, MJ Bourke ¹², AJ Kaffes ¹³, H Neuhaus ¹⁴, J Peetermans ¹⁵, M Rousseau ¹⁵, DN Reddy ¹

Introduction

Benign biliary stricture (BBS) is a complication of chronic pancreatitis (CP). Endotherapy using a single removable fully covered self-expanding metal stent (FCSEMS) or multiple plastic stents (MPS) are acceptable options for patients with biliary obstructive symptoms associated with such BBS. Both options have published long-term results in single arm studies^1,2 and in one randomized controlled study (RCT)³. The need for additional long term RCT data was highlighted in ESGE⁴ and ASGE^5,6 guidelines.

Aims & Methods

Patients with a symptomatic BBS secondary to CP were enrolled at 14 centers in 11 countries in an RCT comparing treatment with one biliary FCSEMS (Fully Covered WallFlex Biliary Stent, Boston Scientific, Marlboro, MA, USA) placed for 12 months compared to 3 or 4 episodes of side-by-side MPS placed for the same cumulative 12 month duration. The primary endpoint was stricture resolution 24 months after initial stent(s) placement, defined as absence of re-stenting after the per-protocol stent-ing period and/or alkaline phosphatase level at 24 months below twice the level at the end of the per-protocol stenting period. Secondary endpoints included adverse events related to stents, stent placement or stent removal, rate of cross-overs, and number of ERCPs during 24 months after initial stent placement.

Results

Enrollment was completed with 80 patients randomized to the FCSEMS arm and 84 to the MPS arm. No significant differences in baseline variables were found. Mean age was 51.8±10.4 and mean years since first diagnosis of CP was 5.1±7.4. At the time of randomization 43.9% (72/164) patients had a prior single plastic stent indwelling and bilirubin level was 2.8±3.8 mg/dL. Technical success at baseline was 96.2% (76/79) for the FCSEMS and 98.8% (80/81) for the MPS arm. During follow-up there were crossovers from FCSEMS to MPS in 8 patients and from MPS to FCSEMS in 11 patients. On an intent-to-treat basis, based on available data to date, stricture resolution at 24 months after randomization was 71.4% (40/56) in the FCSEMS arm and 73.8% (45/61) in the MPS arm. The mean number of ERCPs during 24 months after randomization was 2.5±1.2 in the FCSEMS arm and 4.1±1.2 in the MPS arm. The mean total number of stents placed was 2.4±3.4 in the FCSEMS arm and 7.6±4.6 in the MPS arm. Serious related adverse events occurred in 18 (22.5%) of patients in the FCSEMS arm and in 15 (17.9%) in the MPS arm, mainly cholangitis/jaundice, abdominal pain, cholecystitis, pancreatitis and perforation (respectively 7, 4, 3, 3,0 in the FCSEMS arm and 8, 5, 1, 0, 1 in the MPS arm). There were no stent or procedure related deaths.

Conclusion

Endotherapy of BBS secondary to CP have similar effectiveness and safety when using a single FCSEMS compared to MPS for one year, with the FCSEMS option requiring fewer ERCPs over two years after initial stenting.

Disclosure

The following authors disclosed financial relationships relevant to this publication: G. Costamagna: Grants from Cook Endoscopy, Olympus, and Boston Scientific; member on advisory committees or review panels for Boston Scientific and Olympus. A. Tringali: Consultancy agreements with Boston Scientific. B. Tribl: Consultancy fees from Boston Scientific and AiCuris. J. Devière: Research support from Boston Scientific, Cook Medical, and Olympus; shareholder in Endotools. S. van der Merwe: Grants from Cook Medical and Boston Scientific. W. Laleman: Consultancy agreements with Boston Scientific and Cook Medical. T. Ponchon: Grants from Boston Scientific; personal fees from Olympus, Boston Scientific, Norgine, Fujifilm, and Ipsen. M. Bruno: Grants from Boston Scientific, Cook Medical, 3M, and Pentax Medical; personal fees from Boston Scientific, Cook Medical, 3M, Pentax Medical, Mylan, and SOCAR. M. Bourke: Research support from Cook Medical, Olympus, and Boston Scientific. A. Kaffes: research grants from Boston, Cook Medical, and Olympus. H. Neuhaus: Consultant and speaker for Boston Scientific. J. Peetermans, M. Rousseau: Employee of Boston Scientific. All other authors disclosed no financial relationships relevant to this publication.

References

1.Draganov P., Hoffman B., Marsh W. et al. Long-term outcomes in patients with benign biliary strictures treated endoscopically with multiple stents. Gastrointest Endosc 2002; 55: 680–686 [DOI] [PubMed] [Google Scholar]
2.Lakhtakia S, Reddy N., Dolak W. et al. Long-term outcomes after temporary placement of a self-expanding fully covered metal stent for benign biliary strictures secondary to chronic pancreatitis. Gastrointestinal Endoscopy. 2020; 91: 361–369 [DOI] [PubMed] [Google Scholar]
3.Haapamäki C., Kylänpää L., Udd M. et al. Randomized multicenter study of multiple plastic stents vs. covered self-expandable metallic stent in the treatment of biliary stricture in chronic pancreatitis. 2015; 47: 605–10 [DOI] [PubMed] [Google Scholar]
4.Dumonceau J., Tringali A., Papanikolaou I. et al. Endoscopic biliary stent-ing: indications, choice of stents, and results: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline - Updated October 2017. Endoscopy. 2018; 50: 910–930 [DOI] [PubMed] [Google Scholar]
5.Chandrasekhara V., Chathadi K., Acosta R. et al. The role of endoscopy in benign pancreatic disease. Gastrointestinal Endoscopy. 2015; 82: 203–14 [DOI] [PubMed] [Google Scholar]
6.Chathadi K., Chnadrasekhara, Acosta R. et al. The role of ERCP in benign diseases of the biliary tract. Gastrointestinal Endoscopy. 2015: 81: 795–803. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.34

OP038 Endoscopic Papillectomy: A Delphi Consensus

JA Fritzsche ^1,^✉, P Fockens ¹, M Barthet ², MJ Bruno ³, DL Carr-Locke ⁴, G Costamagna ⁵, GA Cote ⁶, PH Deprez ⁷, M Giovannini ⁸, G Haber ⁹, RH Hawes ¹⁰, JJ Hyun ¹¹, T Itoi ¹², E Iwasaki ¹³, L Kylänpää ¹⁴, H Neuhaus ¹⁵, JY Park ¹⁶, DN Reddy ¹⁷, A Sakai ¹⁸, MJ Bourke ^20,¹⁹, RP Voermans ¹

¹Amsterdam University Medical Centers, University of Amsterdam, Department of Gastroenterology and Hepatology, Amsterdam, Netherlands

²Assistance Publique des Hôpitaux de Marseille, Aix-Marseille Université, Hôpital Nord, Department of Hepatogastroenterology, Marseille, France

³Erasmus MC -University Medical center, Department of Gastroenterology and Hepatology, Rotterdam, Netherlands

⁴Weill Cornell Medicine, New York Presbytarian Hospital, Department of Gastroenterology and Hepatology, New York, United States

⁵Fondazione Policlinico Universitario A. Gemelli, IRCCS, Catholic University, Digestive Endoscopy Unit, Rome, Italy

⁶Medical University of South Carolina, Division of Gastroenterology and Hepatology, Department of Medicine, Charleston, United States

⁷Cliniques universitaires Saint-Luc, Université Catholique de Louvain, Departmemt of Hepato-Gastroenterology, Brussels, Belgium

⁸Paoli-Calmettes Institute, Endoscopic Unit, Marseille Cedex, France

⁹NYU Langone Medical Center, New York University, Division of Gastroenterology and Hepatology, New York, United States

¹⁰AdventHealth Orlando, Center for Interventional Endoscopy, Orlando, United States

¹¹Korea University Ansan Hospital, Division of Gastroenterology, Department of Internal Medicine, Ansan, Korea (Republic of)

¹²Tokyo Medical University, Department of Gastroenterology and Hepatology, Tokyo, Japan

¹³Keio University School of Medicine, Division of Gastroenterology and Hepatology, Department of Internal Medicine, Tokyo, Japan

¹⁴Helsinki University Central Hospital, Department of Gastrointestinal and General Surgery, Helsinki, Finland

¹⁵Evangelisches Krankenhaus Düsseldorf, Department of Gastroenterology, Düsseldorf, Germany

¹⁶Yonsei University College of Medicine, Department of Internal Medicine, Seoul, Korea (Republic of)

¹⁷Asian Institute of Gastroenterology Hospitals, Hyderabad, India

¹⁸Kobe University Graduate School of Medicine, Division of Gastroenterology, Department of Internal Medicine, Kobe, Japan

¹⁹Westmead Hospital, Department of Gastroenterology and Hepatology, Sydney, Australia

²⁰Westmead Clinical School, University of Sydney, Sydney, Australia

^✉

Contact E-Mail Address: j.a.fritzsche@amsterdamumc.nl

Introduction

Consensus on the optimal treatment algorithm for endo-scopic treatment of papillary adenomas has not been established. This is likely due to low patient numbers and lack of high level scientific studies.

Aims & Methods

The aim of this study is to assess the existing degree of consensus among international experts and develop further concordance by means of a Delphi process. Ultimately, the goal is to generate a consensus text for a more standardized treatment algorithm for the endoscopic treatment of papillary adenomas. A total of 52 international experts with either significant scientific contributions or internationally recognised high volume clinical practice in the field of endoscopic papillectomy were invited to participate.

Data were collected between August and December 2019 using an online survey platform. A total of three rounds were conducted and after each round the responses were summarized and anonymously redistributed for discussion in the next round(s). Consensus was defined as >70% agreement. Based on these results a consensus based treatment algorithm is proposed. Strength of the consensus statements was based on the level of evidence of the supporting literature collected after systematic review, according to the definitions of the Oxford Centre for Evidence-Based Medicine.

Results

Twenty-eight experts (54%) joined the first round, 17 the second (33%) and 16 the final round (31%). Consensus was achieved on 47 of the final 79 statements (59%). A selection of consensus statements, percentage of agreement, and level of evidence is shown in the Table. Diagnostic work-up should include at least an upper endoscopy using a duodeno-scope and biopsies. Additional abdominal imaging should only be performed on indication. Patients with (suspected) papillary malignancy or over 1 cm ingrowth in the pancreatic duct (PD) or common bile duct (CBD) should be referred for surgical resection. To prevent pancreatitis, rectal nonsteroidal anti-inflammatory drugs should be administered prior to resection and a PD stent should be placed afterwards. A CBD stent is indicated in case of ongoing bleeding from the papillary region or concerns for a (micro) perforation after resection. Follow-up protocol depends on pathology showing low or high grade dysplasia, in which case first follow-up should be performed within 6 or 3 months, respectively. in both situations follow-up should be continued for at least 5 years. All other statements did not reach consensus.

Conclusion

Useful expert consensus statements on endoscopic treatment for papillary adenoma were derived from this Delphi process. This is the first step in developing an international consensus based algorithm for endoscopic management of papillary adenomas. There were surprisingly many areas where consensus could not be achieved, no scientific data exists, and a wide variety in daily practice continues. These aspects of endoscopic papillectomy should be the focus of future studies.

Disclosure

Nothing to disclose

[Selection of important consensus statements (A; level 1a-b evidence,B; level 2a-3b evidence,C; level 4 evidence,D; level 5 evidence)]

Consensus statement	Agreement	Grading
1. Gastroduodenoscopy with side-viewing instrument should always be performed prior to resection.	100%	D
2. Biopsy should always be performed prior to resection.	94%	D
3. Either MRI/MRCP or EUS should be performed in case of a lesion larger than 2 cm and/or in case of cholestatic laboratory features and CT in case of jaundice.	75%	D
4. Either MRI/MRCP or CT should be performed in case of significant weight loss and/or in case of endoscopic signs of malignancy.	81%	D
5. When a lesion shows ulceration this lesion should be defined as most likely malignant.	94%	D
6. Patient should be referred for surgical management in case of ingrowth in the CBD or PD of >1 cm, considering patient is suitable for surgery.	81%	D
7. If there is ingrowth in the CBD >1 cm, endoscopic papillectomy with radiofrequency ablation can be considered in a patient that is not a surgical candidate due to age and/or comorbidity. Considering the lesion seems favorable for endoscopic resection.	75%	C
8. Submucosal injection should only be performed in case of lateral spreading lesion.	88%	C
9. PD stent should be routinely placed to prevent postintervention pancreatitis.	82%	B
10. CBD stent should only be placed if there are concerns for a perforation or bleeding in the papillary region after resection.	77%	D
11. Biliary sphincterotomy should be performed in case of concomitant bile duct stones and in case drainage is deemed suboptimal.	81%	D
12. Rectal indomethacin or diclofenac should be administered prior to resection.	82%	B

Open in a new tab

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.35

OP039 Indwelling Double-Pigtail Plastic Stents For The Treatment of Disconnected Pancreatic Duct Syndrome-Associated Peripancreatic Fluid Collections: Long-Term Safety and Efficacy Results

P Gkolfakis ^1,^✉, A Bourguignon ¹, M Arvanitakis ¹, A Baudewyns ¹, P Eisendrath ¹, D Blero ¹, A Lemmers ¹, M Delhaye ¹, J Deviere ¹

Introduction

Long-termtransmural double-pigtail stent (DPS) placement is recommended for patients with disconnected pancreatic duct syndrome (DPDS) and peripancreatic fluid collections (PFC). Long-term safety and efficacy of indwelling DPS were evaluated.

Aims & Methods

Medical files of patients treated with transmural DPS for DPDS-associated PFC and a minimum follow-up of 48 months were reviewed. Overall, early (< 30 days) and late complications were evaluated. Long-term efficacy and PFC recurrence were also assessed.

Results

From 2002 to 2014, 116 patients with DPDS were identified and followed for 80.6±34.4 months. Patients underwent 175 transmural drainage procedures (89.7% transgastric access). Seventy (60.3%) of the initially drained PFCs were pseudocysts with the main treatment indication being infection (47.4%). Twenty early complications were observed, 6 of them (3 bleedings and 3 pneumoperitoneum) occurred peri-interventionally. Late complications (n=17), were mainly pain due to DPS-induced ulcer or erosion (n=10) and 14/17 of these were treated either conservatively or by stent removal. Two cases of gastro-pancreatico-colo-cutaneous fistula and one persisting bleed required surgical intervention. All patients had favorable outcomes and no DPS-related deaths were recorded. Overall, the incidence of late complications was 2.2 per 100 patient-years of follow-up. Stent migration occurred in 85/116 (73.3%) patients. PFCs recurred in 27.6% of the patients; with earlier stent migration and chronic pancreatitis identified as risk factors for PFC recurrence. Long-term endoscopic treatment was considered successful in 109/116 (94%) patients.

Conclusion

These results demonstrate that long-term transmural drainage with DPS is a safe and effective treatment for DPDS-associated PFCs.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.36

OP040 Lams For Pancreatic Fluid Collections: The Impact of The Pigtails

T Seoud ^1,^✉, A Syed ², S Umar ², G Kochhar ³, M Dhawan ², A Lupetin ², S Thakkar ²

Introduction

Lumen-apposing metal stents (LAMS) are recently developed devices that allow for transmural drainage of pancreatic fluid collections (PFCs). However, there are limited studies evaluating the effect of placing double pigtail stents (DPS) across LAMS during PFC drainage. Moreover, the negative effect of PPI therapy on the efficacy of PFC drainage remains controversial.

Aims & Methods

We evaluated outcomes of patients who had LAMS deployed with DPS across them as well as the effect of PPI therapy on the need for reintervention and repeat drainage of PFCs in the setting of LAMS with DPS.

Patients with LAMS deployed from Apr. 2015 till Sept. 2019 at a single tertiary care center were reviewed retrospectively. Patients with DPS deployed across LAMS that were used for drainage of pseudocysts (PP) or walled off pancreatic necrosis (WON) were noted. Technical success (TS), clinical success (CS), adverse events and need for re-intervention was recorded. TS and CS were defined as a significant decrease in size of the PP/ WON by at least 50% on CT/MRI and resolution of symptoms at 4-5 weeks follow up, respectively.

Results

112 LAMS were deployed for PFC drainage. 86 were WON (76.8%) and 25 were pseudocysts (22.3%). of those patients, 99 had LAMS with DPS deployed across them. 89.9% of those patients had TS and 87.9% had CS at follow up visits. of the patients who had WON with LAMS and DPS deployed, 31.4% required re-intervention. 48.1% of those patients were on continuous PPI therapy after LAMS deployment until the re-intervention period compared to 51.8% of them who were not on PPI therapy (P= 0.7876). The average number of re-interventions done on WON patients who were on continuous PPI therapy was 1.85 (SD 0.86 Range: 1-4) compared to 1.64 (SD: 0.89 Range: 1-4) re-interventions done on the patients who were not on PPI therapy (P= 0.5392). 61.5% of the patients who were on continuous PPI therapy required at least one re-intervention as compared to 42.8% of the patients who were not on PPI therapy but required at least one re-intervention (P= 0.3403). Only 5.1% of the patients who had LAMS with DPS deployed across them developed adverse events. These included 2 acute bleeding, 1 pneumoperitoneum, 1 migration, and 1 buried LAMS syndrome. 4/5 of those were delayed complications.

Conclusion

Our data showed that there is no significant difference between WON patients’ need for reintervention on or off PPI therapy when DPS are deployed across LAMS. Perhaps the utilization of DPS reduces the impact of withholding PPI therapy on patients with LAMS, possibly from mechanical debridement of the blunt pigtail coupled with gastric contractions. Moreover, our institution's experience with LAMS with DPS deployed across them for draining PPs and WONs has a high TS rate with a low adverse event rate. Larger prospective investigations into LAMS with DPS are needed to further elucidate their benefit in the setting of PPI therapy.

Disclosure

Nothing to disclose

Table 2.

[LAMS Indications,Use of DPS,PPI Therapy,and Adverse Events]

Indications, n (%)
PFC	112
WON	86 (76.8)
PP	25 (22.3)
LAMS with DPS for PFCS, n	99
Need for re-intervention in WON, n (%)	27 (31.4%)
On PPI therapy:	13 (48.1)
Average number of re-interventions:	1.85
Patients requiring >1 re-intervention:	8 (61.5)
Not on PPI therapy:	14 (51.8)
Average number of re-interventions:	1.64
Patients requiring >1 re-intervention:	6 (42.8)
Complications, n (%)	5 (5.1)

Open in a new tab

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.37

OP041 Ultraproactive Therapeutic Drug Monitoring Based On Point-Of-Care Testing of Infliximab Is Not Superior To Reactive Drug Monitoring in Patients with Inflammatory Bowel Disease: 1 Year Results of A Pragmatic Clinical Trial

P Bossuyt ^1,^✉, S Claeys ², S D'haens ¹, E Hoefkens ¹, B Strubbe ², D Marichal ², L Pouillon ¹, H Peeters ²

Introduction

Higher infliximab (IFX) drug levels are associated with better outcomes over time in patients with inflammatory bowel disease (IBD). Algorithm based personalized dosing of IFX may prevent loss of response.¹ When incorporating point of care testing (POCT) for IFX, ultraproactive therapeutic drug monitoring (TDM) with ad-hoc dose optimisation is possible.

Aims & Methods

We aimed to compare the use of an ultraproactive TDM dosing algorithm based on POCT with the use of reactive TDM, by looking at clinical outcomes at 1 year in IFX-treated patients with IBD. This was a pragmatic cluster randomized clinical trial in two large IBD centers. All IBD patients with maintenance IFX treatment were prospectively included between June and August 2018. in cohort A, an ultra-proactive TDM algorithm was applied as follows. All patients had an ELISA TL measurement at baseline, of which the result determined the follow-up pathway: (A) TL between 3-7|g/mL: continuation at same dose and interval; (B) TL >7|g/mL: interval prolongation allowed; (C) TL < 3|g/mL: interval shortening with minimum 2 weeks, with the next IFX TL measured using a POCT. (i) If the POCT showed an IFX TL < 3μg/mL, dose was optimized ad hoc using a linear dosing formula (Dose¹¹ = (TL^target * Dose) / TL^measured), followed by a new POCT test at next visit with the same interval. (ii) If the POCT showed an IFX TL >3|g/mL, no additional dose was given and routine TL testing with ELISA was retaken at next visit. This algorithm was repeated at each visit. in cohort B, a reactive TDM approach was applied with TL measurement only at physician's discretion (retrospectively assessed to avoid treatement bias). Primary endpoint was failure of IFX therapy, defined as IFX discontinuation, IBD surgery, IBD hospitalization, add-on IBD treatment, and allergic reaction to IFX. Secondary endpoints included sustained clinical remission (phycisian's global assessement < 1 at each visit) and mucosal remission (endoscopy and/or fecal calprotec-tin) between 6 and 12 months after initiation of the trial.

Results

One hundred eightyseven patients were included (cohort A/B: 115/72, M/F: 95/92, CD/UC: 135/51). Both cohorts had comparable baseline characteristics for disease type, disease duration, IFX duration, and previous treatment. Cohort A had more TL measurements compared with cohort B (8.8/patient/year vs 1/patient/year; p< .0001) leading to a significant higher number of dose optimizations: interval shortening 48% vs 15% (p< .0001), interval prolongation 21% vs 6% (p=.004), bidirectional 13% vs 1% (p=.006). in cohort A, POCT was required in 27% after the first round of ultraproactive TDM and in a mean of 6.3% (SD 1.9%) in the subsequent rounds. Ad-hoc extra dosing was required in 13% of the POCT. After one year of follow-up, no difference was seen in IFX failure 19% vs 10% (p=.08) or IFX discontinuation 11% vs 6% (p=.18) between cohort A and B. Sustained clinical remission rates were comparable in both cohorts (75% vs 83%; p=.17). Mucosal remission data were available in 71 patients (38%). in this subgroup, mucosal remission was more frequently seen in the reactive TDM cohort (79%) than in the ultraproactive TDM cohort (52%) (p=.021).

Conclusion

This is the first clinical trial comparing an ultraproactive TDM dosing regimen including POCT with reactive TDM dosing during maintenance therapy with IFX in IBD. Although ultraproactive TDM dosing leads to more dose flexibility compared with reactive TDM, this does not result in better clinical outcomes. The value of ultraproactive TDM during induction therapy requires further investigation.

Disclosure

L.P. received personal fees from Abbvie, Ferring, Takeda; and advisory board fees from Takeda. PB has received financial support for research from AbbVie, Mundipharma, Pfizer, Janssen and Mylan; lecture fees from AbbVie, Takeda, Pfizer and Janssen; advisory board fees from Abbvie, Takeda, Hospira, Janssen, MSD, Mundipharma, Roche, Pfizer, Sandoz and Pentax.

References

1.Strik et al. Journal of Crohn's and Colitis 2019: 13(S063) [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.38

OP042 Pharmacokinetics, Immunogenicity, and Exposure-Response Relationship of Ustekinumab in Patients with Crohn'S Disease: Results From The Week 16 Interim Analysis of The Stardust Study

G D'Haens ^1,^✉, OJ Adedokun ², S Danese ³, S Vermeire ⁴, J Panés ⁵, A Dignass ⁶, FJ Magro Dias ^7,⁸, M Nazar ⁹, M Le Bars ¹⁰, M Lahaye ¹¹, L Ni ¹², DR Gaya ¹³, L Peyrin-Biroulet ¹⁴

Introduction

STARDUST is the first treat-to-target (T2T), randomized phase 3b trial of adult patients (pts) with moderate-to-severe, active Crohn's disease (CD), using endoscopy at Week (W) 16 as a first decision point for dose adjustment of ustekinumab (UST). It compares UST treat to target (T2T) with UST standard of care (SoC) in achieving endoscopic response - defined as a >50% reduction in Simple Endoscopic Score for CD (SES-CD) vs baseline - at W48.

Aims & Methods

This interim analysis describes pharmacokinetics (PK), immunogenicity, and exposure-response data through W16 of STARDUST. During induction, all pts received IV weight-based UST (∼6mg/kg [approved label]) at W0 and SC UST 90mg at W8. Blood samples were collected to measure serum UST concentration (conc) and antibodies to UST at W0, W8, and W16 using validated assays. Clinical efficacy was evaluated based on CD Activity Index (CDAI) and centrally read endoscopy. Faecal calprotectin (FCal) and C-reactive protein (CRP) were measured using validated assays. At W16, pts with CDAI reduction >70 points were randomized (1:1) to T2T or SoC. Patients randomized to T2T group underwent endoscopy at W16 as part of the decision on treatment regimen (every 8W or every 12W).

Results

erum UST conc and immunogenicity data were obtained from a set of 494 pts who received UST in STARDUST and had at least 1 blood sample for PK analysis. Median serum UST conc at W0 (1 hour post-infusion), W8 trough, and W16 trough (112.7ug/mL, 6.7ug/mL, and 2.7μg/ mL, respectively) were comparable to those from the pivotal CD studies (UNITI-1, UNITI-2, & IM-UNITI) at the same timepoints (126.1ug/mL, 6.4ug/ mL, and 2.1ug/mL). Serum UST concs were almost comparable between pts who were biologic naïve or had prior exposure to 1 biologic. At W8 and W16, proportions of pts in clinical response and remission were similar across the top 3 UST conc quartiles (Q2-Q4) but numerically lower in the lowest quartile (Q1) (Table). Baseline CRP and FCal levels were inversely associated with UST concs at W8 and W16 suggesting that pts who have the highest inflammatory burden as measured by CRP/FCal clear the drug more rapidly than those with lower inflammatory burden. At W8 and W16, the proportions of pts with normalization of these biomarkers increased with increasing UST conc. UST concs were positively associated with proportions of pts achieving endoscopic response (SES-CD >50%) and endoscopic remission (SES-CD < 3) (Table 1). Incidence of antibodies to UST through W16 visit was 2% and not adversely associated with efficacy.

[Ustekinumab concentration and efficacy outcomes through Week 16]

Week 8 concentration (μg/mL) quartiles^a,b,c	Q1	Q2	Q3	Q4
Clinical response^h	58.7% (71/121)	76.9% (93/121)	70.2% (85/121)	69.7% (85/122)
Clinical remissionⁱ	50.4% (61/121)	62.8% (76/121)	56.2% (68/121)	59.0% (72/122)
Normalization of C-reactive protein (≤3mg/L)^j	9.5% (8/84)	28.6% (24/84)	42.9% (36/84)	43.5% (37/85)
Normalization of faecal calprotectin (≤250μg/g)^k	10.4% (8/77)	23.1% (18/78)	39.5% (30/76)	50.0% (40/80)
Week 16 UST concentration (μg/mL) quartiles^d,e,f,g	Q1	Q2	Q3	Q4
Clinical response^h	74.1% (83/112)	84.1% (95/113)	88.4% (99/112)	80.5% (91/113)
Clinical remissionⁱ	57.1% (64/112)	73.5% (83/113)	73.2% (82/112)	71.7% (81/113)
Endoscopic response^l	22.0% (11/50)	39.2% (20/51)	38.0% (19/50)	51.0% (26/51)
Endoscopic remission^m	2.0% (1/50)	7.8% (4/51)	20.0% (10/50)	19.6% (10/51)
Normalization of C-reactive protein (≤3mg/L)^j	7.8% (6/77)	22.8% (18/79)	42.3% (33/78)	47.4% (37/78)
Normalization of faecal calprotectin (≤250μg/g)^k	16.2% (12/74)	26.4% (19/72)	48.6% (36/74)	53.4% (39/73)

Open in a new tab

^a

Week 8 ustekinumab concentration quartiles for clinical response or clinical remission at Week 8: Q1: <3.9μg/mL, Q2: 3.9 to <6.7μg/mL, Q3: 6.7 to <9.8μg/mL, Q4: >9.8μg/mL

^b

Week 8 ustekinumab concentration quartiles for normalization of C-reactive protein at Week 8: Q1: <3.1ng/mL, Q2: 3.1 to <5.8ng/mL, Q3: 5.8 to <9.1ng/mL, Q4: >9.1ng/mL

^c

Week 8 ustekinumab concentration quartiles for normalization of faecal calprotectin at Week 8: Q1: <3.3μg/mL, Q2: 3.3 to <6.1μg/mL, Q3: 6.1 to <9.5μg/mL, Q4: >9.5μg/mL

^d

Week 16 ustekinumab concentration quartiles for clinical response or clinical remission at Week 16: Q1: <1.5μg/mL, Q2: 1.5 to <2.7μg/mL, Q3: 2.7 to <4.9μg/mL, Q4: >4.9μg/mL

^e

Week 16 ustekinumab concentration quartiles for endoscopic response or endoscopic remission at Week 16: Q1: <1.3ng/mL, Q2: 1.3 to <2.6ng/mL, Q3: 2.6 to <4.8ng/mL, Q4: >4.8|ag/mL. Week 16 ustekinumab concentration quartiles for normalization of C-reactive protein at Week 16: Q1: <1.3ng/mL, Q2: 1.3 to <2.5ng/mL, Q3: 2.5 to <4.2μg/ mL, Q4: >4.2|g/mL.

^g

Week 16 ustekinumab concentration quartiles for normalization of faecal calprotectin at Week 16: Q1: <1.3|μg/mL, Q2: 1.3 to <2.6ng/mL, Q3: 2.6 to <4.7μg/ mL, Q4: >4.7|ag/mL.

^h

Defined as a >100-point reduction from the baseline CDAI score, or a CDAI score <150.

ⁱ

Defined as a CDAI Total score of <150 points.

^j

Includes only subjects with abnormal CRP at baseline.

^k

Includes only subjects with abnormal faecal calprotectin at baseline.

^l

Defined as a reduction from baseline in SES-CD of ≥50%.

^m

Defined as SES-CD of <3.

Conclusion

Serum UST conc and the low incidence of antibodies to UST were consistent with data from pivotal CD randomized controlled trials. Most pts attained clinical efficacy outcomes at UST concs derived from the approved dose regimen; however, pts with higher UST concs were more likely to achieve biomarker and endoscopic improvements.

Disclosure

G D'Haens: has served as advisor for AbbVie, Ablynx, Allergan, Alphabiomics, Amakem, Amgen, AM Pharma, Applied Molecular Therapeutics, Arena Pharmaceuticals, AstraZeneca, Avaxia, Biogen, Bristol-Myers Squibb, Boehringer Ingelheim, Celgene/Receptos, Celltrion, Cosmo, DSM Pharma, Echo Pharmaceuticals, Eli Lilly, Engene, Exeliom Biosciences, Ferring, Dr Falk Pharma, Galapagos, Genentech/Roche, Gilead, GSK, Gossamerbio, Hospira/Pfizer, Immunic, Johnson and Johnson, Kintai Therapeutics, Lycera, Medimetrics, Millenium/Takeda, Medtronics, Mitsubishi Pharma, MSD, Mundipharma, Nextbiotics, Novonordisk, Otsuka, Pfizer/Hospira, Photopill, ProciseDx, Prodigest, Prometheus Laboratories/ Nestlé, Progenity, Protagonist, RedHill, Robarts Clinical Trials, Salix, Samsung Bioepis, Sandoz, Seres/Nestlé, Setpoint, Shire, Teva, Tigenix, Tillotts, Topivert, Versant and Vifor; and reports speaker fees from AbbVie, Biogen, Ferring, Johnson & Johnson, MSD, Mundipharma, Norgine, Pfizer, Samsung Bioepis, Shire, Millenium/Takeda, Tillotts and Vifor. OJ Adedokun is a full-time employee of Janssen Research & Development, LLC S Danese reports consultancy fees from AbbVie, Allergan, Amgen, AstraZeneca, Biogen, Boehringer Ingelheim, Celgene, Celltrion, Eli Lilly, Enthera, Fer-ring Pharmaceuticals Inc., Gilead, Hospira, Janssen, Johnson & Johnson, Merck Sharp Dome, Mundipharma, Mylan, Pfizer, Roche, Sandoz, Sublimity Therapeutics, Takeda, TiGenix, UCB Inc. and Vifor. S Vermieire reports grants/research support from AbbVie, Janssen, MSD, Pfizer and Takeda; honoraria/consulting fees from AbbVie, Amgen, Arena, Celgene, Ferring, Galapagos, Genentech/Roche, Gilead, GlaxoSmithKline, Hospira, Janssen, Lilly, MSD, Mundipharma, Pfizer Inc., Progenity, Second Genome, Shire and Takeda; and participation in company sponsored speakers bureaux with AbbVie, Ferring, Hospira, Janssen, MSD, Pfizer, Takeda and Tillots. J Panés reports grants from AbbVie, MSD and Pfizer; consulting fees/honorarium from AbbVie, Arena, Boehringer Ingelheim, Celgene, Genentech/ Roche, GlaxoSmithKline, Janssen, MSD, Nestlé, Oppilan, Pfizer, Progen-ity, Takeda, Theravance and TiGenix; support for travel to meetings from AbbVie and Takeda during the conduct of the study; payment for lectures including service on speakers bureaux from Abbott, Janssen, MSD, Pfizer and Takeda; and payment for development of educational presentations from Abbott, Janssen, MSD, Pfizer and Roche, outside the submitted work. A Dignass reports fees for participation in review activities, such as data monitoring boards, statistical analysis, end point committees, and the like from Falk, during the conduct of the study; consultancy fees from AbbVie, Amgene, Celgene Tillotts, Falk, Ferring, Fresenius Kabi, Janssen, MSD, Pfizer, Roche/Genentech, Sandoz/Hexal, Takeda and Vifor; grants from Institut für Gemeinwohl; payment from lectures including service on speakers bureaux from AbbVie, Falk Foundation, Ferring, Janssen-Cilag, Med Update GmbH, MSD, Pfizer, Tillotts and Vifor; payment for manuscript preparation from Falk Foundation, Takeda, Thieme and UniMed Verlag; and payment for development of educational presentations from Fer-ring and Tillots, outside the submitted work. F Magro reports fees from lectures to AbbVie, Falk, Ferring, Hospira, Lab Vitoria, MSD, OM Pharma, PharmaKern, Schering Vifor and Takeda. M Nazar is a full-time employee of Janssen-Cilag Polska Sp. z o.o. and reports restricted stocks. M Le Bars is a full-time employee of Janssen-Cilag and reports restricted stocks. M Lahaye is a full-time employee of Janssen-Cilag BV and reports restricted stocks. L Ni is a full-time employee of Janssen-Cilag Russia and reports restricted stocks. DR Gaya reports personal fees from AbbVie, Janssen and Takeda; and a travel grant from Vifor, outside the submitted work. L Peyrin-Biroulet reports personal fees from AbbVie, Allergan, Alma, Am-gen, Applied Molecular Transport, Arena, Biogen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion, Enterome, Enthera, Ferring, Fre-senius, Genentech, Gilead, Hikma, Index Pharmaceuticals, Janssen, Lilly, MSD, Mylan, Nestlé, Norgine, Oppilan Pharma, OSE Immunotherapeutics, Pfizer, Pharmacosmos, Roche, Samsung Bioepis, Sandoz, Sterna, Sublimity Therapeutics, Takeda, Tillots and Vifor; and grants from AbbVie, MSD and Takeda; and stock options from CTMA, outside the submitted work.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.39

OP043 Clinically Adjusted Versus Therapeutic Drug Monitoring Dosing Regimens with Adalimumab in Patients with Moderately To Severely Active Crohn'S Disease: Results From The Serene-Cd Maintenance Study

S Danese ^1,^✉, WJ Sandborn ², EV Loftus Jr ³, SB Hanauer ⁴, S Schreiber ⁵, L Peyrin-Biroulet ⁶, R Panaccione ⁷, J Panés ⁸, F Baert ⁹, J-F Colombel ¹⁰, M Ferrante ¹¹, E Louis ¹², A Armuzzi ¹³, VS Goteti ¹⁴, J Liu ¹⁴, NM Mostafa ¹⁴, T Doan ¹⁴, J Petersson ¹⁴, AM Robinson ¹⁴, AP Song ¹⁴, GR D'Haens ¹⁵

Introduction

Adalimumab (ADA) is well-tolerated and effective in patients (pts) with Crohn's disease (CD)[1]. SERENE-CD (NCT02065570) is a phase 3, double-blind, randomized, multicenter study to evaluate higher versus standard ADA dosing regimens for Induction therapy in adult pts with moderate to severe CD. Induction results have been previously reported [2]. The maintenance study explored outcomes in patients after 44 weeks (wks) ADA treatment using clinically adjusted (CA) dosing based on clinical symptoms and biomarkers, or therapeutic drug monitoring (TDM) based on symptoms, biomarkers, and ADA serum levels. Herein results from the maintenance study are reported.

Aims & Methods

Safety and efficacy of CA and TDM maintenance regimens at Wk56 (12wks induction + 44 wks dosing) were assessed. Clinical responders (N=184) completing Induction were re-randomized at Wk12 and assigned 1:1 to CA or TDM regimens. The CA arm consisted of 40 mg ADA every other wk (EOW), escalating to every wk (EW) based on CDAI (>220) or high-sensitivity C reactive protein (hs-CRP, >10mg/L). All pts in the TDM arm received ADA 40 mg EOW at Wk12 and could undergo dose-escalation to EW at Wks 14, 28, or 42 if ADA serum concentration was < 5 μg/mL or 5-10 μg/mL with CDAi>220 or hs-CRP>10 mg/L. Once escalated, the pt remained at 40 mg EW. Key exploratory efficacy endpoints at Wk56 included measures of clinical remission and endoscopic response and remission (Table 1). Safety analyses were performed for all subjects who took at least 1 dose of maintenance study drug.

Table 1.

[Key Exploratory Efficacy Endpoints (mITT population^#)]

Endpoints (Week 56)	CA n/N (%)	TDM n/N (%)	Nominal p-value^&
Clinical remission (CDAI^a < 150)	65/92 (70.7%)	61/92 (66.3%)	0.497
Steroid-free clinical remission among subjects taking corticosteroid at Induction Baseline	30/39 (76.9%)	41/56 (73.2%)	0.636
Endoscopic response^b	41/92 (44.6%)	40/92 (43.5%)	0.824
Endoscopic remission^c	29/92 (31.5%)	27/92 (29.3%)	0.621
Deep remission (CDAI^a < 150 and endoscopic remission)	27/92 (29.3%)	24/92 (26.1%)	0.507
Sustained clinical remission (CDAI^a < 150) among Week 12 clinical remitters	55/74 (74.3%)	48/66 (72.7%)	0.819
Endoscopic response^b among Week 12 endoscopic responders	30/42 (71.4%)	25/45 (55.6%)	0.132
Endoscopic remission^c among Week 12 endoscopic remitters	21/30 (70.0%)	17/33 (51.5%)	0.146

Open in a new tab

^#

Includes clinical responders who completed 12-week induction period.

^&

p-value for comparison between CA and TDM, calculated using Cochran-Mantel-Haenszel (CMH) test

^a

CDAI - Crohn's Disease Activity Index

^b

Defined as decrease in SES-CD > 50% from Induction Baseline (or for an Induction Baseline SES-CD of 4, > 2-point reduction from Induction Baseline).

^c

Defined as SES-CD ≤ 4 and at least a 2-point reduction from Induction baseline and no subscore greater than 1 in any individual variable.

Results

Of re-randomized pts, 76 (82.6%) CA and 79 (85.9%) TDM pts completed the study. BL characteristics and demographics were well-balanced between treatment arms. in the CA and TDM arms, 26 (28.2%) and 36 (39.1%) pts escalated to EW dosing, respectively, with most in both groups escalating at Wk14 (13/26 CA, 21/36 TDM). No difference in proportion of pts achieving Wk56 clinical remission, steroid-free clinical remission among subjects taking corticosteroid at Induction BL, endoscopic response, endoscopic remission, or deep remission between CA vs TDM (Table 1) was found. No significant differences were observed between the two groups for sustained clinical remission among Wk12 clinical remitters, or for maintenance of endoscopic response and remission among Wk12 responders and remitters respectively (Table 1). Dose escalation was driven mainly by hs-CRP in the CA arm and by serum drug levels in the TDM arm. The observed safety profile was similar between groups; no new safety signals or unexpected trends were identified.

Conclusion

Results from the SERENE-CD maintenance study indicate both CA and TDM maintenance regimens are similar with respect to the key exploratory efficacy endpoints. Both dosing regimens were generally well-tolerated with a similar safety profile as previously described [3]. Addition of TDM as criteria for dose adjustment showed no clinical benefit over use of clinical symptoms and biomarkers alone.

Disclosure

D'Haens: Consultant and/or speaker: AbbVie, ActoGeniX, AIM, Allergan, Amgen, Arena, Boehringer Ingelheim, Celgene/Receptos, Cell-trion, Cosmo Technologies, Elan, Eli Lilly, enGene, Dr Falk Pharma, Ferring, Galapagos, Genentech, Gilead Sciences, Giuliani SpA, Given Imaging, Glaxo-SmithKline, Gossamer Bio, Janssen Biologics, MSD, Neovacs, Norgine, Novo Nordisk, Otsuka, PDL BioPharma, Prometheus, Progenity, Pfizer, Robarts Clinical Trials, Salix, Seres/Nestle, Schering-Plough, SetPoint, Shire, Takeda, Tillotts, Tramedico, UCB, Versant, Vifor; research grants: AbbVie, Dr Falk Pharma, Given Imaging, Janssen, MSD, PhotoPill; Sandborn: Consultant, advisory board member, speaker, and/or stockholder: AbbVie, Allergan, Amgen,

Arena, Avexegen, BeiGene, Boehringer Ingelheim, Celgene, Celltrion, Conatus, Cosmo, Escalier Biosciences, Ferring, Forbion, Genentech, Gilead Sciences, Gossamer Bio, Incyte, Janssen, Kyowa Kirin Pharmaceutical Research, Los Biopharma, Lilly, Oppilan Pharma, Otsuka, Pfizer, Precision IBD, Progenity, Prometheus, Reistone, Ritter, Robarts Clinical Trials (owned by Health Academic Research Trust, HART), Series Therapeutics, Shire, Sienna Biopharmaceuticals, Sigmoid, Sterna, Sublimity Therapeutics, Takeda, Theravance, Tigenix, Tillotts, UCB, Ventyx Biosciences, Vimalan Biosciences, Vivelix; research grants: AbbVie, Amgen, Atlantic Healthcare Limited, Cel-gene/Receptos Genentech, Gilead, Janssen, Lilly, Takeda. Loftus Jr: Consultant: AbbVie, Allergan, Amgen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion Healthcare, Eli Lilly, Genentech, Gilead, Janssen, Pfizer, Takeda, UCB; research grants: AbbVie, Amgen, Genentech, Gilead, Janssen, Medimmune, Pfizer, Receptos, Robarts Clinical Trials, Seres Therapeutics, Takeda, UCB. Hanauer:: Consultant and/or speaker: AbbVie, Actavis, Allergan, Amgen, Arena, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion, Ferring, Genentech, Gilead, GlaxoSmithKline, Hospira, Janssen, Lilly, Merck, Nestle, Novartis, Pfizer, Prometheus, Receptos, Salix, Samsung Bioepis, Sanofi-Aventis, Seres Health, Shire, Takeda, Therakos; research grants AbbVie, Allergan, Amgen, Celgene, Genentech, GlaxoSmithKline, Janssen, Lilly, Novartis, Pfizer, Prometheus, Receptos, Sanofi-Aventis Schreiber: Consultant: AbbVie, Falk Pharma, Ferring, Genentech, GlaxoSmithKline, MSD, Pfizer, Shire, Takeda. Peyrin-Biroulet: Consultant and/or speaker: AbbVie, Amgen, Biogen, Biogaran, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion, Ferring, Forward Pharma, Genentech, HAC-Pharma, Hospira, Index, Janssen, Lilly, Lycera, Merck, Mitsubishi, Norgine, Pfizer, Pharmacosmos, Pilège, Samsung Bioepis, Soz, Takeda, Therakos, Tillotts, UCB, Vifor. Panaccione: Consultant and/or speaker: AI4GI, AbbVie, Arena, Amgen, Atlantic Healthcare, BioBalance, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Coronado Biosciences, Cosmo Technologies, Eagle, Eisai Medical Research, Elan, Eli Lilly, EnGene, Ferring, Genentech, Gilead, Given Imaging, GlaxoSmithKline, Janssen, Lycera, Meda, Merck & Co., Merck Research, MerckSerono, Novo Nordisk, PDL Biopharma, Pfizer, Prometheus, Protagonist, Receptos, Robarts Clinical Trials, Salix, Soz, Sanofi Genzyme, Shire, Sigmoid, Sublimity, Takeda, Theravance. Panés: Consultant and/or speaker: AbbVie, Arena, Boehringer Ingelheim, Celgene, Ferring, Genentech, GlaxoSmithKline, GoodGut, Janssen, MSD, Nestle, Oppilan, Pfizer, Progen-ity, Robarts, Roche, Shire, Takeda, Theravance, TiGenix, Topivert; research grants: AbbVie MSD Baert: Consultant and/or speaker: AbbVie, Falk, Ferring, Janssen, Mundipharma, MSD, Pfizer, Takeda, Vifor; research grants: from AbbVie, Amgen, Chiesi, Ipsen, MSD Colombel: Consultant, speaker, and/or stockholder: AbbVie, Amgen, Allergan, Arena, Boehringer Ingelheim, Celgene, Celltrion, Eli Lilly, Enterome, Ferring, Geneva, Genfit, Genentech, Intestinal Biotech Development, Ipsen, Imedex, Janssen, Landos, Medim-mune, Merck, Novartis, O Mass, Otsuka, Pfizer, Shire, Takeda, Tigenix, Viela Ferrante: Consultant and/or speaker: AbbVie, Boehringer Ingelheim, Chiesi, Ferring, Janssen, Lamepro, Mitsubishi Tanabe, MSD, Pfizer, Takeda, Tillotts, Tramedico, Zeria; research grant: Amgen, Biogen, Janssen, Pfizer, Takeda Louis: Consultant, advisory board member, and/or speaker: Abbvie, Falk, Ferring, MSD, Takeda, Celgene, Hospira, Janssen, Pfizer; research grants: Abbvie, MSD, Takeda, Janssen Armuzzi: Consultant, advisory board member, and/or speaker: AbbVie, Allergan, Amgen, AstraZeneca, Biogen, Bristol-Myers Squibb, Celgene, Celltrion, Chiesi, Ferring, Hospira, Janssen, Lilly, Medtronic, Mitsubishi Tanabe, MSD, Mundipharma, Mylan, Nikkiso, Otsuka, Pfizer, Roche, Samsung Bioepis, Soz, Sofar, Takeda, Tigenix, Zambon; research grants: MSD, Pfizer, Takeda. Danese: Consultant: AbbVie, Allergan, Amgen, Bristol-Myers Squibb, Celgene, Celltrion Healthcare, Eli Lilly, Janssen, Pfizer, Takeda, UCB; research grants: AbbVie, Amgen, Genentech, Gil-ead, Janssen, Medimmune, Pfizer, Receptos, Robarts Clinical Trials, Seres, Takeda, UCB Goteti, Mostafa, Doan, Petersson, Robinson, Song, Butler: Ab-bVie employees, may own AbbVie stock and/or options.

References

1.Colombel J.F. et al. Long-term safety of adalimumab in clinical trials in adult patients with Crohn's disease or ulcerative colitis. Aliment Pharmacol Ther, 2018. 47(2): p. 219–228. [DOI] [PubMed] [Google Scholar]
2.Panés J. C.J.-F., D'Haens G.R., Schreiber S., Panaccione R., Peyrin-Biroulet L., Loftus E. Jr., Danese S., Louis E., Armuzzi A., Ferrante M., Vogelsang H., Lefebvre J., Doan T., Kwatra N.V., Mostafa N.M., Xie W., Huang B., Petersson J., Kalabic J., Robinson A.M., and Sandborn W.J. High versus standard adalimumab induction dosing regimens in patients with moderately to severely active ulcerative colitis: results from the SERENE-UC induction study. United European Gastroenterology Journal, 2019. 7(8S): p. 1. [Google Scholar]
3.Colombel J.F. et al. OP01 Higher vs. standard adalimumab maintenance regimens in patients with moderately to severely active ulcerative colitis: Results from the SERENE-UC maintenance study. Journal of Crohn's and Colitis, 2020. 14(Supplement 1): p. S001-S001. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.40

OP044 Pharmacokinetics and Immunogenics of Maintenance Therapy with Ustekinumab: 2-Year Results From The Unifi Long-Term Extension Study

OJ Adedokun ^1,^✉, R Panaccione ², T Hisamatsu ³, MT Abreu ⁴, RWL Leong ⁵, D Rowbotham ⁶, C Marano ¹, Y Zhou ¹, H Zhang ¹, S Danese ⁷, W Afif ⁸, L Peyrin-Biroulet ⁹, WJ Sandborn ¹⁰, BE Sands, on behalf of the UNIFI Investigators¹¹

Introduction

UNIFI is a phase 3, double-blind, placebo-controlled study that evaluated ustekinumab (UST) treatment in patients (pts) with moderate-severe UC who previously failed biologics (1 or more TNF-blockers or vedolizumab) or conventional therapy (corticosteroids and/or 6-MP/AZA). Pts who completed safety and efficacy evaluations at Week (Wk) 44 of the maintenance study could enter the long-term extension (LTE) study continuing the same treatment regimen that they received at the end of the maintenance study (either placebo or UST 90mg SC q8w or q12w). Starting at Wk56, pts could dose adjust: PBO to UST q8w, UST q12w to q8w, and UST q8w to q8w (sham adjustment).

Aims & Methods

In this analysis, PK, immunogenicity, and their relationships with efficacy and safety were evaluated among pts who received UST SC in the LTE. Serum UST concentrations through Wk 92 and antibodies to UST (positive or negative) through Wk96 were determined using validated assays. Clinical efficacy was assessed with symptomatic and partial Mayo remission at Wk92; safety events (infections, serious infections, and serious AEs) were evaluated through Wk96.

Results

During the LTE, randomized pts who continued to receive 90mg SC UST and did not have a dose adjustment during the LTE (n=271), had median non-trough UST concentrations ranging between 6.49 to 6.66 μg/ mL, 4wks after dosing in the q8w group, and 2.11 to 2.56 μg/mL, 8wks after dosing in the q12w group. High proportions (>80%) of pts were in symptomatic remission and partial Mayo remission at Wk92 in each serum UST concentration quartile, thus no clear exposure-response was demonstrable with these efficacy endpoints; however, proportions of pts with normalized calprotectin and normalized CRP at Wk92 increased with increasing serum UST concentration quartiles (Table). No trends were found between serum UST concentration and the incidence of infections, serious infection, or serious AEs reported through Wk96 (Table). Among 400 pts who received continuous UST in induction, maintenance, and LTE, antibodies to UST were found in 22/400 (5.5%) and were often transient. of these, most (18/22 pts) had titers at or below 1:800; 4 of the 22 pts were positive for neutralizing antibodies. Among pts randomized to UST, the proportions in symptomatic or partial Mayo remission at Wk92 were similar between pts who were positive (80.0%, 80.0%; respectively) and pts who were negative (81.0%, 82.2%; respectively) for antibodies to UST. No relationship was observed between antibodies to UST and injection site reactions.

Conclusion

Following treatment with UST 90mg SC q12w or q8w during the LTE, sustained serum UST levels were observed through Wk92 and were generally consistent with serum UST levels observed for these treatment

groups during the maintenance study. The incidence of antibodies to UST was low and often transient through Wk96 of the LTE, and anti-drug antibodies did not appear to affect efficacy or injection site reactions.

[Table: Efficacy outcomes by serum ustekinumab concentration quartiles at Week 92 during the LTE]

Week 92 Concentration Quartiles^a,b,c
	q12w	q8w
	Q1	Q2	Q3	Q4	Q1	Q2	Q3	Q4
Efficacy Outcome at Week 92
Patients in symptomatic remission, % (n/N)	100% (14/14)	85.7% (12/14)	100.0% (14/14)	85.7% (12/14)	92.3% (12/13)	92.3% (12/13)	92.3% (12/13)	100.0% (14/14)
Patients in partial Mayo remission, % (n/N)	100% (14/14)	85.7% (12/14)	100.0% (14/14)	85.7% (12/14)	92.3% (12/13)	92.3% (12/13)	92.3% (12/13)	100.0% (14/14)
Patients with normalized CRP (≤3 mg/L), % (n/N)	28.6% (2/7)	57.1% (4/7)	57.1% (4/7)	100.0% (7/7)	14.3% (1/7)	62.5% (5/8)	57.1% (4/7)	87.5% (7/8)
Normalized Calprotectin (≤250 mg/kg), % (n/N)	50% (5/10)	45.5% (5/11)	60% (6/10)	90.9% (10/11)	50% (5/10)	45.5% (5/11)	72.7% (8/11)	72.7% (8/11)
Safety Events from Week 44 through Week 96
Patients with ≥1 infection, % (n/N)	78.6% (11/14)	57.1% (8/14)	50.0% (7/14)	57.1% (8/14)	53.8% (7/13)	69.2% (9/13)	61.5% (8/13)	57.1% (8/14)
Patients with ≥1 serious infection, % (n/N)	7.1% (1/14)	0.0% (0/14)	0.0% (0/14)	7.1% (1/14)	0.0% (0/13)	7.7% (1/13)	0.0% (0/13)	0.0% (0/14)
Patients with ≥ 1 serious adverse event, % (n/N)	14.3% (2/14)	7.1% (1/14)	0.0% (0/14)	7.1% (1/14)	7.7% (1/13)	7.7% (1/13)	7.7% (1/13)	0.0% (0/14)

Open in a new tab

^a

Week 92 ustekinumab concentration quartiles for symptomatic or partial Mayo remission: 90 mg SC q12w: Q1: <1.49 μg/mL, Q2: 1.49 to <2.11 μg/mL, Q3: 2.11 to <2.74 μg/mL, Q4: ≥2.74 μg/ mL; 90 mg SC q8w: Q1: <5.22 μg/mL, Q2: 5.22 to <6.66 μg/mL, Q3: 6.66 to <8.53 μg/mL, Q4: ≥8.53 μg/mL

^b

Week 92 ustekinumab concentration quartiles based on patients with abnormal CRP levels (>3 mg/L): 90 mg SC q12w: Q1: <1.39 μg/mL, Q2: 1.39 to <2.03 μg/mL, Q3: 2.03 to <2.72 μg/mL, Q4: ≥2.72 μg/mL; 90 mg SC q8w: Q1: <4.01 μg/mL, Q2: 4.01 to <6.88 μg/mL, Q3: 6.88 to <8.52 μg/mL, Q4: ≥8.52 μg/mL

^c

Week 92 ustekinumab concentration quartiles based on patients with abnormal calprotectin levels (>250 mg/kg): 90 mg SC q12w: Q1: <1.48 μg/mL, Q2: 1.48 to <2.02 μg/mL, Q3: 2.02 to <2.71 μg/mL, Q4: ≥2.71 μg/mL; 90 mg SC q8w: Q1: <5.19 μg/mL, Q2: 5.19 to <6.64 μg/mL, Q3: 6.64 to <8.53 μg/mL, Q4: ≥8.53 μg/mL

Disclosure

This study was supported by Janssen Research & Development, LLC

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.41

OP045 Definition of A Bacterial Signature As A Predictor of Anti Tnf Treatment Response

L Oliver ^1,^✉, D Busquets Casals ², J Amoedo ¹, S Ramió-Pujol ¹, M Serrano ¹, A Bahí ³, A Lluansí ³, R Chavero ⁴, P Gilabert ⁵, JO Miquel-Cusachs ², M Sàbat ⁶, J Guardiola Capon ⁵, M Serra-Pagès ¹, LJ Garcia-Gil ^1,⁷, X Aldeguer Mante ^1,^2,³

Introduction

Crohn's disease (CD) and ulcerative colitis (UC) evolve with alternate outbreaks and remissions of variable duration in both cases. Tumor necrosis factor a antagonists (anti-TNF) have advanced in the treatment of patients with inflammatory bowel disease (IBD), which has improved the patient's quality of life by reducing the number of surgeries and hospitalizations. Despite these advances, about 10-30% of patients do not respond to the treatment after the induction period. Besides, between 20% to 50% further patients need an optimization of the dose to respond the treatment.

Recent studies have pointed gut microbiota can play a role in the anti-TNF treatment response. Gram-positive bacteria can modulate the response of NOD proteins and, on the other hand, gram-negative bacteria can stimulate TLR4 receptors causing activation of NFkß.

The definition of a bacterial signatures for treatment response prognosis would be of great value to configure a personalized therapeutic strategy for these patients and increase the efficacy of this treatment.

Aims & Methods

This study aimed to define a bacterial signature that could be used to predict the response of patients with CD and UC to anti-TNF treatment.

This observational study consisted of obtaining a stool sample from 38 IBD patients before starting 3 different anti-TNF treatments. Patients were recruited in two spanish hospitals: Hospital Universitari Dr. Josep Trueta (Girona) and Hospital Universitari de Bellvitge (Hospitalet de Llobregat). At baseline, it was recorded demographic data, age at diagnosis, disease duration, Montreal classification of disease location and behaviour, and previous related surgeries. At every visit, disease activity score, evolution of faecal calprotectin and anti-TNF antibody levels were analyzed in all patients to evaluate treatment response, differentiating 2 groups: respond-ers and non-responders to biological treatment.

From each sample, DNA was purified and used in a qPCR for the quantification of the following markers: F. prausnitzii (Fpra) and its phylogroups (PHG-I and PHG-II), E.coli (Eco), A. muciniphila (Akk), Ruminococcus sp. (Rum), Bacteroidetes (Bac) and M.smithii (Msm).

Results

In this proof of concept, the predictive ability to identify anti-TNF treatment responders was analysed. Some differences in the abundances of the analysed biomarkers were observed between groups, but none of them demonstrated the ability to differentiate the two groups (responders vs. non-responders) as a single biomarker with high sensitivity and specificity. However, an algorithm consisting in the combination of 4 bacterial markers (Msm, Fpra, PHGII, and Rum) showed a high capacity to discriminate between responders and non responders. The algorithm proved high sensitivity and specificity reporting values of 93.33% and 100% respectively, with a positive predictive value of 100% and a negative predictive value of 75% for predicting response to biologic treatment.

Conclusion

A specific bacterial signature could beneficiate patients with inflammatory bowel disease predicting the therapeutic effectiveness of a anti-TNF treatment, leading to a personalized therapy, improving the patients’ quality of life, saving costs and gaining time in patient improvement.

A larger prospective study will be needed to validate these results.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.42

OP046 European Registry On H. Pylori Management (Hp-Eureg): Empirical First-Line Treatment Use and Effectiveness Trends in Europe in The Period 2013-2020

OP Nyssen ^1,^✉, DS Bordin ², B Tepeš ³, A Perez Arisa ⁴, D Vaira ⁵, M Caldas ¹, L Bujanda ⁶, M Castro-Fernandez ⁷, F Lerang ⁸, M Leja ⁹, L Rodrigo ¹⁰, T Rokkas ¹¹, L Kupcinskas ¹², J Perez-Lasala ¹³, L Jonaitis ¹², O Shvets ¹⁴, A Gasbarrini ¹⁵, H Simsek ¹⁶, AT Roger Axon ¹⁷, M G Buzás ¹⁸, JCL Machado ¹⁹, Y Niv ²⁰, L Boyanova ²¹, A Goldis ²², V Lamy ²³, A Tonkic ²⁴, K Przytulski ²⁵, M Wojciech ²⁶, C Beglinger ²⁷, M Venerito ²⁸, P Bytzer ²⁹, L Capelle ³⁰, V Milivojevic ³¹, L Veijola ³², J Molina Infante ³³, L Vologzhanina ³⁴, G Fadeenko ³⁵, I Ariño Pérez ³⁶, G Fiorini ⁵, A Garre ¹, J Garrido ³⁷, C Fernandez Perez ³⁸, F Heluwaert ³⁹, I Puig ⁴⁰, F Megraud ⁴¹, C O'Morain ⁴², JP Gisbert, On behalf of the Hp-EuReg Investigators.¹

¹Gastroenterology Unit, Hospital Universitario de La Princesa, Instituto de Investigación Sanitaria Princesa (IIS-IP), Universidad Autónoma de Madrid, and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Madrid, Spain

²Gastroenterolgy Unit, A. S. Loginov Moscow Clinical Scientific Center, Department of Outpatient Therapy and Family Medicine, Tver State Medical University and Department of Propaedeutic of Internal Diseases and Gastroenterology, Moscow, Russian Federation

³Gastroenterology Unit, AM DC Rogask, Rogaska Slatina, Slovenia

⁴Digestive Unit, Agencia Sanitaria Costa del Sol, Marbella, Spain

⁵Department of Clinical Medicine, University of Bologna, Bologna, Italy

⁶Department of Gastroenterology. Hospital Donostia/Instituto Biodonostia. Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd). Universidad del País Vasco (UPV/EHU), San Sebastian, Spain

⁷Digestive Unit, Hospital de Valme, Sevilla, Spain

⁸Medical Department, Central Hospital Ostfold, Fredrikstad, Norway

⁹Institute of Clinical and Preventive Medicine & Faculty of Medicine, University of Latvia, Riga, Latvia

¹⁰Gastroenterology Unit, Hospital Universitario Central de Asturias, Oviedo, Spain

¹¹Gastroenterology Unit, Henry Dunant Hospital, Athens, Greece

¹²Lithuanian University of Health Sciences Inst. for Digestive Research, Kaunas, Lithuania

¹³Digestive Service, HM Sanchinarro, Madrid, Spain

¹⁴Internal Diseases Department No. 1, National Medical University named after O.O. Bogomolets, Kyiv, Ukraine

¹⁵Catholic University, Gemelli University Hospital, Internal Medicine, Gastroenterology and Liver Diseases, Rome, Italy

¹⁶Internal Medicine/Gastroenterology Department, Hacettepe University Faculty of Medicine, Ankara, Turkey

¹⁷Gastroenterology Unit, University of Leeds, Leeds, United Kingdom

¹⁸Gastroenterology Unit, Ferencváros Policlinic, Budapest, Hungary

¹⁹Instituto de Investigaçao e Inovaçao em Saúde, Universidade do Porto, and Ipatimup - Institute of Molecular Pathology and Immunology of the University of Porto, Porto, Portugal

²⁰Department of Gastroenterology, Rabin Medical Center, Tel Aviv University, Petach Tikva, Tel Aviv, Israel

²¹Department of Medical Microbiology, Medical University of Sofia, Sofia, Bulgaria

²²Gastroenterology Unit, Timisoara Hospital, Timisoara, Romania

²³Dpt. of Gastroenterology, Hepatology & Nutrition, CHU Charleroi, Charleroi, Belgium

²⁴Department of Gastroenterology, University Hospital of Split, School of Medicine, University of Split, Split, Croatia

²⁵Gastroenterology Unit, Medical Centre for Postgraduate Education, Warszawa, Poland

²⁶Department of Gastroenterology, Pomeranian Medical University, Szczecin, Poland

²⁷Gastroenterology Unit, Hospital de Basel, Basel, Switzerland

²⁸Department of Gastroenterology, Hepatology and Infectious Diseases, Otto-von-Guericke University Hospital, Magdeburg, Germany

²⁹Department of Medicine, Zealand University Hospital, Copenhagen University, Copenague, Denmark

³⁰Gastroenterology and Hepatology, Erasmus MC University, Rotterdam, Netherlands

³¹Medical Department, Clinical Center of Serbia Clinic for Gastroenterology and Hepatology, University of Belgrade, Belgrade, Serbia

³²Internal Medicine, Herttoniemi Hospital, Helsinki, Finland

³³Gastroenterology, Hospital San Pedro de Alcantara, Caceres and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Madrid, Spain

³⁴Gastroenterology Unit Gastrocentre, Moscow, Russian Federation

³⁵Digestive Ukrainian Academy of Medical Sciences, Kyiv, Ukraine

³⁶Gastroenterology Unit, Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain

³⁷Departamento de Psicología Aplicada, Universidad Autónoma de Madrid, Madrid, Spain

³⁸Servicio de Medicina Preventiva, Hospital Clínico San Carlos, Facultad de Enfermería, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain

³⁹Centre Hospitalier Annecy Genvois, Pringy, France

⁴⁰Althaia Xarxa Assistencial Universitaria de Manresa and Universitat de Vic-Universitat Central de Catalunya (UVicUCC), Manresa, Spain

⁴¹Laboratoire de Bactériologie, Hôpital Pellegrin, Bordeaux, France

⁴²Department of Clinical Medicine, Trinity College Dublin, Dublin, Ireland

^✉

Contact E-Mail Address: opnyssen@gmail.com

Introduction

The impact of consensus, prescription choices and efficacy trends on clinical practice over time has not been studied in depth.

Aims & Methods

International multicenter prospective non-intervention-al registry aimed to evaluate the decisions and outcomes of H. pylori management by European gastroenterologists. All infected adult patients were systematically registered at AEG-REDCap e-CRF from June 2013 to April 2020. Variables included: Patient's demographics, previous eradication attempts, prescribed treatment, adverse events, and outcomes. Modified intention-to-treat (mITT) and time trend analyses were performed. Data were subject to quality review.

Results

So far 28,194 patients from 29 European countries have been included, and 24,882 (88%) were first-line empirical prescriptions. Although, overall, the most common prescribed treatments in the 2013-20 period were triple therapies; however, a shift in antibiotic regimens was identified. Triple therapies decreased from over 50% of prescription in 2013/15 to less than 20% in 2018/20; concomitant therapy likewise decreased from 21% in 2013/14 to 13% in 2019/20, while Pylera” increased from 0-1% in 2014/2015 to 18% in 2018/20. An increase in the average duration of treatments from 10.9 days in 2013 to 12.0 in 2020, and of the daily dose of PPI was identified (full description of most common treatments is shown in Table 1).

Table 1.

[Prescriptions and effectiveness trends of first-line empirical treatments in Europe in 2013-2020]

Year	2013	2014	2015	2016	2017	2018	2019	2020
Quadruple-C+A+B	0.5%	0.9%	5.2%	17.2%	10.2%	15.3%	5.2%	5.7%
Pylera®	0.0%	0.0%	0.5%	12%	24.5%	22.3%	17.6%	17.8%
Quadruple-C+A+M/T	20.0%	21.4%	26.9%	22.3%	21.2%	10.6%	11.8%	14.4%
Sequential-C+A+M/T	8.1%	3.4%	1.8%	0.9%	0.3%	0.5%	0.2%	0.3%
Triple-C+M	3.9%	6.4%	9.0%	6.6%	1.4%	1.0%	1.0%	4.0%
Triple-C+A	53.6%	54.3%	42.7%	28.2%	30.5%	34.0%	40.6%	34.3%
7 days/10 days/14 days length	31.3% / 48.3% / 20.5%	28.1% / 52.7% / 19.2%	24.7% / 55.9% / 19.4%	16.7% / 46.4% / 36.8%	7.8% / 46.9% / 45.3%	1.8% / 43.9% / 54.2%	2.3% / 30.8% / 66.8%	10.1% / 31.7% / 58.3%
low / standard / high dose* PPI	62.0% / 18.7% / 19.3%	56.7% / 25.5% / 17.8%	47.1% / 26.5% / 26.4% /	36.2% / 24.5% / 39.4%	39.2% / 23.7% / 37.1%	28.5% / 28.8% / 42.8%	24.0% / 34.6% / 41.4%	33.4% / 24.4% / 42.2%
Eradication rate (mITT)	85.3%	85.1%	85.9%	87.4%	88.2%	90.9%	92.2%	91.3%

Open in a new tab

Regarding the effectiveness of each specific treatment, no trend was identified (data now shown); however, there was a 5% overall improvement in first-line mITT overall effectiveness (Table 1).

Conclusion

European gastroenterological practice is constantly adapting to the newest published evidence and recommendations (reducing the use of triple therapies and increasing the duration of treatment and the dose of PPIs), with a subsequent improvement in overall effectiveness.

PPI: proton pomp inhibitor; mITT: modified intention-to-treat; A - amoxicillin, C - clarithromycin; M - metronidazole; T - tinidazole; L - levofloxacin B; - bismuth salts; Tc - tetracycline. *Low PPI dose - 4.5 to 27 mg omeprazole equivalents, b.i.d., Standard dose PPI - 32 to 40 mg omeprazole equivalents, b.i.d, High dose PPI - 54 to 128 mg omeprazole equivalents, b.i.d.

Disclosure

Dr. Nyssen has received research funding from Mayoly, Allergan. Dr. Gisbert has served as a speaker, a consultant and advisory member for or has received research funding from Mayoly, Allergan, Diasorin.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.43

OP047 European Registry On H. Pylori Management (Hp-Eureg): Analysis of 1,782 Empirical Rescue Therapies On Third and Subsequent Lines

D Burgos-Santamaría ^1,^✉, OP Nyssen ², D Vaira ³, Y Niv ⁴, B Tepeš ⁵, G Fiorini ³, A Perez-Aisa ⁶, LR Rodrigo Sáez ⁷, M Castro Fernandez ⁸, R Pellicano ⁹, I Modolell ¹⁰, P Mata-Romero ¹¹, J-C Delchier ¹², J Ortuno Cortés ¹³, M Areia ¹⁴, J Barrio Andrés ¹⁵, P Phull ¹⁶, L Bujanda Fernández de Piérola ¹⁷, N Brglez Jurecic ¹⁸, J Perez-Lasala ¹⁹, AJ Lucendo ²⁰, L Jonaitis ²¹, J Gomez-Camarero ²², Calvo X Calvet ²³, Fernández J Santos ²⁴, I Puig ²⁵, F Megraud ²⁶, C O'Morain ²⁷, JP Gisbert ²; Hp-EuReg Investigators

¹Hospital Ramón y Cajal, Department of Gastroenterology and Hepatology, Madrid, Spain

²Gastroenterology Unit, Hospital Universitario de La Princesa, Instituto de Investigación Sanitaria Princesa (IIS-IP), Universidad Autónoma de Madrid, and Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Madrid, Spain

³Department of Surgical and Medical Sciences, University of Bologna, Bologna, Italy

⁴Rabin Medical Center, Tel Aviv University, Dept. of Gastroenterology, Petach Tikva, Israel

⁵Abakus Medico d.o.o., Gastroenterology, Rogaska Slatina, Slovenia

⁶Servicio de Gastroenterologia, Agencia Sanitaria Costa del Sol, Red de Investigación en Servicios de Salud en Enfermedades Crónicas (REDISSEC), Marbella, Spain

⁷Servicio de Gastroenterologia, Hospital Universitario Central de Asturias, Oviedo, Spain

⁸Servicio de Gastroenterologia, Hospital de Valme, Digestivo, Sevilla, Spain

⁹Molinette Hospital, Turin, Italy

¹⁰Addenbrookes Hospital, Gastroenterology, Cambridge, United Kingdom

¹¹Hospital Universitario San Pedro de Alcántara, Department of Gastroenterology, Caceres, Spain

¹²Hôpital Henri-Mondor, Service de Gastroenterologie, Créteil, France

¹³Servicio de Gastroenterologia, Hospital Universitari i Politècnic La Fe Valencia, Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Valencia, Spain

¹⁴Portuguese Oncology Institute of Coimbra, Gastroenterology, Coimbra, Portugal

¹⁵Hospital Universitario Rio Ortega, Valladolid, Spain

¹⁶Aberdeen Royal Infirmary, Aberdeen, United Kingdom

¹⁷Instituto Biodonostia, San Sebastian, Spain

¹⁸Diagnostic Center Bled, Bled, Slovenia

¹⁹Servicio de Gastroenterologia, Hospital HM Sanchinarro, Madrid, Spain

²⁰Hospital General de Tomelloso, Dept. of Gastroenterology, Tomelloso, Spain

²¹Lithuanian University of Health Sciences, Gastroenterology, Kaunas, Lithuania

²²Servicio de Gastroenterologia, Hospital Universitario de Burgos, Burgos, Spain

²³Hospital de Sabadell, Institut Universitari Parc Tauli, UAB, Unitat de Malalties Digestives, Barcelona, Spain

²⁴Rio Hortega, Gastroenterology Department, Valladolid, Spain

²⁵Althaia Xarxa Assistencial Universitaria de Manresa and Universitat de Vic-Universitat Central de Catalunya (UVicUCC), Digestive Diseases Department, Manresa, Spain

²⁶HopitalPellegrin, Laboratoire de Bacteriologie, Inserm U1053, Bordeaux, France

²⁷Trinity College Dublin, Faculty of Health Sciences, Dublin, Ireland

^✉

Contact E-Mail Address: diegoburgossantamaria@gmail.com

Introduction

H. pylori treatment's effectiveness decreases as treatment eradication attempts fail.

Aims & Methods

The aim was to evaluate the use and effectiveness of empirical rescue therapies on third and subsequent lines in Europe. This was a sub-study of the the European Registry on H. pylori Management (Hp-EuReg), an international multicenter prospective non-interventional registry starting in 2013 aimed to evaluate the decisions and outcomes of H. pylori management by European gastroenterologists (27 countries, 300 researchers). All infected adult patients were systematically registered at AEG-REDCap e-CRF. All cases with three or more eradication attempts were extracted until June 2019. Only the empirically prescribed therapies (without a result of antibiotic resistance testing) were analyzed. Data were subject to quality review.

Results

In total, 1,782 rescue treatments were included: 1,264, 359, 125 and 34 third-, fourth-, fifth- and sixth-line treatments, respectively. Mean age was 51 years, 69% of patients were women and 5% were allergic to penicillin. Sixty-three different therapy regimens were used, being Pylera” the most commonly prescribed. The most frequent regimens are shown in the table. Overall effectiveness was 73% by modified intention-to-treat (mITT) and 74% by per-protocol (PP) analyses. Three regimens approached an optimal eradication rate (>90%, by mITT): Pylera” prescribed in third line (86%), quadruple PPI-bismuth-tetracycline-metronidazole (95%) and triple PPI-amoxicillin-levofloxacin (90%), these two latter irrespective of line, but only when high PPI doses and 14 days’ treatment duration were used. The use of doxycycline instead of tetracycline was associated with lower eradication rates in classical bismuth quadruple therapies (p < 0.05). Quadruple PPI-amoxicillin-levofloxacin-bismuth therapy was not superior to triple PPI-amoxicillin-levofloxacin as third or subsequent rescue therapy.

Conclusion

Empirical rescue treatments in third and subsequent lines obtain, in general, suboptimal eradication rates in Europe. Only Pylera^® and the optimised versions of triple PPI-amoxicillin-levofloxacin and quadruple PPI-bismuth-tetracycline-metronidazole achieve acceptable outcomes.

[Overall eradication rates of the most prescribed empirical therapies on third and subsequent lines]

Rescue therapy	Use, N (%)	mITT n	mITT effectiveness (95% CI)	PP n	PP effectiveness (95% CI)
Pylera®	416 (23%)	363	84 (80-87)	350	85 (81-88)
Triple PPI-A-L	277 (15%)	213	78 (72-83)	206	78 (72-84)
Triple PPI-A-R	231 (13%)	205	66 (59-72)	198	67 (60-74)
Quadruple PPI-B-Tc-M	171 (9.6%)	162	73 (65-80)	157	73 (66-80)
Quadruple PPI-B-D-M	115 (6.5%)	109	63 (54-72)	105	64 (54-73)
Quadruple PPI-A-L-B	95 (5.3%)	81	78 (67-86)	79	80(69-88)
Quadruple PPI-C-A-M	62 (3.5%)	57	58 (44-71)	54	59 (45-72)
Triple PPI-A-M	54 (3.0%)	47	68 (53-81)	45	69 (53-82)
Triple PPI-C-A	43 (2.4%)	33	67 (48-82)	30	67 (47-83)

Open in a new tab

A, amoxicillin; B, bismuth; C, clarithromycin; D, doxycyclin; L, levofloxacin; M, metronidazole; Mx, moxifloxacin; Tc, tetracycline; R, rifabutin; 95% CI, 95% confidence interval.

Disclosure

Dr. Burgos-Santamaría has no Conflict of Interests. Dr. Gisbert has served as a speaker, a consultant and advisory member for or has received research funding from Mayoly, Allergan, Diasorin.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.44

OP048 Excellent Performance of 2-Weeks Tailored Triple Therapy in H. Pylori Infected Children: Interim Results of The New Europed Hp Registry

TG Le Thi ^1,^✉, K Werkstetter ¹, J Cabral ², K Kotilea ³, P Bontems ³, J Barrio ⁴, ML Cilleruelo Pascual ⁵, M Homan ⁶, M Kori ⁷, P Urruzuno ⁸, N Kalach ⁹, Z Misak ¹⁰, R Lima ¹¹, M Tavares ^11,¹², E Miele ¹³, V Urbonas ¹⁴, A Chiaro ^15,¹⁶, A Papadopoulou ¹⁷, J Sykora ¹⁸, A Krahl ¹⁹, K Matusiewics ²⁰, MK Ugras ²¹, F Rea ²², E Roma ¹⁷, T Casswall ²³, M Klemenak ²⁴, Á Cseh ²⁵, J de Laffolie ²⁶, M Rogalidou ²⁷, AI Lopes ²⁸, H Banoub ²⁹, S Koletzko, on behalf of the Helicobacter pylori Working Group of ESPGHAN¹

¹Dr. von Hauner Children's Hospital, LMU University Hospital of Munich, Munich, Germany

²Dona Estefanía Hospital, University Hospital Centre of Central Lisbon, Lisbon, Portugal

³Université Libre de Bruxelles, Hôpital Universitaire des Enfants Reine Fabiola, Brussels, Belgium

⁴Hospital Universitario de Fuenlabrada Madrid, Madrid, Spain

⁵University Hospital Puerta de Hierro Majadahonda, Madrid, Spain

⁶University Children's Hospital of Ljubljana, Ljubljana, Slovenia

⁷Kaplan Medical Center, Rehovot, Israel

⁸Hospital 12 de Octobre, Madrid, Spain

⁹Saint Antoine Pediatric Clinic, Saint Vincent de Paul Hospital, Catholic University, Lille, France

¹⁰Children's Hospital Zagreb, University of Zagreb, Zagreb, Croatia

¹¹Centro Materno Infantil do Norte, Porto Hospital Center, Porto, Portugal

¹²Hospital de Sao Joao, Pediatria, Porto, Portugal

¹³University of Naples ‘Federico II, Naples, Italy

¹⁴Clinic of Children's Diseases of Vilnius University, Vilnius, Lithuania

¹⁵Giannina Gaslini Children's Hospital and Research Institute, Genoa, Italy

¹⁶Department of Pediatrics, ‘Maggiore'Hospital, Crema, Italy

¹⁷University of Athens, Pediatrics, Athens, Greece

¹⁸Prof Charles University of Prague, Pilsen, Czech Republic

¹⁹Darmstädter Kinderkliniken Prinzessin Margaret, Darmstadt, Germany

²⁰Medical University of Wroclaw, Wroclaw, Poland

²¹Yeditepe University Medical Faculty, Istanbul, Turkey

²²Digestive Endoscopy and Surgery Unit, Bambino Gesù Children's Hospital, Rome, Italy

²³Karolinska University Hospital and Clintec, Karolinska Institutet, Stockholm, Sweden

²⁴University Medical Centre Maribor, Maribor, Slovenia

²⁵Semmelweis University of Medicine, First Department of Pediatrics, Budapest, Hungary

²⁶University Children's Hospital of Giessen, Giessen, Germany

²⁷University Hospital of Ioannina, Pediatrics Department & Ped. Gastroenterology, Ioannina, Greece

²⁸HospitalSanta Maria, Medical Faculty, University of Lisbon, Lisbon, Portugal

²⁹Queen Mary's Hospital for Children, Carshalton, United Kingdom

^✉

Contact E-Mail Address: tlethi@med.lmu.de

Introduction

The updated ESPGHAN/NASPGHAN Guidelines for the management of Helicobacter pylori (H. pylori) in pediatric patients (JPGN 2017;64: 991-1003) recommend a 2-weeks tailored triple therapy (TTT) with higher dosing combining a proton-pump-inhibitor (PPI) with two antibiotics tailored to susceptibility results to reach a primary eradication rate (ER) of ≥90%.

Aims & Methods

The international multicenter prospective EuroPedHp registry was initiated by the H. pylori working group of ESPGHAN for surveillance of antibiotic resistance, to assess the management according to the current guidelines, compliance and success of therapy. Since 2017, 30 centers from 17 European countries reported demographic, clinical and follow up data of H. pylori infected pediatric patients, who were diagnosed with gastric biopsies or non-invasive tests. For this analysis, we included all treatment naïve children aged 3-18.5 years with biopsy proven infection and available antibiotic susceptibility testing for clarithromycin (CLA) and metronidazole (MET), who received TTT with PPI + Amoxicillin + CLA (PAC) or PPI + Amoxicillin + MET (PAM). ER was asccessed 4 to 8 weeks after completed treatment.

Results

Of 756 cases with completed follow up, 419 met inclusion criteria (52% female, median age: 13 years, 37% immigrants). Strains susceptible to CLA and MET were detected in 60% (n=252), double-resistance in 2% (n=10). Primary resistance to CLA and MET was high (24%, 18%, respectively). PAC was given to 59%, and PAM to 41% of cases; 96% were treated for 14 days. Antibiotics were dosed according to guidelines in >80%, while PPI dose was lower than recommended in 52%. The ER in relation to country of living, antibiotic susceptibility, treatment regimen and adherence to therapy is shown in table 1.

Table 1.

[Risk factors for eradication failure in 419 treatment naïve patients with TTT]

Factors	Eradication success	Eradication failed	ORadj (95% CI)	p-value
Number of patients, n (ER%)	376 (90%)	43 (10%)
Country of living
North and West Europe	100 (93%)	7 (7%)	1 (ref)
South Europe	173 (91%)	18 (9%)	1.4 (0.6 - 3.6)	0.4
East Europe	84 (85%)	15 (15%)	2.5 (1.0 - 6.5)	0.05
Israel & Turkey	19 (86%)	3 (14%)	2.2 (0.5 - 9.5)	0.3
Susceptibility groups
MET-S/CLA-S	233 (93%)	19 (7%)	1 (ref)
MET-S/CLA-R	80 (87%)	12 (13%)	1.8 (0.8 - 3.9)	0.1
MET-R/CLA-S	53 (82%)	12 (18%)	2.8 (1.3 - 6.2)	0.01
MET-R/CLA-R	10 (100%)	0 (0%)	na	na
Treatment regimen
PPI + AMO + MET (PAM)	159 (92%)	14 (8%)	1 (ref)
PPI + AMO + CLA (PAC)	217 (88%)	29 (12%)	1.5 (0.8 - 3.0)	0.2
Drug dosing according to guidelines
yes	165 (91%)	17 (9%)	1 (ref)
no	202 (89%)	25 (11%)	1.3 (0.7 - 2.4)	0.5
Compliance to therapy
excellent 90-100%	346 (93%)	27 (7%)	1 (ref)
lower than 90%	15 (63%)	9 (37%)	7.7 (3.0 - 19.2)	<.0001
ORadj obtained from multivariate logistic analysis adjusted for gender and age. Significant results are given in bold characters.

Open in a new tab

In the 252 fully susceptible patients, PAC (ER=92%) tended to have higher risk for treatment failure compared to PAM (ER=95%) (OR adj=1.6, 95%CI: 0.5-5.0, p=0.4).

Conclusion

ER of TTT for 2 weeks tailored to antibiotic susceptibility reached 90% (95%CI: 87% - 93%). The trend for a lower ER with PAC compared to PAM supports our hypothesis that we miss CLA resistant strains by sampling error in cases with mixed infections. Taking at least two gastric biopsies for culture, increasing drug dose, particular PPI as recommended, and supporting measures for compliance may further improve treatment success. Continuous surveillance is necessary to provide more accurate and robust results.

Disclosure

SK reports grants from Mead Johnson, Nestec Nutrition, BioGaia, and personal fees from Nestle, Danone, Biocodex, Shire, Abbvie, R-Biopharm, Vifor, Pharmacosmos, Celgene, ThermoFisher, Janssen, outside of the submitted work. AP reports research grants from BioGaia and Abbvie, speakers’ honoraria from Nestle, Nutricia, Vian, Friesland, Abbvie, Aventis and fees for participating in Advisory Board from Adare Pharmaceuticals and Adacyte Therapeutics, outside of the submitted work. PB reports speakers’ honoraria from Abbvie, Ferring and Nutricia as well as fees for participating in Advisory Board from Biocodex, outside of the sumitted work. ER reports speaker's honorarium from Abbvie, outside the submitted work. EM received grant or research support from Nestle Italy and Nutricia Italy, served as a member of the advisory board for Abbvie, and received payment/honoraria from Ferring, outside the submitted work. ZM received personal fees and travel grants from GlaxoSmithKline, Abbvie, Pharmas, Wurth, outside the submitted work. ÁC reports grants and personal fees from Abbvie, Biogaia, Ferring and Danone, outside of the submitted work. All other remaining authors declare that they have no competing interests.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.45

OP049 Helicobacter Pylori Resistance To Antibiotics in Europe in 2018 and Its Relationship To Antibiotic Consumption in The Community

R Bruyndonckx ^1,², F Megraud ^3,^4,^✉, S Coenen ¹, L Wittkop ⁵, T-DD Huang ⁶, M Hoebeke ⁶, L Bénéjat ^3,⁴, P Lehours ^3,⁴, H Goossens ¹, Y Glupczynski ^1,⁶; H. pylori resistance European multicentric study group*, European Surveillance of Antimicrobial Consumption Network (ESAC-Net)

Introduction

Resistance to antibiotics is the major cause of treatment failure of Helicobacter pylori infection.

Aims & Methods

Our aim was to assess prospectively the antibacterial resistance rates in H. pylori strains in Europe and to study the link between antibiotic consumption in the community collected in the European Surveillance of Antimicrobial Consumption Network (ESAC-Net) and H. pylori resistance levels across Europe. We investigated the proportion of primary antibiotic resistance of H. pylori in 2018 in 24 centres from 18 European countries according to a standardized protocol using similar methods as previously described¹. A uni- and a multivariable logistic regression analysis were also carried out to define the factors of antibiotic resistance. Data on antibiotics for systemic use in the community for the period 2008-2017 were expressed in Defined Daily Doses (DDD) per 1,000 inhabitants per day (DID). The fit of models and the degree of ecological association between antibiotic consumption and resistance data were assessed using generalised linear mixed models. Poisson and negative binomial distribution models were assessed on yearly consumption up to 6 years backwards from 2017 and on cumulative consumption up to 10 years. The model with the best fit was selected by means of the Akaike Information Criteria (AIC) (lowest AIC value).

Results

H. pylori resistance rates for 1,225 adult patients included were 21% for clarithromycin, 16% for levofloxacin and 39% for metronidazole and were significantly higher in Central/Western and Southern than in Northern European countries. in the multivariable logistic regression analysis, the factors associated with resistance were essentially linked to the region of birth: Southern Europe for clarithromycin (OR:3.7, 95%CI [1.4-9.5]) and outside Europe for metronidazole (OR:2.7, 95%CI [1.2-6.2]) compared to Northern Europe. Model fit improved for each year of antibiotic consumption accumulated, but the best fit was obtained for the Poisson distribution using 2013 data. Large variations in consumption of antibiotics in the community were observed between countries: the use of macrolides, mostly intermediate acting macrolides (clarithromycin), ranged from 1.22 DID in The Netherlands to 6.94 DID in Greece and the use of quinolones, mostly second-generation quinolones, ranged from 0.54 DID in Norway to 3.55 DID in Italy. A significant association was found between clarithromycin resistance in H. pylori and consumption in the community of macrolides (OR=1.17 (95%CI: 1.08-1.28); p=0.0003) and intermediate-acting macrolides (OR=1.16 (95%CI: 1.06-1.27); p=0.005), and between levofloxacin resistance and consumption of quinolones (OR=1.57 (95%CI: 1.31-1.89); p=0.0002) and second-generation quinolones (OR=1.63 (95%CI: 1.33-1.99); p=0.0003).

Conclusion

In a large number of countries and regions of Europe the rate of clarithromycin resistance is currently too high (>15%) to allow its empiric use in standard H. pylori eradication regimens. Likewise, the level of levofloxacin resistance (at or above 15%) also jeopardizes its use as second- or third-line treatment at many places. Knowledge of historical macrolide and quinolone consumption in the community may provide a simple and useful tool to predict the susceptibility of H. pylori to clarithromycin and to levofloxacin and to optimize the treatment strategies.

Disclosure

Support was received from bioMerieux for Etests and from Mo-bidiag for real time PCR kits.

References

1.Megraud F., Coenen S., Versporten A., Kist M., Lopez-Brea M., Hirschl A.M., Andersen L.P., Goossens H., Glupczynski Y. on behalf of the Study Group participants. Helicobacter pylori resistance to antibiotics in Europe and its relationship to antibiotic consumption. Gut 2013; 62: 34–42. *List of other ‘H. pylori resistance European multicentric study group’ participants: Makristathis (Austria), L. Boyanova (Bulgaria), M. Tonkic (Croatia), L. Andersen (Denmark), B. Blumel, E. Glocker, A. Link, S. Suerbaum (Germany), A. Mentis (Greece), S. Smith (Ireland), M.P. Dore, R. Monno (Italy), D. Rudzite, M. Leja (Latvia), J. Kupcinskas (Lithuania), K. Melby (Norway), G. Gosciniak, T. Karpinski (Poland), M. Oleastro (Portugal), S. Jeverica (Slovenia), X. Calvet, S. Lario, M. Montes, A. Morilla (Spain), S. Boonstra, P. Schneeberger (The Netherlands)22580412 [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.46

OP050 Changes in Antibiotic Resistance Genes and Gut Microbiota After Primary Or Repeated Courses of H. Pylori Eradication Therapy

LL Wang ^1,^✉, HB Yao ², T Tong ¹, KS Lau ¹, A Chan ³, JWK Ho ², SY Leung ³, WK Leung ¹

Introduction

The antibiotic resistance rates of H pylori (HP) have increased dramatically due to overuse of antibiotics, which pose a major challenge on the success of conventional eradication therapy. Repeated use of antibiotics for HP could potentially alter the gut microbiota and antibiotics resistance genes (ARGs).

Aims & Methods

We determined the short-term and long-term impacts of different HP eradication therapies on the ARGs and gut microbiota of HP-infected patients. 44 HP-infected patients were recruited, including 21 with 1^st line therapy (clarithromycin-based triple therapy), 16 for 2^nd line and 7 for rescue therapy (>2 previous courses). A total of 121 stool samples were collected at three time points: before current HP eradication therapy (baseline), 4-8 weeks after the completion of anti-HP therapy and 6-months afterwards. Each sample had an average of 8 Gb throughput, and 54 million reads from metagenomic shotgun sequencing (Illumina NovaSeq 6000, PE151bp). Raw data were assembled into long contigs using MEGAHIT before applying the Resistance Gene Identifier (RGI) for potential ARGs against the latest version of the Comprehensive Antibiotic Resistance Database (CARD). Taxonomic profiling was performed using Kraken2 against Refseq bacteria genomes as the reference database, and species diversity was computed using vegan package in R.

Results

We found a total of 378 unique ARGs at 90% similarity cutoff, with ErmB and tet (40) detected in all 121 samples. Besides, tetO, tetQ, tetW, ErmF, dfrF, and AAC (6’)-le-APH (2'’)-la were detected in >95% of samples. Compared with baseline, significant changes of ARGs were found in the 2^nd line therapy group with the number of unique ARGs increased significantly (65 to 85, p=0,022), the relative abundance of tetracycline resistance genes decreased (63.3% to 39.9%, p=0.015), fluoroquinolone resistance genes increased (0.37% to 1.63%, p=0.006) 4-8 weeks after the treatment. More specifically, 45 different ARGs had significantly increased relative abundance after 4-8 weeks, and 14 ARGs increased after 6 months in 2^nd line therapy group. Antibiotics had relatively little effects on the ARGs in the 1^st line with only 2 ARGs increased significantly after 4-8 weeks, and 3 ARGs changed after 6 months. in rescue therapy group, no significant difference was found after 4-8 weeks, but 4 ARGs significantly increased after 6 months. The number of bacterial species detected and bacterial diversity significantly declined 4-8 weeks after 1^st line (p=0.011), 2^nd line (p=0.010), and rescue therapy(p=0.012), and returned to baseline level 6 months later. The changes in Proteobacteria phylum in the 2^nd line group was more profound and long-lasting, increased after 4-8 weeks (3.54% to 7.47%, p=0.005), and continued to 6 months (3.54% to 7.96%, p=0.03).

Conclusion

Antibiotics used in various eradication therapies had significant short-term impacts on the gut microbiota and ARGs, which usually resolved at 6-month. However, second-line therapy that contained tetracycline, amoxicillin, levofloxacin, or metronidazole had more profound and long-lasting impacts on gut microbiota and ARGs.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.47

OP051 Endoscopic Management of Post-Operative Anastomotic Leakage Or Fistula After Esophagogastric Resection For Malignancy: A French Multicenter Experience

R Hallit ^1,^✉, C Melanie ², U Chaput ³, D Lorenzo ⁴, A Becq ⁵, M Camus ⁶, X Dray ⁷, J-M Gonzalez ⁸, M Barthet ⁹, J Jacques ¹⁰, R Legros ¹¹, P Cattan ², A Belle ¹², S Chaussade ¹³, R Coriat ¹⁴, F Prat ¹⁵, D Goere ¹⁶, M Barret ¹⁷

Introduction

Esophagectomy remains the standard of care for a vast proportion of esophageal or esophagogastric lesions. However, this procedure is associated in 10% of cases with anastomotic leakages or fistulas. Endoscopic management of post-operative anastomotic complications has been mostly reported with esophageal stents. Recently, internal drainage by double pigtail stents has been introduced in the management of these complications. Our aim was to assess the overall efficacy of the endoscopic treatment for anastomotic leaks and fistulas after esophago-gastric resection for cancer.

Aims & Methods

We conducted a multicenter retrospective study in five reference centers for digestive endoscopy in France including 68 patients operated on between January 2016 and December 2018. We included patients with anastomotic leakage or fistula following surgical resection of cancers of the esophagus or the esophago-gastric junction. The primary outcome was the efficacy of the endoscopic management on fistula closure. The secondary outcomes were the efficacy of internal drainage with a double pigtail stent or anastomotic coverage with a self-expandable metal stent (SEMS), the number of endoscopic treatment sessions needed, the factors associated with treatment success, and the incidence of anastomotic esophageal strictures.

Results

68 patients were included, among which 46 men and 22 women, with a mean±SD age of 61±11 years. 44% had an Ivor Lewis procedure, 16% a tri-incisional esophagectomy, and 40% a total gastrectomy. 51% had received neoadjuvant chemotherapy, 12% neoadjuvant chemoradio-therapy, and 37% did not receive any neoadjuvant treatment. The median (IQR) time between surgery and fistula diagnosis was 9 (6-13) days. Overall, endoscopic treatment was effective in 90% (61/68) of the patients. The

efficacy of internal drainage and anastomotic coverage were 95% (36/38) and 79% (23/29), respectively (p= 0.06). The mortality rate was 3%. in univariate analysis, the only predictor of successful endoscopic treatment was the use of internal drainage (p = 0.002).

Conclusion

Endoscopic management of early postoperative leakages and fistulas is successful in 90% of patients, making it possible to avoid highly morbid surgical revisions. Internal endoscopic drainage should be considered as the first-line endoscopic treatment of anastomotic fistulas complicating esophagogastric cancer resection surgery whenever technically feasible.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.48

OP052 Acute Esophageal Perforation: Does Endoscopic Vacuum Therapy Abandon Surgery?

P Stathopoulos ^1,^✉, S Wächter ², L Schiffmann ³, C Bauer ⁴, TM Gress ⁴, D Bartsch ², G Seitz ⁵, UW Denzer ¹

Introduction

Endoscopic vacuum therapy (EVT) has been proven to be effective for closure of postoperative anastomotic leaks in the GI tract. in case of acute esophageal perforation surgical therapy is currently the gold standard treatment. However casuistic reports showed also promising results of EVT. We report a series of acute esophageal perforations treated with EVT successfully.

Aims & Methods

Mai 2018 - Februar 2019 eight patient (m:3; 3-79 y) experienced acute esophageal perforation, all treated with EVT. After positioning of a gastric feeding tube into the stomach, the esophageal sponge (Esosponge, Braun) was endoscopically inserted into the esophageal lumen covering the perforation site or placed intramediastinal. The sponge was exchanged twice a week until complete endoscopic closure. Patients were covered with a broad spectrum antibiotic iv.

Results

Perforation cases were as follows: One child (3 y) with esophageal stricture four weeks after caustic ingestion emerged a 20 mm perforation in the middle esophagus complicated by a pneumothorax during endoscopic dilatation therapy. in 4 patients (71y - 79 y) acute perforation occurred in the distal esophageal part (30- 50mm) after endoscopic pneumatic balloon dilatation (30 mm, Rigiflex) for achalasia treatment. The other three perforations were located in the proximal esophagus. One female patient (67 y) with esophageal involvement of pemphigus vulgaris showed perforation of approx. 20 cm length during the initial esophageal intubation. A 25 year old man experienced a 20 mm perforation below the upper esophageal sphincter after ingestion of a broken piece of glass under alcoholiziation and in a 79y old patient 15 mm perforation occured during endoscopic Zenker Myotomy.

Endoscopic vacuum therapy was started: Immediately after perforation n = 6, Day 1 after clipping n =1, day 2 due to a delayed diagnosis n =1. All sponges were placed intraluminal, only the patient with delayed diagnosis needed intramediastinal sponge positioning due to abscess formation during the first week. Mean number of sponge exchanges needed were 2.9 (1-5) with a mean duration of EVT therapy of 10.5 days (3-21). After endoscopic diagnosis of closure patients gradual returned to solid food under clinical control. EVT resulted in complete closure of the acute esophageal perforation in all 8 patients.

Conclusion

Endoscopic vacuum therapy is able to close acute esopha-geal perforation within 1 to 3 weeks. Intraluminal positioning of the sponge was mostly sufficient. Immediate start of EVT to prevent abscess formation and induce defect closure is crucial.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.49

OP053 High Proficiency of Oesophageal Endoscopic Submucosal Dissection Thanks To “Tunnel + Clip” Strategy: A Large French Multicentric Study

T Wallenhorst ^1,^✉, J Jacques ², R Legros ³, S Stephant ¹, C Brochard ⁴, M Barret ⁵, A Lupu ⁶, F Rostain ⁶, J Rivory ⁶, M Pioche ⁷

Introduction

Endoscopic Submucosal Dissection (ESD) is the treatment of choice for superficial neoplasms of the oesophagus. ESD is oncologi-cally efficient and associated with less morbidity than the surgical alternative. ESD requires a high level of skill, being technically challenging and time consuming, so its performance is often limited to experts. A combination of a tunnel technique with clip-line traction (1) may enable optimisation of oesophageal ESD.

Aims & Methods

The aims were to describe the feasibility and the safety of the tunnel technique with clip-line traction performed by different operators in a multicenter study. From January 2015 to January 2020, we performed a retrospective three-centre case study of consecutive “tunnel + clip” oesophageal ESD. Six young operators (each of whom had performed fewer than five oesophageal ESDs) treated patients requiring ESD using the tunnel + clip method developed in our team. This involves generation of a classic tunnel beneath the lesion, with constant traction being applied by a clip with a line placed at the oral side of the tunnel (1).

Results

One hundred and seventy-six lesions (56 SCCs and 120 ADK/HGDs complicating Barrett's oesophagus) were consecutively resected. Overall, the en bloc, R0 and curative resection rates were 100.0% (176/176), 82.3% (145/176) and 73.1% (129/176), respectively. The R0 resection rates were comparable between SCCs (76.8%; 43/56) and ADK/HGDs complicating Barrett's oesophagus (85.0 %; 102/120), respectively (p=0.18). The curative resection rate was lower in SCCs (57.1%; 32/56) than in ADK/HGDs complicating Barrett's oesophagus (80.7%; 96/120) (p=0.001). Vertical and lateral margins were free in 89.7% (158/176) and 88.6% (156/176), respectively. One specimen (0.5%) had pathological damage caused by clipping.

One perforation (0.5%) and 6 (3.4%) post-procedural bleeding occurred; these were managed endoscopically. At 3 months, 25/176 (14.2%) patients had stricture treated by endoscopic dilation. The mean ESD velocity was 30.4 mm²/min for lesions of mean area 1831.9 mm² (mean oesophageal circumference: 56.4%). The pathological type, en bloc, R0, curative resection, perforation and stricture rates were comparable between the operators (Table). The rate of bleeding differed between operators but all of them were treated endoscopically.

Conclusion

This study is the largest Western series of oesophageal ESD. The tunnel+clip strategy is effective and safe, with reproductible results even when performed by physicians with little prior experience in oe-sophageal ESD. The results of this procedure will encourage the use of oesophageal ESD in Western countries.

Table 1.

[Results according to the operators.]

1	2	3	4	5	6	p-value
Number of lesions	68	50	24	17	11	6
SCCs / ADK-HGDs-n (%)	23 (33.8)/45 (66.2)	12 (24.0)/ 38 (76.0)	11 (45.8)/ 13 (54.2)	3 (17.6)/14 (82.3)	5 (45.5)/6 (54.6)	2 (33.3)/ 4 (66.7)	0.28
En bloc resection-n (%)	68 (100)	50 (100)	24 (100)	17 (100)	11 (100)	6 (100)	0.0
R0 resection-n(%)	55 (80.9)	41 (82.0)	20 (83.3)	14 (82.4)	9 (81.8)	6 (100)	0.92
Vertical margin free -n(%)	59 (86.8)	48 (96.0)	21 (87.5)	15 (88.2)	9 (81.8)	6 (100)	0.49
Lateral margin free-n(%)	63 (92.7)	42 (84.0)	21 (87.5)	15 (88.2)	9 (81.8)	6 (100)	0.63
Curative resection-n(%)	53 (77.6)	36 (72.0)	15 (62.5)	12 (70.6)	7 (63.6)	6 (100)	0.42
Oesophageal circumference of the resected area(%)	50.6	58.2	68.3	57.6	54.1	58.3	0.01
Perforation - n (%)	0 (0)	0 (0)	2 (4.2)	0 (0)	0 (0)	0 (0)	0.27
Bleeding-n (%)	1 (1.5)	0 (0)	2 (8.3)	1 (5.9)	2 (18.2)	0 (0)	0.04
Stricture- n (%)	11 (17.5)	2 (4.5)	3 (12.5)	3 (16.7)	4 (36.4)	2 (33.3)	0.08

Open in a new tab

Disclosure

Nothing to disclose

References

1.Jacques J., Legros R., Rivory J., Charissoux A., Sautereau D., Ponchon T., Pioche M. The “tunnel+clip” strategy standardised and facilitates oesophageal ESD procedures: a prospective, consecutive bi-centric study. Surg Endosc. 2017. Nov; 31: 4838–4847. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.50

OP054 Long Term Outcomes of Peroral Endoscopic Myotomy: A Single Center Experience with More Than 350 Procedures

D Dražilova ^1,^✉, Z Rabekova ², Z Vackova ³, T Hucl ⁴, P Štirand ³, J Krajciova ⁵, J Spicak ⁶, E Kieslichová ⁷, R Janoušek ⁷, J Martinek ⁸

Introduction

Peroral endoscopic myotomy (POEM) has become a standard treatment of esophageal achalasia due to its excellent efficacy and safety, however, long term durability of treatment success needs to be further investigated as POEM is still relatively new procedure.

Aims & Methods

The aim of our analysis was to assess the long-term clinical outcomes of patients undergoing POEM for esophageal achalasia with a minimum follow up of 3 years. Our study is a retrospective analysis of prospectively collected data of all consecutive patients who underwent POEM at our institution between December 2012- March 2020. All patients were scheduled for follow up at 3 and 12 months after the procedure and every year thereafter. Upper GI endoscopy, high-resolution manometry (HRM) and 24-hour pH monitoring were performed 3 months after POEM; endoscopy was then repeated between 24-36 months. Main outcomes were treatment success defined as Eckardt score < 3, recurrence rate and parameters of post-POEM gastroesophageal reflux (24h pH monitoring, presence of reflux esophagitis, reflux symptoms and use of proton pump inhibitors (PPIs).

Results

A totalof 348 patients underwent 364 POEM procedures, an initial treatment success (at 12 months) was present in 96% (95% CI 92-98) of patients. Follow- up visits at 36, 48, 60 and 72 months were completed in 122, 76, 42 and 13 patients. At 36, 48, 60 and 72 months, treatment success was achieved in 88% (95% CI 82- 92), 83% (75- 89), 81% (72- 88) and 81% (68-89) of patients. A total of 6 patients experienced treatment failure (no symptomatic response after POEM) and 22 patients experienced a recurrence. At 3 months, reflux esophagitis was observed in 128/317 (40.4%, LA C/D in 15 patients). At 24-36 months, reflux esophagitis was present in 38 /102 patients (37.3%, no LA C/D). PPIs were administered at 36, 48 and 60 months to 42.8%, 51.4% and 57.1% of patients, respectively

Conclusion

POEM is a highly effective endoscopic treatment for achalasia with favorable treatment success of 80% at 6 years after the procedure. Higher rate of post-POEM reflux esophagitis seems to decrease over time. However, approximately 50% of patients need to take PPIs on a long term basis.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.51

OP055 Comprehensive Motility Analysis in Patients with Decompensated Gastroparesis Undergoing Gastric Per-Oral Endoscopic Pyloromyotomy: Relation To Treatment Outcomes

JM Conchillo ^1,^✉, JWA Straathof ¹, Z Mujagic ¹, N Bouvy ², D Keszthelyi ¹, AA Masclee ¹

Introduction

Gastric per-oral endoscopic pyloromyotomy (G-POEM) has been recently introduced as an alternative therapeutic modality for gastroparesis patients with refractory symptoms. However, there are no reliable data to predict which patients would benefit the most from G- POEM.

Aims & Methods

Aim of the present study is to assess whether antro-duodenal motility patterns and pyloric distensibility can predict the outcome of G-POEM in patients with decompensated gastroparesis. From September 2017 to September 2019 we conducted an open-label study in a tertiary center at Maastricht University Medical Center. Patients with gastroparesis and refractory symptoms were eligible for treatment with G-POEM if treatment attempts according to a standardized stepwise protocol had failed and all were on enteral feeding support. Baseline assessment included Gastroparesis Cardinal Symptom Index (GCSI), C13 gastric emptying breath test and 6-hr high-resolution antro-duodenal manometry (ADM). Antral hypomotility was defined as antral motility index < 13.67. The following neuropathic patterns were scored: presence of bursts, retrograde peristalsis, clustered contractions and absence of phase III contractions. Pyloric dystensibility using EndoFlip measurements was assessed at baseline and 3 months after the procedure.

Clinical success was defined as decrease of one point in GCSI score. Explorative analyses were performed on potential predictors of response using logistic regression analyses.

Results

Forty-five patients with gastroparesis were screened. Twenty-four patients (21 women, mean age 55.3 ± 3 yrs.) with decompensated gas-troparesis (11 idiopathic, 7 post-surgical, 6 diabetes mellitus) fulfilled the inclusion criteria and underwent G-POEM under propofol deep sedation. Mean disease duration was 44 ± 7 months and baseline GCSI score was 5.7 ± 0.3. During baseline ADM, 78.3% of patients showed antral hypomotility. Thirteen out of 21 patients (61.9%) showed at least 1 neuropathic motor pattern: 28.6% of patients showed bursts, 19% retrograde peristalsis, 19% absence of phase III contractions and 19% clustered contractions. A significant correlation was found between idiopathic gastroparesis and presence of neuropathic patterns during ADM (X²= 6.4, p=0.01). Technical success rate of G-POEM was 100% (24/24). Clinical success rate was 87.5% at 3 months, 75% at 6 months and 50% at 12 months after G-POEM. Mean GCSI improved significantly at 3, 6 and 12 months after G-POEM (p=0.007). Median [Q1;Q3] distensibility index (DI, mm²/mmHg) at 40mL distension volume improved significantly as compared with baseline (7.5 [6.9;11.7] vs. 5.3 [3.1;8.1], p=0.004). Similarly, both pyloric cross-sectional area (CSA) and pressure improved significantly 3 months after G-POEM (p=0.02 and p=0.04, respectively). A significant correlation was found between clinical response at 6 months and pyloric DI improvement (X²=9.4, p=0.002). No potential predictors of clinical response after G-POEM could be identified among antro-duodenal motility parameters and pre-G-POEM pyloric distensibility in an explorative analysis.

Conclusion

Per-oral endoscopic pyloromyotomy improved pyloric distensibility patterns in patients with decompensated gastroparesis. Clinical response at 6 months after G-POEM was associated with pyloric distensibility improvement. However, no potential predictors of response could be identified from either antro-duodenal motility patterns or pre-G-POEM pyloric distensibility.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.52

OP056 Impact of Prenatal Stress On Visceral Sensitivity and Intestinal Homeostasis in Adulthood

C Petitfils ^1,^✉, G Payros ², G Dietrich ², J-P Motta ², M Serino ², P Le Faouder ³, N Cenac ⁴

Introduction

Environmental factors, and predominantly stress during pregnancy, could lead to gut microbiota dysbiosis in adult offspring and increase the risk of developing irritable bowel syndrome (IBS). However, a causal link between prenatal stress-induced compromised sensitivity and dysbiosis has yet to be established. We hypothesized that prenatal stress (PS) in mice would predispose adult offspring to overreact to stress leading to visceral hypersensitivity and intestinal homeostasis disruption as observed in IBS.

Aims & Methods

Prenatal stress (PS) was induced by using restriction stress with bright light for 30 minutes, three times a day, in dams, between day 13 and 18 of gestation. Eight-week-old offspring, both male and female, underwent 1 session of water avoidance stress (WAS) for 1 hour, 24 hours before assessing visceral sensitivity to colorectal distention. Cxcl2, Tgfb, Ccl5, Reg3g, Muc2, Occln, Ttf3, Mmp7, Penk and Ifng expressions in the colon were monitored by qRT-PCR and intestinal bioactive lipids by mass spectrometry. The gut microbiota was assessed by MiSeq-based microbial taxonomic analysis to determine the faecal microbiota composition and by 16S RNA FISH staining to visualise its organization as a biofilm.

Results

WAS induced a significant increase of visceral sensitivity to colorectal distension in male and female PS mice. PS alone or WAS in control mice did not significantly modify visceral sensitivity. While there were no changes in gene expression in the colons of both sexes, PS-male colons were characterised by a decrease of the 5-lipoxygenase pathway metabolites. PS alone induced a disruption of the biofilm in 8-week-old male and female adult offspring, marked by a low number of bacterial infiltrations into the sterile mucus layer. WAS worsened the disruption of the biofilm in male and female PS mice characterized by numerous bacterial infiltrations in the sterile mucus layer and bacteria in contact with the epithelial cells. However, WAS did not induce any changes in the biofilm organisation in control mice. While Akkermansia muciniphila was increased in both male and female stressed mice, the strains Desulfovibrio sp and Lactobacillus animalis were decreased in male and in female stressed mice, respectively.

Conclusion

Prenatal stress is sufficient to induce gut microbiota dysbiosis and biofilm disruption. WAS worsened the gut microbiota alteration and induced mechanical visceral hypersensitivity without modification of mucus, tight junction proteins, anti-microbial peptides, chemokines and cytokines genes expression. This study shows that prenatal stress could represent an important priming event for the development of IBS in response to harmless stress for adult controls.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.53

OP057 Early Response To Plecanatide Predicts Overall and Sustained Efficacy in Patients with Irritable Bowel Syndrome with Constipation

N Martinez de Andino ¹, R Patel ², S Lorenzen ³, EM Quigley ^4,^✉

Introduction

Irritable bowel syndrome with constipation (IBS-C) is a common and bothersome functional gastrointestinal disorder. Plecanatide, a uroguanylin analog indicated for the treatment of adults with IBS-C or chronic idiopathic constipation, has demonstrated efficacy and safety in two phase 3 studies in IBS-C. This pooled analysis evaluated the predic-tiveness of treatment response during Week 2 or Week 4 on overall treatment response after 12 weeks in patients with IBS-C in the two studies.

Aims & Methods

Data were pooled from two double-blind, placebo-controlled studies of patients with IBS-C (NCT02387359, NCT02493452). Patients were randomized to placebo, plecanatide 3 mg, or plecanatide 6 mg for 12 weeks. The primary endpoint was the percentage of overall responders (ORs), defined as weekly responders (WRs) for >6 of 12 treatment weeks; WRs were both abdominal pain responders (>30% decrease in worst abdominal pain vs baseline) and stool frequency responders (increase >1 complete spontaneous bowel movement [CSBM] vs baseline) in the same week. Weekly response during Weeks 2 or 4 were assessed as predictors of overall and sustained response after 12 weeks in the intent-to-treat population (duplicates removed).

Results

A total of 2176 patients with IBS-C (placebo, n=729; 3 mg, n=724; 6 mg, n=723) were included. At baseline, mean CSBMs/week and SBMs/ week were ∼0.25 and ∼1.46, respectively, across groups; mean abdominal pain severity score was ∼6.25, indicating moderate severity. During Week 2, 11.9% (placebo), 20.7% (3 mg), and 20.1% (6 mg) of patients were WRs, increasing to 16.2%, 24.2%, and 24.3%, respectively, during Week 4. Across Weeks 1-12, significantly more plecanatide-treated patients were ORs than placebo patients (placebo, 16.0%; 3 mg, 25.6%, P< 0.001; 6 mg, 26.7%, P< 0.001). Being a WR during either Week 2 or 4 was significantly predictive of OR status during weeks 1-12. Patients experiencing Week 2 response were 29.5 times (95% CI: 21.96, 39.58; P< 0.001) more likely to be ORs than those not experiencing Week 2 response. Patients with week 4 response were 61.3 times (95% CI: 45.25, 82.93; P< 0.001) more likely to be ORs than those not experiencing week 4 response. Similar results were seen in sustained efficacy responders (ORs who are WR for >2 of the last 4 weeks).

Conclusion

Plecanatide 3 mg, as well as 6 mg, is an effective treatment for IBS-C and its symptoms. A clinical response to plecanatide as early as Week 2 appears to be predictive of overall response after 12 weeks of treatment.

Disclosure

Funded by Salix Pharmaceuticals, LLC, Bridgewater, NJ, USA. This presentation has been modified from Digestive Disease Week 2020.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.54

OP058 Expression and Localization of Cannabinoid Receptors and The Effect of Olorinab, A Peripherally Acting, Highly Selective, Full Agonist of The Cannabinoid Receptor 2, On Visceral Hypersensitivity in Rodent Models of Irritable Bowel Syndrome and Inflammatory Bowel Disease

J Castro ^1,^✉, S Garcia-Caraballo ¹, J Maddern ¹, G Schober ¹, B Lindstrom ², S Schmiel ², SM Brierley ¹

Introduction

Patients with inflammatory bowel disease (IBD) and irritable bowel syndrome (IBS) commonly experience abdominal pain. Treatment options for abdominal pain associated with IBD and IBS are suboptimal, representing a significant unmet need. in preclinical models of gastrointestinal inflammation, increased expression of cannabinoid receptor 2 (CB₂) in the colon has been demonstrated; additionally, CB₂ is upregulated in gut tissue from patients with IBD and IBS. Olorinab—a highly selective, peripherally acting, full agonist of CB₂—has shown antinociceptive activity in animal models of pain and is under investigation for the treatment of chronic visceral pain in patients with IBD and IBS.

Aims & Methods

Here we evaluated the effects of olorinab on visceral hypersensitivity in rodent models of colitis (IBD-like) and chronic visceral hypersensitivity (CVH; IBS-like) and determined potential sites of olorinab activity. Colitis and CVH were induced in male Sprague Dawley rats or C57BL/6 mice by intracolonic enema of 2,4,6-trinitrobenzene sulfonic acid (TNBS) or 2,4-dinitrobenzene sulfonic acid (DNBS), respectively. Colitis was evaluated on Day 4 following induction, while CVH was evaluated on Day 28. in the colitis model, control and colitis rodents were administered vehicle or olorinab (3 or 30 mg/kg twice daily) orally for 5 days starting 1 day before colitis induction. in the CVH model, control and CVH rodents were administered vehicle or olorinab (3, 10, or 30 mg/kg twice daily) orally on Days 24 to 28 following colitis induction. Visceral hypersensitivity was measured through in vivo visceromotor responses (VMR) to colorectal distention (CRD; 0 to 80 mm Hg) on Day 4 (colitis) and Day 28 (CVH). The mRNA expression of CB₁ and CB₂ was determined by quantitative real-time polymerase chain reaction (qRT-PCR) in colonic tissue (mucosa; muscle and enteric nervous system [ENS]) and dorsal root ganglia (DRG) from healthy, colitis, and CVH rodents, and in healthy and IBD human donor DRG. The localization of CB₂ in healthy, colitis, and CVH mice was confirmed through RNA in situ hybridization (ISH).

Results

Vehicle-treated colitis and CVH rodents exhibited pronounced visceral hypersensitivity compared with healthy controls. Treatment with olorinab significantly reduced VMR to CRD in colitis and CVH rodents compared with vehicle-treated counterparts (P < 0.05). CB₁ and CB₂ mRNA levels were detected in all tissues analyzed. CB₂ was the predominant transcript in colonic mucosa, while in the colonic muscle + ENS and rodent DRG, CB₁ mRNA was the most abundant. There were no differences in CB₂ expression in colonic mucosa, colonic muscle + ENS, and DRG observed across the healthy, colitis, or CVH states. Additionally, no differences were found in CB₂ mRNA levels in healthy and IBD human donor DRG. The presence of CB₂ in mouse colonic tissues and mouse DRG was verified by ISH.

Conclusion

Olorinab reduced visceral hypersensitivity in colitis and CVH rodents; these results suggest that CB₂ activation causes antinociceptive actions in visceral sensory pathways. The presence of CB₂ in colonic tissue and in primary sensory DRG neurons was validated through qRT-PCR and ISH studies, implying that both sites may contribute to the antinociceptive activity of olorinab. The colocalization of CB₂ receptors in specific cell types are under investigation to identify the mechanisms involved in

olorinab-mediated responses. These results support the further investigation of olorinab for the treatment of IBD- and IBS-associated abdominal pain.

Disclosure

S. Schmiel and B. Lindstrom are employees of Arena Pharmaceuticals, Inc. S.M. Brierley received research funding from Arena Pharmaceuticals to conduct the study. J. Castro, S. Garcia-Caraballo, J. Maddern, and G. Schober have nothing to declare.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.55

OP059 Long-Term Effects of Faecal Microbiota Transplantation (Fmt) in Patients with Irritable Bowel Syndrome

M El-Salhy ^1,^✉, A Brâthen Kristoffersen ², J Valeur ³, C Casén ⁴, JG Hatlebakk ^5,^6,⁷, OH Gilja ^8,⁹, T Hausken ¹⁰

Introduction

A recently published randomized double-blind placebo-controlled study from our group found faecal microbiota transplantation (FMT) to be an effective and safe treatment for irritable bowel syndrome (IBS) after 3 months (1). The present follow-up study investigated the efficacy and safety of FMT at 1 year after treatment.

Aims & Methods

This study included 77 of the 91 IBS patients who had responded to FMT in our previous study (i.e. 14 patients were either excluded or dropped out). Patients provided a faecal sample and completed five questionnaires to assess their symptoms and quality of life at 1 year after FMT. Abdominal symptoms, fatigue and quality of life were assessed using the IBS Severity Scoring System (IBS-SSS), Birmingham IBS Symptom, Fatigue Assessment Scale (FAS), IBS Quality of Life and the Short-Form Ne-pean Dyspepsia Index questionnaires. The dysbiosis index (DI) and faecal bacterial profile were analysed using a CE marked 16S rRNA gene-based DNA probe hybridisation method (2). The levels of faecal short-chain fatty acids (SCFAs) were determined by gas chromatography.

Results

The response to FMT was maintained at 1 year after treatment in 32 (86.5%) and 35 (87.5%) patients who received 30-g and 60-g FMT, respectively. in the 30-g FMT group, 12 (32.4%) patients showed complete remission (had an IBS-SSS total score of ≤75) at 1 year, compared with 8 (21.6%) after 3 months. in the 60-g FMT group, 18 (45%) patients were in complete remission at 1 year, compared with 11 (27.5%) after 3 months. Abdominal symptoms, fatigue and the quality of life were improved at 1 year compared with 3 months after FMT. These findings were accompanied by a significant improvement in the DI and comprehensive changes in the faecal bacterial profile. The levels of Alistipes spp. — which belong to the phylum Bacteroidetes— were significantly lower in the relapsed patients at baseline than in the responders and patients in remission at 1 year after FMT. Furthermore, Alistipes spp. levels increased as early as 1 month after FMT and remained high in responders at 1 year after FMT. Moreover, they were strongly correlated with the total scores on both IBS-SSS and FAS (r=-0.479 and P< 0.001, and r=0.436 and P< 0.001, respectively). Changes in the levels of faecal SCFAs indicated that the microbial metabolism changed from a saccharolytic to a proteolytic fermentation pattern in IBS patients at 1 year after FMT. The level of faecal acetic acid was reduced compared with that at baseline. The clinically relapsed patients had significantly lower fatigue scores and significant changes in the bacterial profile and the levels of SCFAs compared with those at baseline. No adverse events were reported.

Conclusion

Most of the IBS patients who responded to FMT after 3 months maintained a response at 1 year after FMT. Moreover, the improvements in symptoms and quality of life increased significantly over time. Changes in the faecal bacterial profile and SCFAs also increased over time. The finding that FMT induced remission in about half of the patients with IBS emphasizes the importance of the intestinal microbiota as an aetio-logical factor of IBS. Alistipes spp. seem to play a central role in the improvements seen after FMT and levels of Alistipes spp. could probably be used to predict the outcome of FMT. The reduction of acetic acid levels could be relevant since acetic acid has been found to induce visceral hy-persensitivity in rodents (3).

Disclosure

Nothing to disclose

References

1.El-Salhy M., Hatlebakk J.G., Gilja O.H. Brathen Kristoffersen A and Hausken T: Efficacy of faecal microbiota transplantation for patients with irritable bowel syndrome in a randomised, double-blind, placebo-controlled study. Gut 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Casen C., Vebo H.C., Sekelja M. et al. Deviations in human gut microbiota: a novel diagnostic test for determining dysbiosis in patients with IBS or IBD. Alimentary pharmacology & therapeutics 42: 71–83, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Winston J., Shenoy M., Medley D. Naniwadekar A and Pasricha PJ: The vanilloid receptor initiates and maintains colonic hypersensitivity induced by neonatal colon irritation in rats. Gastroenter-ology 132: 615–627, 2007. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.56

OP060 Effects of Eluxadoline Treatment On Abdominal and Bowel Symptoms in Patients with Irritable Bowel Syndrome with Diarrhea Reporting Inadequate Symptom Control with Loperamide: Aggregated Results From Phase 3 and 4 Clinical Trials

DM Brenner ¹, G Sayuk ², M Saeed ³, P Janiszewski ⁴, L Liu ^5,^✉

Introduction

The heterogeneous nature of irritable bowel syndrome with diarrhea (IBS-D) presents a complex challenge in the management of a wide range of symptoms. Despite the need for more effective treatments that provide global IBS-D symptom relief, many therapies currently available in Europe only treat individual IBS-D symptoms. Eluxadoline (ELX) is a mixed μ- and κ-opioid receptor agonist and 5-opioid receptor antagonist for the treatment of IBS-D in adults.

Aims & Methods

To analyze the efficacy of ELX for managing a constellation of IBS-D symptoms among patients (pts) self-reporting inadequate responses to loperamide (LOP) from Phase 3 and 4 trials of ELX. Adults with IBS-D were randomized to placebo (PBO) or ELX (75 or 100 mg) twice daily (BID) for 26 or 52 weeks in two Phase 3 trials (NCT01553591 and NCT01553747); the data from a subset reporting inadequate IBS-D symptom control with prior LOP use were pooled and prospectively analyzed. in the Phase 4 trial (RELIEF: NCT02959983), IBS-D pts with self-reported inadequate symptom control with LOP over the previous 12 months were randomized to PBO or ELX 100 mg BID for 12 weeks. in all trials, pts recorded their daily IBS-D symptoms including worst abdominal pain (0-10 scale), stool consistency (Bristol Stool Form Scale score 1-7), and daily urgency

episodes. Results for ELX 100 mg vs. PBO were evaluated after 12 weeks of treatment, which included the generation of radar plots to visualize multiple efficacy endpoints.

Results

Of pts enrolled in the Phase 3 trials with prior LOP use, inadequate control of IBS-D symptoms was reported in 58.9% (n=166) and 58.8% (n=174) of pts in the PBO and ELX groups, respectively; in the Phase 4 trial, 174 and 172 pts were treated with PBO and ELX, respectively. ELX-treated pts had a greater proportion of responders compared to PBO-treated pts for all outcomes, including a significant increase in the proportion of pts achieving both >50% and >75% urgency-free days in the Phase 3 trials. Analysis of the global symptoms of pts enrolled in the Phase 4 trial further supported these results. Across Phase 3 and 4 trials, adverse events occurred in similar proportions regardless of treatment, with the most common being nausea.

Conclusion

In pts with inadequate responses to LOP, a greater proportion of ELX-treated pts were responders compared to PBO-treated pts in all Phase 3 and 4 trials for the respective primary composite endpoints, abdominal pain, stool consistency, and urgency after 12 weeks of treatment. These results validate the effectiveness of ELX in alleviating IBS-D symptoms in this population.

Disclosure

This study was sponsored by Allergan plc. Editorial assistance was provided to the authors by Mayuri Kerr, DDS. MS, AbbVie Inc. All authors met ICMJE authorship criteria. Neither honoraria nor payments were made for authorship. Financial arrangements of the authors with companies whose products may be related to the present report are listed below, as declared by the authors. Darren M. Brenner, MD, is a consultant and speaker for Allergan plc (now AbbVie) and Ironwood Pharmaceuticals. Gregory S. Sayuk, MD, is a consultant and speaker for Allergan plc (now AbbVie) and Ironwood Pharmaceuticals. Mazen Saeed, PharmD, is a fulltime employee and stockholder of AbbVie Inc. Peter Janiszewski, PhD, is a full-time employee and stockholder of AbbVie Inc. Louis W.C. Liu, MD, PhD, has received honoraria from Allergan plc (now AbbVie), Lupin, Medtronic, and AbbVie as a consultant and speaker, from Cipher as a consultant, and from Knight and Pendopharm as a speaker.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.57

OP061 Transplant-Free Survival in Patients with Hepatocellular Carcinoma Undergoing Locoregional Therapy - The Impact of Sarcopenia and Sarcopenic Obesity

C Lavender ^1,^✉, MV Beheshti ², JA Dranoff ¹, Group Thandassery RB, CAVHS Liver Sudy ¹

Introduction

There are no large studies describing sarcopenic obesity (SO, presence of sarcopenia and obesity) in patients with hepatocellular carcinoma (HCC).

Aims & Methods

We aimed to study the impact of sarcopenia and SO on the transplant-free survival (TFS) in patients with HCC undergoing locore-gional therapy (LRT). We also aimed to compare the sarcopenia MELD scores (derived from natural MELD-Na) with LRT prognostic scores [NIACE score (tumor Nodularity,Infiltrative tumor,serum Alpha-fetoprotein level,Child-Pugh stage,ECOG performance status),hepatoma arterial embolization prognostic score (HAP),and ABCR score (alpha-fetoprotein,BCLC,change in Child-Pugh stage and radiological response after LRT)]. We retrospectively analyzed a multi-centric United States veteran cohort (n=207), who had undergone LRT between January 2012 and March 2019. Sarcopenia was assessed by measuring lumbar-3 skeletal muscle index (SMI) from abdominal Magnetic Resonance Imaging (GE Healthcare, Waukesha, WI). Sarcopenia was defined as SMI ≤ 50 cm ²/m ² for men and ≤ 39 cm²/m² for women, and obesity was defined as BMI >30. MELD-sarco-penia (MELD-Na + 10xSarcopenia score) and MELD-SMI (MELD-0.3065xL3-SMI) were calculated along with the pre-LRT (NIACE, HAP) and post-LRT (ABCR) scores.

Results

The mean age was 65.6 ± 6.3 years (73.4% Caucasians and 26.6% African Americans). The predominant etiology of liver disease was nonalcoholic fatty liver disease (52.6%). 79 (48.2%) had no sarcopenia or obesity (group-1), 55 (35.5%) had obesity (group-2), 27 (16.5%) had sarcopenia (group-3), and 3 (1.8%) had SO (group-4). SMI was comparable between different etiological groups (p=0.73). Mean natural MELD-Na score prior to LRT was 14.1 ± 6.8.

31.4%, 20.3%, 40.6% subjects had BCLC A, B and C stages respectively. 70.5% underwent transarterial chemoembolization, 5.8% underwent ablation, and 20.3% underwent transarterial radioembolization as first LRT. At a median follow-up of 15.2 months, the overall TFS was 34.6%. BCLC stages A, B and C had 47.5%, 26.3% and 26.2% TFS respectively. The median (±SE) TFS for groups 1, 2 and 3 were 40 (±5.0), 27 (±4.9), and 17 (±5.6) months respectively (log rank 0.03, figure1a). On subgroup analysis (based on BCLC stages) the groups showed a similar trend in median TFS (log rank p=0.043), i.e., group 1 > 2 > 3. MELD-SMI had the highest AUROC (0.753) to predict TFS (p=0.041, figure 1b), more than MELD sarcopenia (0.716), MELD-Na (0.703) and the pre-LRT (NIACE-0.616, HAP-0.671) and post-LRT (ABCR-0.598) scores.

Conclusion

Sarcopenia and obesity are common in patients with HCC undergoing LRT. Both sarcopenia and obesity lower median TFS significantly (lowest with sarcopenia), independent of BCLC stages. Natural MELD-Na modified for sarcopenia (MELD-sarcopenia and MELD-SMI) have higher accuracy than natural MELD-Na and LRT prognostic scores for predicting TFS.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.58

OP062 Telemedicine in The Covid-19 Era For Liver Transplant Recipients: An Italian Lockdown Area Experience

A Santonicola ^1,^✉, F Zingone ², S Camera ¹, M Siniscalchi ¹, L Fortino ¹, C Ciacci ¹

Introduction

Data on COVID-19 in Liver Transplant (LT) recipients are scanty, but one can hypothesize that they are more susceptible to infection due to chronic immunosuppression. Telemedicine could be an alternative to the routine clinical care in this difficult period.

Aims & Methods: Aim

To investigate in a cohort of LT recipients the access to remote consultation and the attitude towards Telemedicine using an internet-based survey.

Methods

We invited LT recipients from the Liver Transplant Follow-up Center of the University Hospital of Salerno to access remote consultation. A subgroup of them also participated in a self-administered, internet-based survey evaluating demographics; LT data and immunosuppressive therapy;

comorbidities; attitudes towards COVID-19 infection; their perceptions of the need for health care, and their approach to telemedicine.

Results

Seventy-four/one hundred and fifty three (48.3%) LT recipients were unable to access remote consultation. They showed a significantly higher mean age and a higher percentage of low school degree compared to those who accessed it (p=0.03 and p=0.001, respectively). Among the remaining LT recipients, 50/79 (63.3%) responded to the survey; mean time from liver transplantation was 12±7 years; 94% of the sample reported at least one comorbidity; 44% of LT recipients declared to be “very much/much” worried because of COVID-19 infection, and 64% considered themselves more vulnerable than the general population. Forty percent of responders refused the routine follow-up visit, and 62% were very interested in using Telemedicine.

Conclusion

LT recipients were anxious because of COVID-19 infection and considered themselves more vulnerable than the general population, refusing the routine clinical visit. For continuing services during the CO-VID-19 era, new strategies are needed and telemedicine shows promise.

Disclosure

Nothing to disclose

References

Serper M., Volk M.L. Current and Future Applications of Tele-medicine to Optimize the Delivery of Care in Chronic Liver Disease. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association 2018; 16(2): 157–161 e158 [PMID: 29389489 PMCID: PMC6334286 DOI: 10.1016/j.cgh.2017.10.004] [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.59

OP063 Prevalence and Predictors of Acute On Chronic Liver Failure in Cirrhotic Patients Hospitalized For Acute Decompensation

N Trad ^1,^✉, M Ghanem ², B Ben Slimane ¹, K Boughoula ², SN Bizid ³, H Ben Abdallah ², R Bouali ⁴, N Abdelli Mohamed ²

Introduction

Acute on chronic liver failure (ACLF) is a clinical syndrome characterized by acute deterioration of chronic liver disease associated with organ failures and high short-term mortality.

Aims & Methods

Our objective was to determine the prevalence and the predictors of ACLF in cirrhotic patients hospitalized for acute decompensation.

We performed a retrospective analysis of data from consecutive patients followed in our department for cirrhosis recruited from January 2012 to December 2019. The diagnosis of ACLF was defined according to the criteria of CANONIC study; Grade 1: Isolated kidney failure or isolated organ failure associated with renal dysfunction (creatinine >15mg and < 20mg) and/or hepatic encephalopathy grade 1 or 2, or cerebral failure associated with renal dysfunction (creatinine >15mg and < 20mg); Grade 2: two organ failures; Grade 3: Three organ failures or more.

Results

A total of 195 patients were included. The mean age was 65.4 ±13.1 years and the sex-ratio was 1.7. The main etiology of cirrhosis was viral infection C (30.8%) followed by viral infection B (23.1%) and non-alcoholic steatohepatitis (13.3%). The average MELD score was 15.75. These patients were admitted to our department 460 times during the study period. ACLF occurred in 28.6% of admissions. Grade 1 was the most common (55.3%) followed by Grade 2 (22.7%) and Grade 3 (21.9%). Intra-hospital mortality was 36.3% in patients with ACLF. The predictive factors of ACLF occurrence in our series were: bacterial infection (p< 0.001), CHILD-pugh score (p=0.001), high MELD score on admission (p=0,001), advanced age > 50 years (p=0,02), history of high blood pressure (p=0,04). in multivariate analysis, only bacterial infection (p< 0.001, OR=23.6), CHILD-pugh score (p=0.02, OR=20.3) and MELD score (p=0.003,OR=10.2) were independent predictors of ACLF occurrence.

Conclusion

In our study, the prevalence of ACLF was 38% of admissions with high intra-hospital mortality. As a major precipitating factor, Bacterial infections would require rapid and effective management as well as close monitoring.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.60

OP064 An End-To-End Artificial Intelligence Model Accurately Diagnosing Hepatocellular Carcinoma On Computed Tomography

W-K Seto ^1,^2,^3,^✉, WHK Chiu ^4,⁵, P Yu ⁶, W Cao ⁶, H-M Cheng ¹, EM-F Wong ⁷, J Wu ², GC Lui ¹, X-P Shen ⁵, L-Y Mak ¹, WK Li ⁸, M-F Yuen ^1,³

Introduction

Hepatocellular carcinoma (HCC) is the fourth leading cause of cancer death worldwide. Diagnosis often requires highly characteristic dynamic patterns on contrast-enhancing cross-sectional imaging. The Liver Imaging Reporting and Data System (LI-RADS), used to interpret radiological findings related to HCC, is often limited by a lack of definitive diagnosis, resulting in treatment delays. We developed an artificial intelligence deep learning algorithm applying state-of-the-art convoluted neural network (CNN) architecture in detecting and classifying liver lesions on computed tomography (CT).

Aims & Methods

We retrieved archived tri-phasic contrast liver CT images and clinical information from different medical centers in Hong Kong and Shenzhen. We followed American Association for the Study of Liver Diseases recommendations for HCC diagnosis and employed LI-RADS classification in lesion categorization, with diagnosis validated by a clinical composite reference standard based on patients’ outcomes over the subsequent 12 months. Each liver lesion was manually contoured and labelled with diagnostic ground-truth. After data augmentation and lesion segmentation over all CT phases, we applied different architectural networks for deep machine learning (NVIDIA Tesla V100 GPUs, Dell Technologies, Singapore), including a densely-CNN consisting of five dense blocks, each with 64 con-volutional layers, followed by a fully-connected layer leading to a 186-layer network, with softmax as the activation function for classification.

Results

As of April 2020, we have retrieved and contoured 1,334 scans with 2,631 liver lesions. The cohort's mean age was 59.3±13.6 years, 63.3% male. Mean lesion size was 36.6±44.5 mm, with 871 lesions (33.1%) confirmed as HCC. Heat maps were generated and superimposed on original images based on the model's prediction of likelihood of lesion location. Such heat map hot-zones appropriately covered the location of lesions in CT images, allowing the detection of existing lesions without further annotated contours. An interim analysis of 1,693 lesions, divided in a 7:3

ratio into training and testing sets, found the densely-CNN to achieve a diagnostic accuracy of 97.0%, negative predictive value (NPV) 99.0%, positive predictive value (PPV) 92.4%, sensitivity 97.3% and specificity 96.9% in the binary classification of HCC. This was compared to the diagnostic accuracy of 86.2%, NPV 86.0%, PPV 86.5%, sensitivity 84.6% and specificity 87.8% via LI-RADS. Over 115 million radiological parameters were optimized in the development of our deep learning model.

Conclusion

Our deep learning model achieved a high diagnostic accuracy based on an end-to-end densely-CNN in detecting and classifying HCC vs. non-HCC. Further confirmatory analyses will be performed in an external validation group with >1,000 scan images. Artificial intelligence has potential in improving the diagnostic capabilities of HCC and prevent delays in treatment.

References

Supported by the Innovation and Technology Fund, the Government of the HKSAR (ITS/122/18FP); and United Ally Research Limited, a subsidiary of Hong Kong Sanatorium and Hospital Limited.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.61

OP065 Checkpoint Inhibitor Induced Hepatitis and The Relation with Liver Metastasis and Outcome in Metastasized Melanoma Patients in A Real-Life Cohort: Results From The Dutch Melanoma Treatment Registry

M Biewenga ^1,^✉, MK van der Kooij ², MW Wouters ³, MJB Aarts ⁴, FWPJ van den Berkmortel ⁵, JWB de Groot ⁶, MJ Boers-Sonderen ⁷, GAP Hospers ⁸, D Piersma ⁹, RS van Rijn ¹⁰, KPM Suijkerbuijk ¹¹, AJ ten Tije ¹², AAM van der Veldt ¹³, G Vreugdenhil ¹⁴, J Haanen ¹⁵, AJM van den Eertwegh ¹⁶, B van Hoek ¹, HW Kapiteijn ²

Introduction

Checkpoint inhibitor induced hepatitis is an immune related adverse event of programmed cell death protein 1 (PD-1) inhibition, cytotoxic T-lymphocyte associated 4 (CTLA-4) inhibition or the combination of both.

Aims & Methods

Aim of this study was to assess whether checkpoint inhibitor induced hepatitis is related to liver metastasis and outcome in a real-life nation-wide cohort. Data from the prospective nationwide Dutch Melanoma Treatment Registry (DMTR) was used to analyse incidence, risk factors, current treatment of checkpoint inhibitor induced grade 3-4 hepatitis and outcome.

Results

2585 metastatic cutaneous melanoma patients were treated with checkpoint inhibitors between May 2012 and January 2019. Severe hepatitis (grade 3/4) occurred in 30/1697 (1.8%) patients treated with PD-1 inhibitors, in 29/1108 (2.6%) patients treated with CTLA-4 inhibition and in 80/402 (20.5%) patients treated with combination therapy. Patients with hepatitis had a similar prevalence of liver metastasis compared to patients without hepatitis (42% vs 29%; p=0.155 for CTLA-4 inhibitors; 32% vs 27%; p=0.583 for PD-1 inhibitors; 38% vs 43%; p=0.497 for combination therapy). Treatment with corticosteroids was less often started in patients with PD-1 induced hepatitis compared to CTLA-4 and combination therapy induced hepatitis (77% vs 83% vs 95%; p=0.006). Second line immunosuppressive therapy was given in 7% of patients with PD-1 induced hepatitis and in 21% of patients with CTLA-4 or combination therapy induced hepatitis (p=0.066). There was no difference in median progression free and overall survival between patients with and without hepatitis (6.0 months vs 5.4 months progression free survival; p=0.608; 17.0 and 16.2 months overall survival; p=0.442).

Conclusion

Incidence of hepatitis in a real-life cohort is 1.8% for PD-1 inhibitor treatment, 2.6% for CTLA-4 inhibitor treatment and 20.5% for combination therapy. Corticosteroid treatment and second-line treatment is less given in PD-1 induced hepatitis. Checkpoint inhibitor induced hepatitis had no relation with liver metastasis and had no negative effect on outcome.

Disclosure

A.J.M. van den Eertwegh has consultancy/advisory relationships with Bristol-Myers Squibb, MSD Oncology, Amgen, Roche, Novartis, Sanofi, Pfizer, Ipsen, Merck, Pierre Fabre and received study grants not related to this abstract from Sanofi, Roche, Bristol-Myers Squibb, TEVA, Idera H.W. Kapiteijn has consultancy/advisory relationships with Bristol-Myers Squibb, Novartis, Merck, Pierre Fabre and received research grants not related to this abstract from Bristol-Myers Squibb.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.62

OP066 Characterization of Hereditary Gastric Cancer: Preliminary Analysis of The First National Multicentre Study

A Pocurull Aparicio ^1,^✉, A Sánchez ¹, M Diaz Centeno ¹, L Aguilera ², A Lopez ², I Salces ³, G Llort ⁴, C Yagüe ⁴, CG Hernández Mesa ⁵, M Carrillo ⁵, L Bujanda Fernández de Piérola ⁶, M Herrainz ⁷, O Prat ⁷, A Suárez González ⁸, E Barreiro Alonso ⁸, O Murcia Pomares ⁹, MD Pico Sala ¹⁰, V Piñol ¹¹, L Peries ¹¹, P Diez Redondo ¹², J Cubiella ¹³, C Alvarez Urturi ¹⁴, D Rodríguez-Alcalde ¹⁵, J Reyes ¹⁶, F Martinez ¹⁷, C Poves ¹⁸, A Guerra ¹⁹, J Llach ¹, C Herrera-Pariente ¹, R Moreira De Abreu ¹, T Ocaña ¹, L Moreno ¹, L Rivero Sánchez ¹, M Pellisé Urquiza ¹, F Balaguer ¹, S Carballal ¹, L Moreira Ruiz ¹

Introduction

Familial aggregation can be observed in approximately 10% of patients with gastric cancer (GC) and 5% of all GCs occurs in context of a germline mutation (Hereditary Gastric Cancer; HGC). The clinical features and preventive strategies in individuals with HGC had been poorly studied.

Aim

To describe the clinical features, preventive strategies and findings of the screening in individuals with HGC, as well as estimate the risk of GC in this context.

Methods

A multicentric nationwide study with a retrospective inclusion of individuals diagnosed with HGC syndromes (germline mutation identified). Personal clinical data, oncological personal and family history were collected. Screening strategies performed and their results were registered based on medical reports.

Results

We included 817 patients (572 families) in nineteen Spanish centers: 482 (59%) were females and the median age was 52 (range 40-63) years old.

Ninety-nine (17.3%) families had history of GC and 44 (5.3%) individuals [corresponding to 32 (5,6%) families] presented personal history of GC (Table 1).

[Description of familial aggregation and personal features of patients with gastric cancer depens on hereditary syndrome]

Hereditary syndrome (Number of patients/ families)	Familial aggregation of gastric cancer (99 families)	Personal history of gastric cancer (44 patients/32 families)	Screening of gastric cancer (515 patients)
Familial adenomatous polyposis (173 patients/ 121 families)	8 (6,6%) families	1 (0,5%) patients/ 1 (0,8%) families	151 (87,2%) patients
MUTYH associated polyposis (53 patients/ 45 families)	3 (6,6%) families	1 (1,8%) patients/ 1 (2,2%) families	37 (69,8%) patients
Juvenile Polyposis (9 patients/8 families)	1 (12,5%) families	0	7 (77,7%) patients
Cowden syndrome (22patients/17 families)	2 (11,7%) families	0	11 (50%) patients
Hereditary breast and ovarian cancer syndrome (92 patients/72 families)	11 (15,2%) families	1 (1%) patients/ 1 (1,3%) families	9 (9,7%) patients
Peutz-Jeghers syndrome (23 patients/22 families)	3 (13,6%) families	0	18 (78,2%) patients
ATM heterocigosis associated syndrome (49 patients/25 families)	11 (44%) families	4 (8,1%) patients/4 (16%) families	10 (20,4%) patients
Hereditary diffuse gastric cancer (54 patients/19 families)	18 (94,7%) families	23 (42,5%) patients/11 (57,8%) families	50 (92,5%) patients
Lynch syndrome (340 patients/241 families)	42 (17,4%) families	14 (4,1%) patients/14 (5,8%) families	222 (65,3%) patients

Open in a new tab

GC screening was performed in 515 (63%) individuals: 504 (97.8%) cases thorough gastroscopy [61% associated with Helicobacter pylori (HP) analysis] and 11 (2.2%) by HP determination without upper endoscopy. Median age at first screening procedure was 43 years old (range 30-55) with a median number of gastroscopies of three (range 1-5) and a median follow-up period or 5 years (range 3-7).

During endoscopic screening, six (1.2%) patients were diagnosed with CG (3 patients with CDH1 mutation and two patients with Lynch syndrome). Twenty-two prophylactic gastrectomies were performed in patients with CDH1 mutation, where GC or dysplasia was observed in 77% of the cases, all of them with previous normal gastroscopy.

Conclusion

This study represents the largest cohort of patients with HGC in our country.

Up to 15% families of HGC presented familial aggregation of GC and we detected a marked heterogenicity in the preventive strategies used. Six (1.2%) individuals of HGC under endoscopic screening were diagnosed with GC. Within CDH1 carriers gastroscopy failed to detect dysplasia or GC in 77% of the cases undergoing prophylactic gastrectomy. Final results of this study will help us to better define features, and risk of developing GC in these families, in order to optimize preventive strategies in GC high risk patients.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.63

OP068 Tracing The Genetic and Epigenetic History of Intestinal Metaplasia in The Chronically Inflamed Stomach To Develop A Model of Pre-Tumour Progression

W Waddingham ^1,^✉, W Cross ¹, S Ecker ², M Williams ³, P Barmpoutis ¹, O Shutkever ¹, D Graham ⁴, M Banks ⁴, TA Graham ³, S Beck ², S Sekine ⁵, M Jansen ⁶

Introduction

Up to half of the world's population is infected with Helicobacter pylori. Chronic inflammation is associated with an increased risk of gastric cancer. The main precursor to gastric adenocarcinoma is intestinal metaplasia (IM), an often multifocal condition occurring in the context of chronically inflamed atrophic gastric mucosa. The clonal origin of IM and its neoplastic potential remains poorly understood.

Aims & Methods

We performed a multi-region tissue sampling strategy to study pre-tumour progression at clonal resolution. Investigating the clonal history, population dynamics, mutational patterns and epigenetic changes of IM in the chronically inflamed stomach. Serial tissue sectioning and immunohistochemistry (CDX2 expression) of snap frozen, en face gastric epithelium was performed to quantify and microdissect multiple IM clones from individuals that had progressed to gastric cancer (6 patients, 4 - 10 IM patches per patient). Paired whole exome sequencing (WXS) and methylation analysis (Infinium EPIC array 850k) was then undertaken. Finally, we performed 3D reconstruction to visualise the clonal expansion of emerging intestinal metaplasia in the Helicobacter stomach.

Results

We perform 3D reconstruction to show that patches of IM arise from a single gastric stem cell that undergoes metaplastic transformation. Expansion of intestinal lineages within niches then follows neutral drift dynamics. WXS analysis showed a heterogenous mutation burden (between 0.5 and 8.0 mut/Mb) and a lack of shared driver mutations between IM clones. IM patches showed higher global methylation variability than gastric controls, as well as significantly increased methylation variability in a number of key genes that have a role in regulating stem cell function and intestinal differentiation (including CDX2, HNF4a, RSPO3). We describe the mutational patterns and signatures in IM patches, these were predominantly C-> T single base substitutions. We create patient-specific phylogenetic trees, reconstructing the star-shaped adaptive radiation seen in response to chronic Helicobacter infection. We then study sub-clonal mutations in a subset of IM samples to develop a model of mutation accumulation over time and space.

Conclusion

Here we trace at high resolution, the pre-tumour evolution of intestinal metaplasia in the chronically inflamed stomach, and by combining spatial clone size data with genetic sequencing data we study the relationship between IM clone size and mutation accumulation, and develop a new spatiotemporal model of pre-tumour evolution occurring within expanding IM clones.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.64

OP069 Gastric Non-Helicobacter Commensal Bacteria and Their Products, Short-Chain Fatty Acids, Could Enhance Cd44V9-Positive Gastric Cancer Stem-Like Cell Development in H. Pylori-Infection

H Suzuki ^1,^✉, H Tsugawa ²

Introduction

Cancer stem cell marker CD44v9-positive cancer stem-like cells are strongly associated with the recurrence and drug resistance of gastric cancer (Br. J. Cancer 109:379, 2013, Anticancer Res. 38: 6163, 2018). Our recent study showed that overexpression of the capping ac-tin protein of muscle Z-line alpha subunit 1 (CAPZA1) inhibits autophagy by inhibiting LAMP1 expression, causing accumulation of Helicobacter pylori-derived CagA oncoprotein in CAPZA1-overexpressing gastric epithelial cells (Cell Host Microbe 12:764, 2012; Autophagy 15:242, 2019). Moreover, the expression of CD44v9 is evoked in CAPZA1-overexpressing cells following CagA accumulation (Cell Mol. Gastroenterol. Hepatol. 8:319, 2019).

These findings indicate that deregulation of CAPZA1 expression leads to an increased risk of gastric carcinogenesis through the development of CD44v9-positive cancer stem-like cells.

We previously showed that CAPZA1 expression is regulated by histone acetylation and is increased by histone deacetylase (HDAC) inhibitors (Autophagy 2019). Short chain fatty acids (SCFAs) derived from commensal bacteria are known to inhibit HDAC (J. Biol. Chem. 253:3364, 1978). The present study examined whether SCFAs derived from gastric non-Helico-bacter commensal bacteria increase the development of CD44v9-positive cancer stem-like cells by inducing CAPZA1-overexpressing cells in H. pylori-infected gastric mucosa.

Aims & Methods

Specific pathogen-free (SPF)-mice were inoculated with H. pylori TN2GF4. Gastric polarized epithelial monolayers (mucosoids) were constructed from gastric organoids based on a previous report (Gut 68:400, 2019). Mucosoids were infected with H. pylori G27 or H. pylori G27 AcagPAI. SCFAs in the gastric mucosa were quantified by liquid chroma-tography-mass spectrometry (LC-MS). A sequencing library of the 16Sr-RNA gene V3-V4 regions was established by extracting and sequencing DNA from the gastric contents.

Results

SCFAs such as propionate and butyrate enhanced CAPZA1 expression in AGS cells in a concentration -dependent manner by enhancing histone acetylation of the CAPZA1 promoter. CD44v9 expression in AGS cells was induced by H. pylori G27 infection under SCFA treatment but not by H. pylori G27 AcagPAI. in H. pylori G27-infected mucosoids, CagA accumulation was caused by treatment with SCFAs, and the development of CD44v9-positive cells was detected. On the other hand, CD44v9-positive cells in H. pylori G27 AcagPAI-infected mucosoids were scarcely detected after treatment with SCFAs. Although localization of butyrate in the gastric mouse mucosa was detected by LC-MS, this localization completely disappeared upon antibiotic administration, suggesting a possible supply of butyrate to the gastric mucosa from commensal bacteria. The expression of CAPZA1 was enhanced by external administration of propionate or butyrate to antibiotic-treated mice.

Among the butyrate-producing gastric commensal bacteria, Eisenbergi-ella tayi and Lachnoanaerobaculum saburreum were presumed to contribute to CAPZA1 induction in murine gastric epithelial cells. Interestingly, these bacterial families are significantly increased in patients with gastric cancer (Sci. Rep. 4:4202, 2014). The 16S rRNAs of E. tayi and L. saburreum were detected in human gastric juice.

Conclusion

Butyrate-releasing gastric commensal bacteria could influence the risk of gastric cancer through promotion of the development of CAPZA1-overexpressing epithelial cells in the gastric mucosa with H. pylori infection.

Disclosure

The author HS received scholarshipfunds for the research from Otsuka Pharmaceutical Co, Ltd, and received service honoraria from Astellas Pharm, Astrazeneca K.K., Daiichi-Sankyo Co., EA Phama Co, Ltd, Mylan EPD, Otsuka Pham, Takeda Pharm, Tsumura Co.

References

Tsugawa Suzuki et al. Cell. Mol Gastroenterol Hepatol. 8(3): 319–334, 2019; [DOI] [PMC free article] [PubMed] [Google Scholar]
Tsugawa Suzuki et al. Autophagy 15(2): 242–258, 2019; [DOI] [PMC free article] [PubMed] [Google Scholar]
Tsugawa Suzuki et al. Cell Host Microbe 12(6): 764–777, 2012. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.65

OP070 Stromal R-Spondin 3 Is Essential For Chief Cell Recovery Upon Injury, Allowing Their Reprogramming Into Premalignant Lesions - A Double Edged Sword

A-S Fischer ^1,^2,^3,^✉, S Müllerke ², M Sigal ^1,^2,³

Introduction

The gastrointestinal epithelium is surrounded by stromal myofibroblasts that produce and secrete multiple signaling molecules. One subset of myofibroblasts produces R-spondin 3 which reaches the epithelial cells in the base of the epithelial glands and promotes Wnt signaling in this compartment. Wnt signaling is essential for proliferation and self-renewal of stem cells throughout the gastrointestinal tract, and deregulation is linked to malignant tissue transformation. in the gastric corpus, the gland base is characterized by the presence of chief cells, that do not act as bona fide stem cells but have the potential to be recruited to the stem cell pool upon injury (1,2). While these cells express Wnt/R-spondin target genes, the function of this signaling pathway in the corpus is not well understood.

Aims & Methods

We asked whether and how R-spondin 3 affects the gland homeostasis in the corpus glands in the healthy state and in the context of acute and chronic injury.

Using conditional mouse models that enable manipulation of R-spondin 3 gene expression in gastric myofibroblasts together with histology, single molecule RNA in situ hybridization, immunofluorescence labelling, microarray analysis and qPCR, we investigated the effect of R-spondin 3 in corpus glands. To study its role in chief cell recovery, we applied a mouse model to simultaneously deplete Lgr5-expressing chief cells and stromal R-spondin 3.

Furthermore, we used models of acute tamoxifen-induced and chronic H. pylori-driven corpus injury to explore the relevance of R-spondin 3 for tissue integrity and its role for recovery from metaplasia.

Results

We find that stromal R-spondin 3 is essential for chief cell identity and differentiation. Loss of R-spondin 3 expression leads to a loss of gland base chief cells, whereas overexpression results in an expansion of the chief cell compartment. Upon depletion of Lgr5+ gland base cells, recovery of these cells requires endogenous stromal R-spondin 3. We demonstrate that differentiated chief cells are lost in the context of acute tamoxifen-driven injury, which triggers an increased expression of stromal R-spondin 3.

This in turn results in epithelial reprogramming leading to a recovery of the gland base. Accordingly, overexpression of R-spondin 3 in stromal cells prevents from tamoxifen-induced injury by stabilizing the chief cell signature.

Chronic infection with H. pylori also induces chief cells loss in the gland base in mice lacking R-spondin 3 and in wildtype mice. in mice that over-express R-spondin 3, chronic H. pylori-driven inflammation reprograms the accumulating chief cells leading to highly proliferative adenomatous lesions in the gland bases of infected R-spondin 3 knockin mice.

Conclusion

Together, our data demonstrate that R-spondin 3 represents a critical endogenous driver of chief cell differentiation in the corpus. This signal is crucial for the maintenance of these cells and their recovery upon short-term injury. However, in the context of chronic inflammation, chief cells that accumulate in a R-spondin 3-dependent manner can be further reprogrammed into a proliferative state, leading to development of pre-malignant lesions.

Disclosure

Nothing to disclose

References

1.Leushacke M., Tan S.H., Wong A. et al. Lgr5-expressing chief cells drive epithelial regeneration and cancer in the oxyntic stomach. Nat Cell Biol 2017; 19: 774–786. [DOI] [PubMed] [Google Scholar]
2.Stange D.E., Koo B.K., Huch M. et al. Differentiated Troy+ chief cells act as reserve stem cells to generate all lineages of the stomach epithelium. Cell 2013; 155: 357–68. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.66

OP071 Mccune-Albright Syndrome and Ipmn Formation Modelled in Gnas-Mutated Duct-Like Organoids Derived From Human Pluripotent Stem Cells

M Hohwieler ^1,^✉, J Merkle ¹, M Breunig ¹, MK Melzer ^1,², PC Hermann ¹, L Perkhofer ¹, S Heller ¹, M Müller ¹, T Seufferlein ¹, M Wagner ¹, M Meier ³, A Kleger ¹

Introduction

One of the early key drivers frequently mutated in intraductal papillary mucinous neoplasia (IPMN) of the pancreas is the GNAS gene with a hotspot mutation at codon 201. McCune-Albright syndrome (MAS) is caused by such a dominant activating GNAS mutation occurring in a post-zygotic mosaic pattern. Besides other manifestations, including characteristic polycystic fibrous dysplasia, patients show an increased risk of developing pancreatic cysts and PDAC.

Aims & Methods

We aimed to develop an appropriate MAS model using human pluripotent stem cells (hPSCs) to investigate the molecular basis by which GNAS mutations cause IPMN formation. Here, we have implemented a novel in vitro differentiation platform to efficiently induce pancreatic duct-like organoids (PDLOs) from hPSCs. MAS-patient-derived cells were reprogrammed to generate induced pluripotent stem cell (iPSC) lines only differing in the GNAS mutation status. Second, using a piggyBac system, hESCs were engineered to inducibly express GNAS^R201H and KRAS^G12D.

Results

Upon ductal commitment, GNAS^WT/R201C MAS-iPSCs formed large cystic PDLOs dramatically increasing in size compared to PDLOs from isogenic GNAS^WT/WT lines. Besides enhanced proliferation, we identified the GNAS-PKA-VASP signalling axis to be important for cystic growth of GNAS^WT/R201C PDLOs. Overexpression of GNAS^R201H within hESC-derived PDLOs faithfully recapitulated MAS patient data evident by pronounced cystic growth in vitro. After transplantation into the mouse pancreas, mutant GNAS expressing PDLOs developed well-differentiated cystic ducts resembling human IPMNs. Additive KRAS^G12D induction led to far larger IPMN-like structures with a higher degree of cellular atypia, more pseudo-papillae, and less proliferation.

Conclusion

These results indicate that we were able to recapitulate IPMN features in hPSC-derived organoids in vitro and in vivo. Together, our approach provides analytical access to early stages of disease development and allows to determine the effect of specific cancer driver mutations on duct-like cells with the genetic background of a cancer predisposition syndrome. This will offer new possibilities for studying tumour initiation and progression in a genetically defined humanized disease model as perquisite for personalized therapies.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.67

OP072 Parp Inhibitor Resistance Induces Massive Genome Alterations Responsible of The Acquisition of Multidrug Resistance in Dna Damage Repair-Deficient Pancreatic Cancer

J Gout ^1,^✉, L Perkhofer ¹, M Morawe ¹, F Arnold ¹, E Roger ¹, M Müller ¹, T Seufferlein ¹, P-O Frappart ², A Kleger ¹

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is the fourth leading cause of cancer-associated death in the Western World. PDAC bears a high accumulation of oncogenic mutations. Recent sequencing studies found ATM serine/threonine kinase (ATM) as the most frequently mutated DNA damage response gene, involved in homologous recombination (HR), in PDAC. Our lab previously showed that the loss of ATM accelerates EMT and promotes genomic instability (1,2). in line with the altered genomic integrity of ATM-mutated PDAC, we demonstrated the efficacy of genotype specific strategies like PARP- and ATR- inhibition. Moreover, synergy was found upon inhibition of PARP, ATR, and DNA-PK leading to synthetic lethality in ATM-deficient murine and human PDAC (Gout, Perkhofer et al,in revision; see second abstract by Gout, Perkhofer et al). Unfortunately, long-term maintenance monotherapy with PARP inhibitor-might promote resistance acquisition as already shown in non-pancreatic tumors (2,3).

Aims & Methods

We decided to explore PARP-inhibitor resistance within the ATM-null background in primary murine PDAC cells. To elucidate the mechanisms of olaparib resistance in ATM-deficient PDAC, several PARPi-resistant AKC (Atm^fl/fl; LSL-Kras^G12D/>;Ptf1a^Cre/>) cell lines were generated, molecularly characterized, and submitted to RNA- and whole exome-se-quencing (WES).

Results

Resistance was confirmed by viability and clonogenicity assays, and persisted even after olaparib removal. PARPi-resistant AKC cells exhibited impaired proliferation, aneuploidy, and increased genomic instability. These observations together with persisting PARPi resistance beyond drug release suggest permanent genetic and not only temporary alterations. Thus, we performed a WES in a set of parental and PARPi resistant-AKC cells. Interestingly, it revealed more indels but even more pronounced a higher number of single-nucleotide variants as well as copy number alterations in the resistant ATM-deficient counterparts. Intrigu-ingly, we observed a structural aberration pattern with consistent amplification of a particular locus described as the Abcb1 amplicon (4) in PARPi resistant-AKC. Numerous genes are located at this locus (5 A1; 5 3.43 cM region in mouse; 7q21.12 region in humans) e.g., Abcb1, Abcb4, Sri, Dbf4, all previously associated with tumorigenesis and multidrug resistance (MDR) in various cancer (4). RNA-sequencing followed by qPCR validation confirmed that structural amplification patterns in PARPiresistant-AKC cells directly translate into a transcriptional regulation of upregulated MDR genes including Abcb1(Mdr1) and Abcg2(Brcp).Whole exome sequencing also provided evidence of clonal evolution during acquired resistance, suggesting that massive genomic instability in the context of homologous recombination-deficiency can lead to massive genetic rearrangements favoring PDAC drug resistance and aggressivity.

Conclusion

Our data suggest that PARPimonotherapy needs to be monitored with caution, particularly in case of complete remission as instead of preventing relapse it can trigger a general drug resistance inevitably rendering subsequent strategies inefficient.

Disclosure

Nothing to disclose

References

1 Russell R., Perkhofer L. et al. Loss of ATM accelerates pancreatic cancer formation and epithelial-mesenchymal transition. Nat Commununications 2015; 6: 7677. [DOI] [PMC free article] [PubMed] [Google Scholar]
2 Perkhofer L., Schmitt A. et al. ATM Deficiency Generating Genomic Instability Sensitizes Pancreatic Ductal Ad-enocarcinoma Cells to Therapy-Induced DNA Damage. Cancer research 2017; 77: 5576–90. [DOI] [PubMed] [Google Scholar]
3 Bouwman P., Jonkers J. Molecular pathways: how can BRCA-mutated tumors become resistant to PARP inhibitors? Clin Cancer Res 2014; 20: 540–7. [DOI] [PubMed] [Google Scholar]
4 Genovese I., Ilari A. et al. Not only P-glycoprotein: Amplification of the ABCB1-containing chromosome region 7q21 confers multidrug resistance upon cancer cells by coordinated over expression of an assortment of resistance-proteins. Drug Resist Update 2017; 32: 23–46 [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.68

OP073 Exploring Roles of Atm-Deletion in Pancreatic Cancer Vulnerabilities and Cancer-Associated Fibroblast/Tumor Cell Crosstalk

J Gout ^1,^✉, E Roger ¹, F Arnold ¹, M Müller ¹, L Perkhofer ¹, T Seufferlein ¹, A Kleger ¹

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is the fourth leading cause of cancer-associated death in the Western World. Recent sequencing studies found ATM serine/threonine kinase (ATM) frequently mutated in PDAC. Our laboratory previously showed that the loss of ATM accelerates EMT and promotes genomic instability in an oncogenic KRAS-^G12Dcontext (Atm^fl/fl; LSL-Kras^G12D/+; Ptf1a^Cre/+, AKC) (Russell, Nat Commun, 2013; Perkhofer, 2017, Cancer Res). We could identify a new targeted and synergistic route to reach highest efficiency with lowest toxicity by allowing synthetic lethality in ATM-deficient PDAC patients (triple PARP/ ATR/DNA-PKcs inhibition; Gout, Perkhofer, in revision). Surprisingly, a stabilization of P53 was observed in AKC tumors suggesting a response to genomic instability.

Aims & Methods

We postulated that this P53 activation could slow down the pancreatic tumorigenesis. Because some human PDAC displaying both TP53 and ATM mutations represent a PDAC subtype that could be more resistant to DNA damage-inducing treatment, we studied the effect of P53 inactivation in this Atm-deleted context. For that, we generated an Atm^fl/fl; Trp53^fl/“; LSL-Kras^G12D/+; Ptf1a^Cre/+(APKC) mouse model displaying a shortened survival.

Results

A drug screening showed that responsiveness of AKPC cells to DNA damage-inducing drugs is reduced. As well, the efficiency of our PAD tritherapy was dramatically decreased when applied on APKC cells, proving that PAD-induced apoptosis in AKC cells was at least partially mediated by P53 and highlighting that TP53 status should be carefully verified before considering a treatment with DNA damaging agents. Nevertheless, APKC cells showed a higher sensitivity to DNA damage drugs when compared to PKC cells, showing that a certain P53-independent workable vulnerability is created by Atm deletion.

In parallel, systematic characterization of pancreatic tumors and isolated primary tumor cells revealed the more mesenchymal phenotype of APKC cells. Very interestingly, cancer-associated fibroblasts (CAFs) content of APKC tumors also differs when compared to other tumors as shown by the higher proportion of inflammatory CAFs (iCAFs) present in APKC tumors.

This suggests a putative role of this CAF subpopulation in the development of the aggressive ATM^KO/P53^KOphenotype. Trying to decipher the role of ATM deficiency in tumor phenotype and in turn, in CAF/tumor crosstalk, we first performed a protein array on tumor cell conditioned medium which found that ATM-deficient PKC (APKC) cells effectively secrete more inflammatory factors as interleukin-6, in comparison to ATM-proficient PKC cells.

Conclusion

To summarize, preliminary but coherent in vivo and in vitro data provided first evidence that tumor ATM status could play a role in switching CAF phenotype which in turn could dramatically promote tumor aggressiveness.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.69

OP074 Patient-Derived Pancreatic Cancer Organoids As A Personalized Preclinical Tool For Drug Response Prediction

AK Beutel ^1,^✉, J Gout ¹, J Scheible ¹, R Koehntop ¹, E Roger ¹, L Perkhofer ¹, L Schulte ¹, MM Haenle ¹, M Müller ¹, TJ Ettrich ¹, T Seufferlein ¹, A Kleger ¹

Introduction

Given the dismal prognosis of pancreatic ductal adenocarci-noma (PDAC) there is an urgent need for personalized therapeutic strategies in order to improve survival. A promising approach is a culture set-up of organoids generated from tumor tissue that enables specific drug testing and the development of individualized treatment regimens. Organ-oids are three-dimensional cultures derived from tissue specific stem cells with the capacity to self-organize into structures that recapitulate features of tissue architecture and function found in vivo.

Aims & Methods

We recently demonstrated that patient-derived xenograft tumor (PDX) and corresponding PDX-derived organoids show high concordance on a morphological, immunohistochemical and deep molecular profiling level concluding that patient-derived organoids (PDOs) may represent a feasible approach to predict drug response. Tissue samples for generation of PDOs were acquired by means of ultrasound-guided biopsy of the primary tumor and its metastases, EUS (endoscopic ultrasound)-guided fine needle biopsy and surgical resection. To evaluate drug response of standard-of-care chemotherapeutics, a viability assay was used to calculate the cell death ratio (dead cells upon drug treatment/ dead cells upon vehicle treatment) reflecting the cytotoxic effect. Repetition of standardized drug screens confirmed the robustness of our method. Drug-dose response tests with ten different dosages covering two orders of magnitude allowed us to calculate the AUC (area under the curve) and to cluster in vitro response for each drug.

Results

PDOs from more than 30 treatment-naïve and pretreated PDAC patients were successfully isolated and expanded with a satisfying PDO generation efficiency above 75%. PDOs exhibited heterogeneous responses to standard-of-care substances and PDO drug prediction profiles paralleled the observed patient response.

Conclusion

Thus, we present a feasible preclinical tool for drug response prediction to pave the way towards an organoid-based precision medicine in clinical routine.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.70

OP075 Compared To Duodenal Carcinomas Mismatch Repair Deficiency Is Rare in Early-Onset Pancreatic and Ampullary Cancer But A Good Indicator For Lynch Syndrome

V Kryklyva ^1,^✉, MAJ Marijnissen-van Zanten ¹, MJE Kempers ², LAA Brosens ^1,³, MJL Ligtenberg ^1,², ID Nagtegaal ¹

Introduction

Limited information is available regarding pancreatic cancer (PC) occurring at a young age. Up to 10% of all PCs develop due to familial basis of disease or within hereditary syndromes, such as, Lynch syndrome (LS). LS is a major cancer syndrome and the most common cause of hereditary colorectal cancer. LS is caused by inactivating germline variants in DNA mismatch repair (MMR) genes (MLH1, PMS2, MSH2 or MSH6) or by germline deletions of the 3’ end of the EPCAM gene leading to the epigenetic silencing of MSH2. in EPCAM-associated LS patients an increased number of pancreatic and duodenal carcinomas (PCs and DCs, respectively) was observed. Therefore, we hypothesized that early-onset PCs and DCs might be associated with EPCAM-related LS.

Aims & Methods

We performed a retrospective nationwide search in the Nationwide network and registry of histo- and cytopathology in the Netherlands (PALGA), to identify all patients < 50 years of age diagnosed with PC or DC between January 2002 and December 2012. A substantial part of PCs were reclassified as ampullary carcinomas (ACs) and assigned to a separate group.

Tissues were requested and immunohistochemical analysis was performed to evaluate the presence of MMR and EPCAM proteins. Microsatellite instability (MSI) was assessed using a panel of five mononucleo-tide markers. in doubtful or complex cases, additional MSI analysis was performed with an extensive smMIP-based Next-Generation Sequencing (NGS) panel of 55 sensitive MSI markers. Methylation-specific multiplex ligation-dependent probe amplification (MS-MLPA) was performed to assess MLH1 promoter hypermethylation. Tumour and normal DNA from the patients with MMR deficiency (MMRd) was sequenced for MLH1, MSH2, and MSH6 genes to determine the presence of pathogenic germline variants (i.e. LS). MLPA was used to detect large genomic aberrations in MLH1 gene. Findings in the cohort of early-onset DC patients were compared with the internal control cohort of late-onset (above the age of 50) DCs.

Results

In total, 162 applicable patients were identified through the PALGA search. After the application of exclusion criteria and based on tissue availability 91 patients (PC, n=46; DC, n=22; AC, n=23) were included. Deficient MMR protein expression and/or high level of MSI (MMRd/MSI-H) was observed in 3/46 (6.5%) PCs, 10/22 (45.5%) DCs, and 1/23 (4.3%) ACs. Six MMRd/MSI-H patients carried pathogenic germline variants in MSH6 (PC, n=2); MSH2 (DC, n=1), MLH1 (DC, n=1; AC, n=1), and PMS2 (DC, n=1). Moreover, in a number of tumours somatic inactivation of MMR genes MLH1 (DC, n=2) and MSH2 (DC, n=1) was observed. in the control cohort of late-onset DCs only 1/18 (5.6%) was MMRd/MSI-H, and this DC had a somatic inactivation of MLH1.

Conclusion

Our findings suggest that MMRd/MSI-H is a rare feature in early-onset PC and AC in comparison to DC, but when present is highly suggestive of LS. MMRd/MSI-H is rarely observed in late-onset DC.

Disclosure

Presented at the ENABLE, 2nd European PhD and postdoc symposium, November 7-9, 2018 in Copenhagen, Denmark. Further nothing to disclose.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.71

OP076 Emerging Incidence Trends of Eosinophilic Esophagitis Over 25-Years: Results of A Nationwide Register-Based Pathology Cohort

WE de Rooij ^1,^✉, ME Barendsen ¹, MJ Warners ^1,², BD van Rhijn ^1,³, AH Bruggink ⁴, J Verheij ⁵, AJ Bredenoord ¹

Introduction

Eosinophilic esophagitis (EoE) has emerged from a case-reportable illness in the early 1990s to a distinct clinicopathological entity. Increasing worldwide incidences have been observed, although, due to various study designs and heterogeneous reporting, estimates are inconsistent.

Aims & Methods

To determine population-based annual incidence rates over a time period of 25 years. A nationwide register-based pathology (PALGA) search was performed to identify reports describing esophageal eosinophilia between 1995 and 2019. EoE was identified if the diagnosis was confirmed by the pathologist and/or the presence of > 15 eosinophils per high power field (hpf) was reported, with exclusion of other causes of esophageal eosinophilia (e.g. gastroesophageal reflux disease) based on additional information on the indication of esophagogastroduodenos-copy (EGD). Crude incidence rates were estimated by the number of new EoE cases per year and matched with population data.

Results

Between 1995 and 2019, 7361 unique patients’ reports mentioned esophageal eosinophilia and were further screened, of which 4061 were classified as EoE (71% male, mean age 37.9±18.4 years). in total, 639 (16%) children (< 18 years) and 3422 (84%) adults were diagnosed. The majority of cases were diagnosed between the ages of 20 and 49 years. The number of annual new EoE diagnoses increased from 0.006 in 1995 (95% Confidence Interval (CI) 0.002 - 0.036) to 3.16 (95% CI 2.90 - 3.44) per 100.000 inhabitants in 2019. EoE was significantly more prevalent in males (Odds Ratio (OR) 2.48 | 95% CI 2.32 - 2.65; vs. females P < 0.001) and adults (OR 1.42 | 95% CI 1.31 - 1.55; vs. children P < 0.001). Highest incidence rates were observed in 2019, being 4.37 per 100.000 in males (95% CI 3.94 - 4.84) and 1.97 per 100.000 in females (95% CI 1.68 - 2.29) (P < 0.001). No seasonal variation was observed.

Conclusion

Over the past quarter century, the annual rates of newly diagnosed EoE patients raised dramatically and this increase has not reached a deceleration yet.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.72

OP077 Efficacy of Proton-Pump Inhibitor Therapy For Eosinophilic Oesophagitis in Real-World Practice: Results From The Eoe Connect Registry

EJ Laserna-Mendieta ^1,^2,³, S Casabona-Francés ^2,⁴, D Guagnozzi ^5,⁶, EV Savarino ⁷, MA Perello ⁸, A Guardiola Arévalo ^10,⁹, J Barrio Andrés ¹¹, I Pérez-Martínez ¹², AL Krarup ¹³, J Alcedo ¹⁴, S De la Riva ¹⁵, E Rey-Iborra ¹, C Santander ^2,^6,¹⁶, Á Arias Arias ^2,¹⁷, AJ Lucendo ^2,^6,^18,^✉

Introduction

Proton-pump inhibitors (PPIs) are the most commonly used first-line therapies for patients with eosinophilic oesophagitis (EoE), a chronic immune-mediate inflammatory disorder of the oesophagus in response to a non-IgE-mediated food allergy. The ability of PPIs to reduce both symptoms and eosinophilic infiltration in patients with EoE has been repeatedly documented. However, many of the aspects related to the efficacy of PPIs in EoE are still unknown, largely because all the evidence has been provided by observational studies, generally involving small numbers of patients.

Aims & Methods

Here, we aimed to assess the effectiveness of PPI therapy for EoE in real-world clinical practice. in this cross-sectional study, we collected data on PPI efficacy from the multicentre EoE CONNECT database that included patients recruited at 13 hospitals in Spain, Italy and Denmark. Clinical remission was defined by a decrease of >50% in dysphagia symptom score; histologic remission was defined as a peak eosinophil count below 15 eosinophils per high-power field (eos/hpf). Factors associated with effectiveness of PPI therapy were identified by chi-square or Fisher's univariate tests and binary logistic regression multivariate analyses.

Results

Overall, 630 patients (76 children) received PPIs as initial therapy (n=600) or after failure to other therapies (n=30). PPI treatment achieved eosinophil density below 15 eos/hpf in 48.8% of patients and decreased symptom score >50% from baseline in 71.0% of them. PPI dose (p=0.027), treatment length (p=0.022), EoE phenotype (p=0.011) and presence of fi-brotic changes at baseline endoscopy (rings and/or strictures) (p=0.016) were identified as significantly associated to PPI effectiveness to induce clinico-histological remission of EoE in the univariate tests. in the multivariate analysis, more EoE patients with an inflammatory rather than stricturing phenotype accomplished for clinico-histological remission after PPI therapy (OR 3.7; 95% CI: 1.4-9.5); as well as those who prolonged treatment length from 56-to-70 days up to 71-to-90 days (OR 2.7; 95% CI: 1.3-5.3). No differences in effectiveness among the different PPI drugs were noted. After achieving clinico-histological remission of EoE, PPI dosage reduction was prescribed in 138 patients and resulted effective to maintain it in 69.9% of the cases. Maintenance PPI treatment trended to be less effective among patients with a stricturing phenotype and those who achieved partial histological remission (5-15 eos/hpf; instead of deep remission with less than 5 eos/hpf) after initial therapy with PPIs. in addition, a trend toward an increased clinico-histological response rate was observed for adults compared to children, for both induction and maintenance therapies with PPIs, although statistically significant differences were not reached.

Conclusion

We provide evidence that PPIs are an effective anti-inflammatory therapy that achieves histologic and clinical remission in half of patients with EoE, with around 70% of responding patients being able to maintain long term remission after dose reduction. All PPI drugs resulted similarly effective, with high doses used for between 71 and 90 days in patients without fibrotic findings at endoscopy providing the highest benefit for induction of remission. Patients with stricturing EoE were less likely to respond to PPI therapy initially and in the long term, so they should be considered candidates for alternative anti-inflammatory options.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.73

OP078 Effect of Elemental Nutrition Added To Four-Food Elimination in Adult Eosinophilic Esophagitis Patients: A Randomized Controlled Trial

WE de Rooij ^1,^✉, B Vlieg - Boerstra ^2,³, MJ Warners ^1,⁴, MTJ van Ampting ⁵, BCAM van Esch ^5,⁶, SRBM Eussen ⁵, AJ Bredenoord ¹

Introduction

Dietary therapies such as empiric elimination of common causative foods or elemental nutrition with amino acid-based formulas (AAF) are often prescribed for induction and maintenance of disease remission in eosinophilic esophagitis (EoE). However, an exclusive AAF diet is difficult to adhere to long-term, and elimination diets are not always effective. A combination of empiric elimination with AAF might improve adherence and therefore efficacy of dietary management.

Aims & Methods

To evaluate whether addition of AAF to a Four-Food-Elimination-Diet (FFED) is more effective than a standard FFED in decreasing Peak Eosinophil Counts (PEC), symptoms and endoscopic signs. This prospective randomized controlled trial enrolled 41 adult patients with active EoE, defined as having symptoms related to esophageal dysfunction and >15 eosinophils per microscopic high-power field (HPF) on baseline biopsy. Patients were randomized to either a standard FFED or FFED with addition of 30% of their daily energy needs by AAF (FFED+AAF).

Histological disease activity, clinical response (Straumann Dysphagia Instrument patient-reported outcome (SDI-PRO)), endoscopic signs (Endo-scopic reference score (EREFS)) and disease specific health-related quality of life (Adult Eosinophilic Esophagitis Quality of Life (EoE QoL-A) were measured at baseline and after 6 weeks of dietary intervention. AAF intake was evaluated by a physician and dietitian.

Results

Forty-one patients (60% male, age 34.5 (interquartile range (IQR) 29-42.8 years) were randomized to FFED (n=20) or FFED+AAF (n=21), 40 patients completed the diet. Overall compliance rate of prescribed AAF was 87%. Complete histological remission (< 15 eos/HPF) was achieved in 50% (10/20) of patients treated with FFED+AAF compared to 25% (5/20) with FFED (P=0.095). Median PEC decreased after FFED (60 (IQR) 41.3-80) - 26.5 (IQR 14-48.8); P=0.008) and after FFED+AAF (50 (IQR 42.5-100) - 17 (IQR 3.3-35.5); P=0.001). Primary endpoint analysis on change in PEC from baseline to 6 weeks signified a possible trend towards the FFED + AAF treated patients vs. the FFED group (-41.5 (IQR -50 - -23) vs. -29 (IQR -49 - -10.8); P=0.127). A significant reduction of the SDI-PRO scores was observed in both groups, FFED (5 (IQR 4-7) - 2 (IQR 0-4); P< 0.001) and FFED+AAF (5 (IQR 3.3-6) - 3 (IQR 0.3-3.8); P=0.002), respectively. Change of SDI-PRO scores did not differ between both groups (P=0.618). A significant reduction of the EREFS after FFED (4 (IQR 3.3-5) - 4 (IQR 1.3-4); P=0.008) and FFED+AAF (4.5 (IQR 3-5) - 3 (IQR 1.3-4); P=0.005) was observed. Reduction of the EREFS was similar in both groups (P=0.687). Total EoE-QoL-A score only significantly improved from baseline to week 6 in the FFED + AAF group (P=0.003). Change of the EoE-QoL-A score did not differ between both groups (P=0.128).

Conclusion

Although statistical significance was not reached (P=0.095), these results suggests that AAF added to FFED results in a higher number of patients achieving complete histological remission. A possible association between decrease in PEC and addition of AAF (P=0.127) was observed. SDI-PRO scores and EREFS showed no significant difference between both groups. A significant improvement of EoE-QoL-A score was only observed in the FFED + AAF treated patients.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.74

OP079 The Gene Expression Signature Modulated By Dupilumab Is Highly Correlated with Eosinophilic Esophagitis Histological Severity

MH Collins ^1,^✉, JD Hamilton ², WK Lim ², S Hamon ², MF Wipperman ², A Radin ², M Chehade ³, I Hirano ⁴, ES Dellon ⁵, S Harel ², W Brian ⁶, LP Mannent ⁷, ME Rothenberg ¹

Introduction

Eosinophilic esophagitis (EoE) is a chronic type 2 inflammatory disease characterized by eosinophilic inflammation and histo-logical abnormalities. Patients with EoE exhibit an altered esophageal transcriptome compared with healthy controls. Dupilumab, a fully human monoclonal antibody, blocks the shared receptor component for inter-leukin (IL)-4 and IL-13, key and central drivers of type 2 inflammation in multiple diseases. in a double-blind, placebo-controlled, phase 2 study (NCT02379052), adults with active EoE were randomized 1:1 to receive 12 weeks of subcutaneous dupilumab 300mg weekly or placebo. Dupilumab significantly improved dysphagia and histological and endoscopic measures of disease with an acceptable safety profile. The aim of this analysis was to evaluate the correlation between gene expression and histological severity.

Aims & Methods

RNA was extracted for transcriptome analysis from pinch biopsies collected from the proximal, mid, and distal esophagus at baseline and Week 12 of 41 patients. Gene expression was averaged per patient across the 3 samples at each time point. The top 50 most upregulated and top 50 most downregulated genes in EoE (that were normalized by dupilumab treatment) were used to generate a single-sample normalized enrichment score (DpxOme-EoE™ NES). Spearman correlation analysis compared individual gene expression and DpxOme-EoE™ NES with total EoE histology scoring system (EoE-HSS) score at each time point. EoE-HSS assesses the severity of eosinophil density, basal zone hyperplasia, eosinophil abscesses, eosinophil surface layering, surface epithelial alteration, dyskeratotic epithelial cells, and dilated intercellular spaces.

Results

Overall, a strong correlation was observed between the DpxOme-EoE™ NES and the EoE-HSS (correlation coefficient [ρ]=0.832; P< 0.001). The highest positive single gene correlation was seen with cathepsin C (CTSC; p=0.826; P< 0.001), a pro-inflammatory protease, and the highest negative correlation with CRISP3 (p=-0.813; P< 0.001; Table). Other highly correlated genes were related to smooth muscle contraction (e.g. ANO1, p=0.823; P< 0.001), barrier function (e.g. SPINK8, p=-0.798; P< 0.001), and neuroinflammation (e.g. NTRK1, p=0.809; P< 0.001).

Conclusion

The DpxOme-EoE™ NES score was strongly correlated with histological severity, and strong correlations were also observed for individual genes associated with tissue remodeling, barrier function, and inflammation. These data demonstrate that the genes most strongly associated with histological severity in EoE are among those normalized by dupilumab. in this study, dupilumab improved histological measures of disease in EoE patients; this analysis suggests that these improvements are, at least in part, due to the effect of dupilumab on esophageal gene expression. Further correlation analyses using the individual components of the EoE-HSS are needed to assess if histology can be used as a surrogate for abnormal gene expression.

[Table. Top 10 highest positive and negative correlations between the post-dupilumab transcriptome and EoE-HSS. P<0.001 for all assessments.]

Positive correlations		Negative correlations
Gene	Correlation ρ value	Gene	Correlation ρ value
CTSC	0.826	CRISP3	–0.813
ANO1	0.823	C1orf177	–0.807
ORAI1	0.819	SPINK8	–0.798
ATP13A5	0.815	NUCB2	–0.796
C1orf74	0.814	ZNF416	–0.794
GCNT2	0.813	AMACR	–0.793
TRPM6	0.812	TP53I3	–0.792
AP2M1	0.810	AMFR	–0.792
NTRK1	0.809	BC107108	–0.790
TRAPPC3	0.808	MUC21	–0.785

Open in a new tab

Disclosure

Collins MH: Allakos, AstraZeneca, BMS, Esocap, GSK, Regeneran Pharmaceuticals, Inc., Shire - consultant; Receptos/BMS, Regeneron Pharmaceuticals, Inc., Shire - research funding. Hamilton JD, Lim WK, Hamon S, Wipperman MF, Radin A, Harel S: Regeneron Pharmaceuticals, Inc. - employees and shareholders. Chehade M: Adare, Allakos, Nutricia, Regeneron Pharmaceuticals, Inc., Shire - consultant; Allakos, Regeneron Pharmaceuticals Inc., Shire - research funding; Medscape, Nutricia – honoraria for lectures. Hirano I: Adare, Receptos/BMS, Regeneron Pharmaceuticals, Inc., Shire - consultant; Meritage, Receptos/BMS, Regeneron Pharmaceuticals, Inc., Shire - research funding. Dellon ES: Abbott, Adare, Aimmune, Alivio, Allakos, Arena, AstraZeneca, Banner, Biorasi, Calypso, Enumeral, EsoCap, Gossamer Bio, GSK, Receptos/BMS, Regeneron Pharmaceuticals, Inc., Robarts, Salix, Shire/Takeda - consultant; Adare, Alla-kos, GSK, Meritage Pharma, Miraca, Nutricia, Receptos/BMS, Regeneron Pharmaceuticals, Inc., Shire - research funding; Allakos, Banner, Holoclara - educational grant. Brian W, Mannent LP: Sanofi - employees, may hold stock and/or stock options in the company. Rothenberg ME: Allakos, AstraZeneca, BMS, ClostraBio, Pulm One, Spoon Guru - consultant; ClostraBio, Pulm One, Spoon Guru - equity interest; Teva Pharmaceuticals – royalties from reslizumab; Mapi Research Trust - royalties from PEESSv2; UpToDate – royalties; an inventor of patents owned by Cincinnati Children's Hospital.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.75

OP080 Changes in Oesophageal Microbiome in Eosinophilic Oesophagitis Before and After Different Therapy Options

EJ Laserna-Mendieta ^1,^2,^3,^✉, J Fitzgerald ^4,⁵, L Arias-González ^1,², D Bernardo ⁶, A Montalban-Arques ⁷, MJ Claesson ^4,⁵, AJ Lucendo ^2,^8,⁹

Introduction

Eosinophilic oesophagitis (EoE) is a chronic immune-mediate inflammatory disorder of the oesophagus in response to a non-IgE-mediated food allergy. The incidence and prevalence of EoE is increasing in the last few years, without a clear explanation for this fact. Many experts believe environmental factors could be the predominant cause, and among them, some have pointed to dysbosis as a main trigger. However, characterization of oesophageal microbiota with culture-independent techniques is a relatively recent field and studies about esophageal micro-biota in EoE are scarce.

Aims & Methods

Here, we aimed to investigate microbiota changes in oesophageal biopsies of patients with EoE before and after treatment-induced disease remission (proton pump inhibitors -PPI-, swallowed topic steroids -STS- and food-elimination diets -FED-; n=10 per group) referred to non-EoE controls (n=10). Hence, and to our knowledge, this is the first microbiota research in EoE differentiating among the three types of normally used treatments. We applied 16S rRNA gene sequencing (V1-V2 regions) and bioinformatic pipelines to assign reads to Amplicon Sequence Variants (ASVs). Three negative controls were included to remove contaminant reads in patient and control samples.

Results

After quality checking, trimming and decontamination processes, 65 samples had enough high-quality reads for microbiota analysis. EoE did not induce significant changes in alpha (intra-individual) diversity compared to controls, and only patients who underwent FED therapy experienced a decrease in their microbial diversity. Comparison of microbiome beta (inter-sample) diversity was performed using non-metric dimensional scaling (NMDS) on un-agglomerated ASV features. NMDS showed centroid position of control samples separated from pretreatment samples. Post-PPI and post-FED treatment groups were closer but more disperse than pre-treatment samples, while post-STS occupied a more nearby position to controls. Hierarchically-clustered Jaccard dissimilarities defined five subtypes of community composition (clusters 1-5), arising within two broader clusters (clusters “A” and “B”). in Arm/ cluster “A” (41 samples) could be found most of control (7 out of 10) and post-STS (7 out of 9) samples; in contrast, arm/cluster “B” comprised 24 samples predominantly from pre-treatment, post-PPI and post-FED groups. Significant changes were identified for Actinobacillus, Gemella, Parvimonas and Porphyromonas genera in the differential analysis of mi-crobial abundance. Parvimonas and Porphyromonas were less abundant in pre-treatement EoE patients than in controls, with Parvimonas displaying a significant recovery after treatment, while patients with EoE before therapy displayed a higher abundance of Gemella and Actinobacillus than controls. When corrected for multiple testing, only Parvimonas (p=0.004) and Porphyromonas (p=0.047) remained significantly different. Finally, functional analysis predicted alterations in degradation of aromatic, sugar-derived and amine compounds and abundance of sulphur-cytochrome oxidoreductases.

Conclusion

We have identified two non-previously associated with EoE genera, Parvimonas and Porphyromonas, whose abundance is decreased in untreated EoE patients. Our findings showed that microbiota changes in EoE were modest but existed, and that reversion of those alterations was affected by treatment, with STS leading to a relatively different microbiota composition compared to PPI and FED and more similar to that observed in controls.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.76

OP081 Timing of Ercp and Outcomes in Patients with Acute Gallstone Cholangitis Graded By Severity in Accordance with Tokyo 2018 Criteria

W On ^1,^✉, C Watters ¹, LK Dwyer ¹, S Hood ¹, R Saleem ¹, R Sturgess ¹, N Stern ¹

Introduction

The optimal timing of endoscopic retrograde cholangiopancreatography (ERCP) in the management of acute gallstone cholangitis is not known. Severity of cholangitis can be classified with the Tokyo 2018 criteria. The European Society of Gastrointestinal Endoscopy published guidance on the recommended timing of ERCP guided by the severity of cholangitis¹; stipulating that biliary drainage should occur within the following timeframes: mild - elective, moderate - within two to three days and severe - as soon as possible.

Aims & Methods

We aim to analyse the clinical outcomes of patients with acute cholangitis who have been admitted to our tertiary hepatobiliary unit when categorised by severity. A retrospective analysis of patients admitted to our hospital with acute cholangitis over a 3 year period from June 2016 to June 2019 was carried out. Patients were identified via coding department and endoscopy reporting tool. All patients met 2018 Tokyo

criteria for definite cholangitis. Only patients with choledocholithiasis without concurrent biliary pathology were included for analysis. Case notes and electronic database interrogation yielded information for calculation of severity of cholangitis. Statistical analyses were carried out with Kruskall-Wallis test or chi-squared tests where appropriate.

Results

A total of 218 patients were identified and 199 patients who underwent ERCP during the index admission were included for analysis. There was a female preponderance (55.8%) and the median age was 73 years (range 19-96). The proportion of severity of cholangitis at presentation was as follows: 51.3% (n=102) mild, 32.6% (n=65) moderate and 16.1% (n=32) severe. The median time taken from admission to ERCP for the 199 patients was 4.8 days (mild 4.4 days, moderate 5.4 days, severe 4.8 days; p=0.31). The median length of stay 7.8 days (mild 7.2 days, moderate 7.8 days, severe 9.5 days; p=0.009). 31.3% of patients with severe cholangitis (n=10) were admitted to intensive care (ITU); 6 of whom required urgent ERCP. For patients with severe cholangitis, the median time in those who required urgent ERCP was 1.5 days vs 5.6 days in those who did not. The overall 30-day all-cause mortality amongst the 199 patients was 1% (n=2; both with severe cholangitis who underwent successful ERCP at 23 hours and 42 hours). 30-day all-cause mortality was 6.3% in the severe group and 0% in both mild and moderate groups (p=0.005).

Conclusion

Our results demonstrate no difference in timing to ERCP in patients with acute gallstone cholangitis when categorised by severity. Deaths were observed only in patients with severe cholangitis although the majority of patients with severe disease did not require urgent ERCP. Provision for urgent ERCP has to be available especially for those admitted to intensive care.

Disclosure

Nothing to disclose

References

1.Manes G., Paspatis G., Aabakken L. et al. Endoscopic management of common bile duct stones: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 2019; 51: 472–91. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.77

OP082 Should Prophylactic Antibiotics Be Administered When Cannulation Fails in Ercp?

G Olsson ^1,^✉, B Andersson ², F Swahn ³, L Enochsson ⁴

Introduction

Endoscopic Retrograde Cholangiopancreatography (ERCP) is associated with a considerable rate of postoperative adverse events, occurring in about 10-15 % of the cases. When cannulation fails, the risk of postoperative complications is considered to be even higher, but the role of using prophylactic antibiotics in this particular situation remains unclear.

Aims & Methods

The aim of this study was to evaluate the rate of postoperative complications in ERCPs where cannulation failed in relation to if prophylactic antibiotics had been administered or not. We analyzed data from 96 818 ERCP procedures, collected from The Swedish National Quality Registry of Cholecystectomy and Endoscopic Retrograde Cholangiopancreatography (GallRiks) between 2006 and 2018. After exclusion of procedures due to successful cannulation (n=88 743), missing data (n=2 014) or on-going antibiotic therapy (n=1 062), a cohort of 4 996 ERCPs remained for the final investigation. The factors affecting the postoperative adverse events were analyzed through multivariable logistic regression modelling and all variables were tested uni- and multivariably.

Results

In the study population (n=4 996), 2 124 (42.5%) patients received antibiotic prophylaxis and the remaining 2 872 (57.5%) did not. There were no demographic differences between the groups regarding sex, age, ASA or if the ERCP procedure was due to an acute or planned indication. There were less overall postoperative adverse events in the group receiving prophylactic antibiotics (13.6% vs 17.1%; p< 0.001). The multivariate analysis demonstrated a 24% overall reduction in postoperative adverse events if prophylactic antibiotics were administered (OR 0.76; 95% CI 0.65-0.89), although the absolute reduction in complications was only 3.5%. in the subgroup of postoperative abscesses there was also a significant reduction in postoperative adverse events if prophylaxis was administered (0.80% vs 1.43%; p=0.040; OR 0.54 95% CI 0.31-0.96) but the absolute number of events was numerically low (n=17 vs n=41). On the contrary, no differences were seen regarding the rate of postoperative cholangitis in relation to antibiotic prophylaxis (1.27% vs 1.74%; p=0.182) and the intraoperative complications did not differ either (7.7% vs 6.3%, p=0.059).

Conclusion

This national quality registry study demonstrates a significant reduction of postoperative ERCP associated complications when profylactic antibiotics are given in cases where cannulation failed. These results indicate that antibiotics should perhaps be offered to all patients with a failed cannulation in ERCP.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.78

OP084 The Current Diagnostic Accuracy and Inter-Observer Agreement of Visual Impression with Digital Single-Operator Cholangioscopy For The Diagnosis of Indeterminate Biliary Strictures

PMC Stassen ^1,^✉, G Goodchild ², PJF de Jonge ¹, N Erler ¹, A Anderloni ³, V Cennamo ⁴, N Church ⁵, I Fernández-Urién Sainz ⁶, M Huggett ⁷, M James ⁸, D Joshi ⁹, L Kylänpää ¹⁰, W Laleman ¹¹, M Nayar ¹², K Oppong ¹², J-W Poley ¹, J Potts ¹³, A Repici ¹⁴, M Udd ¹⁵, JJ Vila ¹⁶, T Wong ¹⁷, M Bruno ¹, G Webster ¹⁸; European Cholangioscopy Group

Introduction

Digital single-operator cholangioscopy (d-SOC) has shown promising results for detection of cholangiocarcinoma in indeterminate biliary strictures, with higher sensitivity rates for visual diagnosis as compared to targeted biopsies. However, in previous studies endoscopists were not blinded for patients’ clinical background and results of previous investigations. We aimed to investigate the diagnostic accuracy and inter-observer agreement (IOA) of d-SOC in the visual appraisal of biliary strictures, whilst blinding the assessors.

Aims & Methods

A multicentre, international cohort study was performed amongst 19 endoscopists experienced in cholangioscopy. Videos of biliary strictures of patients, in whom a final diagnosis of benign or malignant was established, based on pathology or at least 1 year of follow-up without progression, were systematically scored. Videos were anonymised and without cholangiography. Assessment was performed in 2 steps: 1: blinded for patients’ medical history and previous investigations and 2: unblinded. A standardised scoring form was used in which various diagnostic cholangioscopic features were systematically scored.

Results

Forty-four videos were reviewed (59% male, mean age 62 years) of 19 benign and 25 malignant strictures. The sensitivity and specificity for the diagnosis of a malignant biliary stricture was 74.1% and 47.1% for the blinded assessment and 72.8% and 63% for the unblinded assessment. When primary sclerosing cholangitis (PSC) cases were excluded (n=11), sensitivity and specificity was 73.4% and 47.1% (blinded) and 75.1% and 61.1% (unblinded). The IOA of the final diagnosis of a malignant stricture was fair for the blinded assessment (Fleiss’ kappa (F'k) 0.243) and fair for the unblinded assessment (F'k 0.325). For individual cholangioscopic features, the IOA ranged from slight to moderate (F'k 0.058 - 0.401) for the blinded assessment and from slight to moderate (F'k 0.030 - 0.450) for the unblinded assessment. There was moderate agreement for circumferential lesions (F'k 0.401) in the blinded assessment and for villous projections (Fleiss’ kappa 0.450) in the unblinded assessment.

Conclusion

This study shows low sensitivity and specificity rates for blinded as well as unblinded d-SOC video appraisal of indeterminate biliary strictures, with considerable inter-observer variation. This indicates that morphological features of inflammatory and malignant disease overlap and are non-discriminative emphasising the need to establish reproducible consensus opinion on the visual features of biliary strictures and their individual importance in the diagnosis of cholangiocarcinoma as well as the need for a standardized biopsy protocol to improve diagnosis of malignant disease by tissue acquisition.

Disclosure

The European Cholangioscopy Group has received an unrestricted grant from Boston Scientific Inc.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.79

OP085 Equivalent Performance of Single-Use and Reusable Duodenoscopes in A Randomized Trial

JY Bang ^1,^✉, U Navaneethan ¹, R Hawes ¹, S Varadarajulu ¹

Introduction

Single-use duodenoscopes have been recently developed to eliminate risk of infection transmission from contaminated reusable duodenoscopes.

Aims & Methods

We compared performances of single-use and reusable duodenoscopes in patients undergoing endoscopic retrograde cholangiopancreatography (ERCP). Patients with native papilla requiring ERCP were randomized to single-use (EXALT Model D^TM, Boston Scientific) or reusable (Olympus TJF 180) duodenoscope. Primary outcome was comparing number of attempts to achieve successful cannulation of desired duct. Secondary outcomes were technical performance that measured duode-noscope maneuverability, mechanical-imaging characteristics and ability to perform therapeutic interventions, need for advanced cannulation techniques or cross-over to the alternate duodenoscope group to achieve ductal access and adverse events.

Results

98 patients were treated using single-use (n=48) or reusable (n=50) duodenoscopes. While there was no significant difference in cannulation rates, the median number of attempts to achieve successful cannulation was significantly lower for single-use cohort (2 vs. 5, p=0.013). There was no significant difference in adverse events, need to cross-over or need for advanced cannulation techniques to achieve ductal access, between cohorts. While there was no significant difference in ability to undertake therapeutic interventions and overall maneuverability, resistance to passage into stomach (P=0.047), image quality (p< 0.001), image stability (p< 0.001) and air-water valve functionality (p< 0.001) were significantly worse for single-use duodenoscopes.

On multivariate linear regression analysis, only duodenoscope type (single-use) was associated with fewer than 6 attempts to achieve selective cannulation (p=0.012) when adjusted for patient demographics, procedural complexity and type of intervention.

Conclusion

Given the overall safety profile and similar technical performance, single-use duodenoscopes can represent an alternative to reusable duodenoscopes for performing ERCP procedures.

Disclosure

JYB: Consultant for Olympus America Inc., Boston Scientific Corporation. SV: Consultant for Boston Scientific Corporation, Olympus America Inc., Covidien, Creo Medical. RH: Consultant for Boston Scientific Corporation, Olympus America Inc., Covidien, Creo Medical, Nine Points Medical, Cook Medical. UN: No disclosures to declare.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.80

OP086 Higher Vs Standard Adalimumab Induction and Maintenance Treatment in Patients with Moderately-To-Severely Active Ulcerative Colitis: Results From The Phase 3 Serene-Uc Japan Study

S Tanida ^1,^✉, Y Okuyama ², M Watanabe ³, T Hibi ⁴, K Kobayashi ⁵, T Doan ⁶, K Ikeda ⁵, B Huang ⁶, J Petersson ⁶, J Kalabic ⁷, AM Robinson ⁶, J Panés ⁸

Introduction

SERENE-UC (NCT02065622) was a Phase 3 study of higher vs standard adalimumab dosing regimens for induction (HIR vs SIR) and maintenance therapy in patients with moderately-to-severely active ulcerative colitis. Data from the Main (excluding Japan) study were previously reported; here we report data on the Japan study conducted to demonstrate consistency with the integrated (Main plus Japan) population.

Aims & Methods

Patients from Japan were randomized 2:1 to adalim-umab HIR (160mg/week [Wk] at Wk0-3, and 40mg at Wk4 and Wk6) or adalimumab SIR (160mg at Wk0, 80mg at Wk2, and 40mg at Wk4 and Wk6) for 8 weeks induction. Patients completing induction were re-randomized 1:1 to adalimumab 40mg every week (40EW) or adalimumab 40mg every other week (40EOW) for 44 weeks maintenance. The primary endpoint (EP) of the induction study was proportion of patients achieving clinical remission (CR; Full Mayo score [FMS] ≤2 with no subscore >1) at Wk8. The primary EP of the maintenance study was proportion of Wk8 responders (decrease in FMS >3 and >30% from baseline [BL], plus decrease in rectal bleeding subscore [RBS] >1 or absolute RBS ≤1) achieving CR at Wk52. Safety was assessed throughout.

Results

100 patients were enrolled. At Wk8, rates of CR were similar with HIR vs SIR (18.0% vs 17.9%), whereas discontinuation rates were lower with HIR vs SIR (6.6% vs 17.9%); proportions of patients meeting ranked secondary EPs are shown in the Table. 89 patients, including 41 Wk8 re-sponders, completed induction and were re-randomized to maintenance therapy. At Wk52, a greater proportion of Wk8 responders receiving 40EW vs 40EOW achieved CR (52.2% vs 38.9%). Key ranked secondary EPs were numerically higher with 40EW vs 40EOW (Table). Most other ranked secondary EPs were also higher with 40EW vs 40EOW, n/N (%): Wk8 remitters achieving CR, 5/10 (50.0) vs 5/8 (62.5); Wk8 remitters achieving endoscopic improvement, 5/10 (50.0) vs 6/8 (75.0); Wk8 remitters taking steroids at BL who are steroid free for >90 days, 4/8 (50.0) vs 4/6 (66.7); Wk8 remitters taking steroids at BL who are steroid free for >90 days and in CR, 3/8 (37.5) vs 3/6 (50.0); Wk8 responders with inflammatory bowel disease questionnaire (IBDQ) response, 14/23 (60.9) vs 12/18 (66.7); Wk8 nonre-sponders with CR, 4/23 (17.4) vs 3/25 (12.0); Wk8 nonremitters with CR 11/36 (30.6) vs 5/35 (14.3); Week 8 responders achieving endoscopic subscore of 0, 8/23 (34.8) vs 4/18 (22.2); Wk8 remitters achieving endoscopic subscore of 0, 3/10 (30.0) vs 2/8 (25.0); Wk8 responders with response in IBDQ, bowel symptom domain, 19/23 (82.6) vs 12/18 (66.7); Wk8 respond-ers with response in IBDQ fatigue item, 11/23 (47.8) vs 12/18 (66.7). Both maintenance regimens showed similar discontinuation rates. Higher dosing regimens were generally well tolerated.

Conclusion

The Japanese population of SERENE-UC demonstrated similar efficacy with HIR and SIR at Wk8 and numerically greater efficacy for 40EW vs 40EOW maintenance regimens, consistent with the integrated population. All dosing regimens were well tolerated.

[Ranked secondary endpoints for induction and maintenance dosing regimens]

Ranked secondary endpoints, Induction study	Adalimumab higher induction regimen (N=61), n (%)	Adalimumab standard induction regimen (N=39), n (%)
Endoscopic improvement (endoscopic subscore of 0/1)	16 (26.2)	10 (25.6)
Fecal calprotectin below 150mg/kg	14 (23.0)	7 (17.9)
Inflammatory bowel disease questionnaire response (increase of inflammatory bowel disease questionnaire ≥16 from baseline)	30 (49.2)	23 (59.0)
Clinical response (per Full Mayo score)	30 (49.2)	11 (28.2)
Endoscopic remission (endoscopic subscore of 0)	7 (11.5)	4 (10.3)
Key ranked secondary endpoints, Maintenance study	Adalimumab 40mg every week, n/N (%)	Adalimumab 40mg every other week, n/N (%)
Week 8 responders achieving endoscopic improvement (endoscopic subscore of 0 or 1)	13/23 (56.5)	8/18 (44.4)
Week 8 responders taking steroids at baseline who are steroid free for ≥90 days	9/13 (69.2)	7/11 (63.6)
Week 8 responders taking steroids at baseline who are steroid free for ≥90 days and in clinical remission	6/13 (46.2)	4/11 (36.4)

Open in a new tab

Disclosure

AbbVie funded this study and participated in the study design, research, analysis, data collection, interpretation of data, reviewing, and approval of the publication. All authors had access to relevant data and participated in the drafting, review, and approval of this publication. No honoraria or payments were made for authorship. Medical writing assistance was provided by Gregory Bezkorovainy of 2 the Nth, which was funded by AbbVie Inc. Satoshi Tanida: Received research grants from EA Pharma Co., Ltd., and AbbVie GK. Toshifumi Hibi: Is an editor of Intestinal Research, has received grants, personal fees, and/or other funds from AbbVie GK, Celltrion Healthcare, EA Pharma, Eli Lilly Japan K.K., Ferring Pharmaceuticals, Gilead Sciences K.K., Janssen Pharmaceutical K.K., JIMRO, Kissei Pharmaceutical, Kyorin Pharmaceutical, Mitsubishi Tanabe Pharma Corporation, Miyarisan Pharmaceutical, Mochida Pharmaceutical, Nichi-Iko, Nippon Kayaku, Otsuka Holdings, Pfizer Japan,

Takeda Pharmaceutical, and Zeria Pharmaceutical. Mamoru Watanabe: Has received grants, personal fees, and/or other funds from AbbVie GK, Alfresa Pharma Corporation, Asahi Kasei Medical, Astellas Pharma, Ayumi Pharmaceutical Corporation, Celgene K.K., Celltrion Healthcare, Chugai Pharmaceutical, Daiichi Sankyo, EA Pharma, Eisai, Eli Lilly Japan K.K., Fu-jirebio, Gilead Sciences, Janssen Pharmaceutical K.K., JIMRO, Kaken Pharmaceutical, Kissei Pharmaceutical, Kyorin Pharmaceutical, Kyowa Hakko Kirin, Mitsubishi Tanabe Pharma Corporation, Miyarisan Pharmaceutical, Mochida Pharmaceutical, MSD K.K., Nippon Kayaku, Pfizer Japan, Taiho Pharmaceutical, Takeda Pharmaceutical, Toray Industries, and Zeria Pharmaceutical. Yusuke Okuyama: Reports no conflicts of interest. Julián Panés: Financial support for research: AbbVie, MSD, and Pfizer; lecture fee(s): AbbVie, Ferring Pharmaceuticals, Janssen, MSD, Pfizer, Shire Pharmaceuticals, Takeda, and Theravance; consultancy: AbbVie, Arena Pharmaceuticals, Boehringer Ingelheim, Celgene, Ferring Pharmaceuticals, Genentech, GlaxoSmithKline, GoodGut, Janssen, MSD, Nestlé, Oppilan, Pfizer, Progenity, Robarts, Roche, Takeda, Theravance, TiGenix, and Topi-vert. Thao Doan, Kimitoshi Ikeda, Bidan Huang, Jasmina Kalabic, Anne M. Robinson, Joel Petersson, and Kazuya Kobayashi: AbbVie employees and may own stock or options.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.81

OP087 Long-Term Safety and Efficacy of Ozanimod in Patients with Moderate-To-Severe Ulcerative Colitis: Results From The Touchstone Open-Label Extension

W Sandborn ^1,^✉, BG Feagan ², SB Hanauer ³, S Vermeire ⁴, S Ghosh ⁵, WJ Liu ⁶, A Petersen ⁶, L Charles ⁶, V Huang ⁶, K Usiskin ⁶, DC Wolf ^7,⁸, G D'Haens ⁹

Introduction

The efficacy and safety of ozanimod, a potent, selective, sphingosine-1-phosphate receptor subtype 1 (S1P1) modulator, was previously demonstrated in the randomized, placebo-controlled, phase 2 TOUCHSTONE study (NCT01647516) in patients (pts) with moderate-to-severe ulcerative colitis (UC). Ozanimod 1 mg/day improved partial Mayo score (PMS) by the first assessment (wk 8),¹ which was sustained through open-label extension (OLE) wk 104 of TOUCHSTONE.² Here, we report longer-term follow-up from TOUCHSTONE OLE, including efficacy data with up to 4 years of follow-up and safety through end of OLE.

Aims & Methods

Pts in TOUCHSTONE received placebo or ozanimod (0.5 mg or 1 mg daily) during the 8-wk induction and 24-wk maintenance periods and could enter the optional OLE (ozanimod 1 mg/day) if they were nonresponders at the end of the induction period, lost response during the maintenance period, or completed the maintenance period. Eligible pts entered OLE May 2013-March 2015. in 2019, the sponsor ended the OLE and rolled over all active pts who consented to a phase 3 program (all had completed at least OLE wk 200). During OLE, pts were followed for safety and efficacy at OLE wks 4, 8, 12, and 12-wk intervals thereafter. PMS (stool frequency, rectal bleeding, physician's global assessment subscores [overall scale: 0-9; higher scores=worse outcomes]) was assessed at all visits. Endoscopy was performed approximately annually in pt subsets, mainly at OLE wks 56 and 104 per protocol amendments; thus, total Mayo score (TMS, including PMS subscores plus endoscopy), endoscopic improvement (endoscopic subscore of ≤1; scale: 0=normal/inactive disease to 3=severe disease), and histologic remission (Geboes score ≤ 2; scale: 0-5 [0=structural change only,1=chronic inflammation]) were summarized at these time points. Clinical response based on PMS or TMS was defined as a reduction from baseline of >2 (>3 for TMS) and >30%, with reduction in rectal bleeding subscore (RBS) of >1 or absolute RBS of ≤1. Clinical remission was defined as PMS or TMS score of ≤2 with no individual subscore >1. Biomarkers of disease activity were assessed (C-reactive protein [CRP], all visits; fecal calprotectin [FCP], OLE wk 8, end of study).

Results

Of 170 pts entering OLE, 123 (72%), 102 (60%), 84 (49%), and 71 (42%) completed wks 56, 104, 152, and 200, respectively. Using non-responder imputation (NRI) for missing data, PMS clinical response and remission rates at OLE wk 56 were 71% and 55%, decreasing to 41% and 37%, respectively, at OLE wk 200. TMS clinical response and remission rates (observed case analysis) were 83% and 40%, respectively, at OLE wk 56 and 82% and 41% at OLE wk 104. Observed PMS decreased substantially over time, plateauing at < 2. Endoscopic improvement rates were 46% and 47% and histologic remission rates were 51% and 38% at OLE wks 56 and 104, respectively. CRP was reduced by a median of 25% by OLE wk 4 and maintained through wk 200; FCP was reduced by a median of 68% by OLE wk 8 (only available time point). The most common TEAEs during OLE were UC (6%), hypertension (6%), upper respiratory infection (6%), and increased gamma glutamyl transferase (5%).The most common serious TEAEs during OLE (>1 patient) were UC (4%), anemia (1%), and ischemic stroke (1%); no serious TEAEs of cardiac arrhythmias or macular edema were reported.

Conclusion

Data from TOUCHSTONE OLE demonstrate durable efficacy by clinical, endoscopic, histologic, and biomarker measures with ozanimod 1 mg/day. No new safety risks were identified with ≥4 years of follow-up.

Disclosure

WS: Atlantic Healthcare Limited, Amgen, Genentech, Gilead Sciences, AbbVie, Janssen, Takeda, Lilly, Celgene - research funding; AbbVie, Allergan, Amgen, Boehringer Ingelheim, Celgene, Conatus, Cosmo, Escalier Biosciences, Ferring, Genentech, Gilead, Gossamer Bio, Janssen, Lilly, Miraca Life Sciences, Nivalis Therapeutics, Novartis Nutrition Science Partners, Oppilan Pharma, Otsuka, Paul Hastings, Pfizer, Precision IBD, Progenity, Prometheus Laboratories, Ritter Pharmaceuticals, Robarts Clinical Trials, Salix, Shire, Seres Therapeutics, Sigmoid Biotechnologies, Takeda, Tigenix, Tillotts Pharma, UCB Pharma, Vivelix - consultant; Ritter Pharmaceuticals, Oppilan Pharma, Escalier Biosciences, Gossamer Bio, Precision IBD, Progenity - shareholder. BF: AbbVie, ActoGeniX, Albireo, Amgen, AstraZeneca, Avaxia Biologics, Baxter, Biogen Idec, Boehringer Ingelheim, BMS, Calypso, Celgene, Elan, EnGene, Ferring Pharma, Roche/ Genentech, GiCare, Gilead, Given Imaging, GSK, Ironwood, Janssen, Johnson & Johnson, Lexicon, Lilly, Merck, Millennium, Nektar, Novo Nordisk, Pfizer, Prometheus, Protagonist, Celgene, Sanofi, UCB - consultant; Ro-barts Clinical Trials - director. SH: Celgene, AbbVie, Janssen, UCB, Shire, Actavis, Salix, BMS, Merck, Pfizer, Boehringer Ingelheim, Sanofi-Aventis, Ferring, Caremark - consultant. SV: AbbVie, MSD, Centocor - research funding; AbbVie, MSD, Takeda, Pfizer, Genentech/Roche, Ferring, Mundiphar-ma, Celgene - consultant. SG: AbbVie - research funding; AbbVie, MSD - lecturer; Pfizer, Novo Nordisk, BMS, Janssen, AbbVie, Celgene - consultant. WL, AK, LC, VH & KS: are employed by Bristol-Myers Squibb. DW: AbbVie, Amgen, Elan, Given Imaging, Genentech, Janssen, Millennium, Pfizer, Prometheus, Celgene, UCB - research funding; AbbVie, Janssen, Prometheus, Santarus, Salix, Takeda, UCB - lecturer; AbbVie, Genentech, Given Imaging,

Janssen, Prometheus, Salix, Takeda, UCB - consultant. GD: AbbVie, Ablynx, Amakem, AM Pharma, Avaxia, Biogen, Bristol-Myers Squibb, Boehringer Ingelheim, Celgene, Celltrion, Cosmo, Covidien/Medtronic, Ferring, Dr. Falk Pharma, EnGene, Galapagos, Genentech/Roche, Gilead, GlaxoSmith-Kline, Hospira, Immunic, Johnson & Johnson, Lycera, Medimetriks, Millen-nium/Takeda, Mitsubishi Pharma, Merck Sharp & Dohme, Mundipharma, Novo Nordisk, Otsuka, Pfizer, Prometheus Laboratories/Nestle, Protagonist, Receptos, Robarts Clinical Trials, Salix, Sandoz, SetPoint, Shire, Teva, Tigenix, Tillotts, TopiVert, Versant, Vifor - consultant; AbbVie, Biogen, Fer-ring, Johnson & Johnson, Merck Sharp & Dohme, Mundipharma, Norgine, Pfizer, Millennium/Takeda, Tillotts, Vifor - speaker; Robarts Clinical Trials - director; EnGene - shareholder.

References

1.Sandborn W.J. et al. Ozanimod induction and maintenance treatment for ulcerative colitis. N Engl J Med. 2016; 374: 1754–1762. [DOI] [PubMed] [Google Scholar]
2.Feagan B. et al. Ozanimod, an Oral S1P Receptor Modulator, Is Effective and Well-Tolerated in the Long-Term Treatment of Moderate to Severe Ulcerative Colitis. Am J Gastroenterology. 2017; S3: Abst P-012. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.82

OP088 Composite and Individual Measures of Efficacy and Safety After 2 Years of Upadacitinib Treatment For Crohn'S Disease: Results From The Ongoing Phase 2 Celest Open-Label Extension Study

GR D'Haens ^1,^✉, J Panés ², E Louis ³, Q Zhou ⁴, J Liu ⁴, AP Lacerda ⁴, EV Loftus ⁵

Introduction

In patients with moderate to severe Crohn's disease (CD), phase 2 CELEST study data showed safety and efficacy of upadacitinib (UPA), an oral selective Janus kinase 1 inhibitor, for induction and maintenance therapy at 12 months (M) [1]. The CELEST open-label extension (OLE) study showed sustained safety and efficacy following an additional 12M of UPA. [2]. To further validate long-term efficacy, additional, more stringent, composite endpoints, as well as the separate disease indices from the 12M CELEST OLE study were examined.

Aims & Methods

Patients completing the 52-week (W) CELEST study were enrolled in the CELEST OLE study, receiving a UPA modified release formulation once daily (QD) for up to 8 years. Participants on double-blind therapy in CELEST received 15 mg QD in this OLE study; those on open-label rescue therapy received 30 mg QD. We investigated the proportions of patients: 1) in clinical and endoscopic remission/response; 2) in steroid-free and CD Activity Index (CDAI) remission among patients receiving steroids at CELEST baseline (BL); and 3) displaying no extraintestinal manifestations (EIMs, see Table 1) at W0 and 12M. Mean daily very soft/liquid stool frequency (SF), abdominal pain (AP) score, Simple Endoscopic Score for Crohn's Disease (SES-CD), and Inflammatory Bowel Disease Questionnaire (IBDQ) domain scores were also assessed at W0 and 12M. Missing data were imputed via last observation carried forward for clinical endpoints; observed-case analyses were performed for endoscopic-related endpoints, steroid-free remission and EIMs.

Table 1.

[Clinical and endoscopic endpoints at Week 0 and Month 12 in the CELEST OLE study]

	15 mg QD		30 mg QD
Endpoint	W0, n/N (%)	12M, n/N (%)	W0, n/N (%)	12M, n/N (%)
Clinical and Endoscopic Remission	10/69 (14.5)	15/41 (36.6)	4/28 (14.3)	6/19 (31.6)
Clinical and Endoscopic Response	45/69 (65.2)	30/41 (73.2)	19/28 (67.9)	16/19 (84.2)
Steroid-Free CDAI Remission	19/27 (70.4)	12/14 (85.7)	5/14 (35.7)	9/12 (75.0)
Resolution of Extraintestinal Manifestations	24/39 (61.5)	19/26 (73.1)	9/16 (56.2)	8/11 (72.7)
	W0, Mean (SD)	12M, Mean (SD)	12M, Mean (SD)	12M, Mean (SD)
Stool Frequency	2.0 (2.1)	1.7 (2.0)	2.4 (2.3)	2.2 (2.3)
Abdominal Pain Score	0.8 (0.7)	0.7 (0.8)	0.7 (0.7)	0.7 (0.8)
SES-CD	6.0 (4.4)	5.4 (6.7)	8.4 (7.4)	5.3 (6.3)
IBDQ - Total	169.8 (34.3)	173.6 (41.1)	163.7 (32.8)	163.1 (31.6)
IBDQ - Bowel domain	54.4 (10.2)	56.2 (12.3)	55.2 (9.8)	54.3 (9.7)
IBDQ - Systemic domain	23.1 (6.7)	23.9 (7.3)	22.2 (5.8)	21.1 (6.1)
IBDQ - Social domain	28.8 (6.6)	29.4 (7.5)	27.7 (6.6)	28.7 (6.4)
IBDQ - Emotional domain	63.5 (13.8)	64.2 (16.6)	59.5 (16.2)	60.0 (14.8)

Open in a new tab

Intent-to-treat population UPA= Upadacitinib; QD= once daily; CDAI= Crohn's disease activity index; SES-CD= Simple Endoscopic Score for Crohn's Disease; IBDQ= Inflammatory Bowel Disease Questionnaire

Clinical and Endoscopic Remission - average daily very soft/liquid stool frequency (SF) ≤2.8 and abdominal pain (AP) score ≤1.0, both not worse than baseline (BL) of the CELEST study, in patients with SF >4 or AP score >2.0 at baseline; AND SES-CD≤4 and >2 point reduction versus BL, no subscore less than 1.

Clinical and Endoscopic Response - >30% reduction from BL in SF or AP; neither worse than baseline; AND SES-CD >25% reduction from BL

Steroid-Free CDAI Remission - patients who discontinue steroids and achieve CDAI < 150 among patients receiving steroids at CELEST BL

Resolution of EIMs - zero EIMs at W0 or 12M, in patients with EIMs at BL

Results

Of 107 patients, 76 received 15 mg QD and 31 received 30 mg QD UPA. Clinical and endoscopic remission was observed in 14% of subjects at W0 for both groups. At 12M, clinical and endoscopic remission occurred in 36.6% of the 12M visit completers in the 15 mg QD group and 31.6% in the 30 mg group (Table 1). The proportion of patients with clinical and endoscopic response appeared to remain stable from W0 to 12M in the 30 mg QD group (Table 1). Steroid-free CDAI remission rates (Table 1) were similar at W0 and 12M in the 15 mg QD group and higher at 12M in the 30 mg QD group. Mean daily SF and AP were maintained between W0 and 12M, as were improvements in SES-CD and EIMs compared to baseline. IBDQ total and its domains were maintained (Table). The incidence of AEs, serious AEs, and severe AEs was similar across UPA groups, as previously reported [1-3].

Conclusion

Subjects with moderate-to-severe CD receiving long-term therapy with UPA may achieve tight disease control, with deep remission and improved quality of life.

Disclosure

Geert D'Haens: Consulting and/or lecture fees from AbbVie, ActoGeniX, AIM, Allergan, Amgen, Arena Pharmaceuticals, Boehringer Ingelheim GmbH, Celgene/Receptos, Celltrion, Cosmo Technologies, Elan Pharmaceuticals, Eli Lilly, enGene, Dr Falk Pharma, Ferring, Galapagos, Genentech, Gilead Sciences, Giuliani SpA, Given Imaging, GlaxoSmith-Kline, Gossamer Bio, Janssen Biologics, MSD, Neovacs, Novo Nordisk, Otsuka, PDL BioPharma, Prometheus Laboratories, Progenity, Pfizer, Ro-barts Clinical Trials, Salix, Seres/Nestle, Schering-Plough, SetPoint, Shire Pharmaceuticals, Takeda, Tillotts Pharma, UCB Pharma, Versant, and Vifor Pharma; research grants from AbbVie, Dr Falk Pharma, Given Imaging, Janssen, MSD, and PhotoPill; speaking honoraria from AbbVie, Fer-ring, MSD, Norgine, Shire, Tillotts Pharma, Tramedico, and UCB Pharma. Edward V. Loftus Jr: Consultancy: AbbVie, Allergan, Amgen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion Healthcare, Eli Lilly,

Genentech, Gilead, Janssen, Pfizer, Takeda, and UCB; research support: AbbVie, Amgen, Genentech, Gilead, Janssen, Medimmune, Pfizer, Receptos, Robarts Clinical Trials, Seres Therapeutics, Takeda, and UCB. Julian Panés: Financial support for research: AbbVie and MSD; lecture fee(s): Ab-bVie, Ferring, Janssen, MSD, Pfizer, Shire Pharmaceuticals, Takeda, and Theravance; consultancy: AbbVie, Arena Pharmaceuticals, Boehringer Ingelheim, Celgene, Ferring, Genentech, GlaxoSmithKline, GoodGut, Janssen, MSD, Nestle, Oppilan, Pfizer, Progenity, Robarts, Roche, Takeda, Theravance, TiGenix, and Topivert. Edouard Louis: Research Grant: Take-da, Pfizer, Janssen; Educational Grant: Abbvie, MSD, Takeda, Janssen; Speaker Fees: Abbvie, Ferring, MSD, Falk, Takeda, Hospira, Janssen, Pfizer; Advisory Board: Abbvie, Ferring, MSD, Takeda, Celgene, Hospira, Janssen, Pfizer; Consultant: Abbvie Ana Lacerda, Qian Zhou, and Jiangzhou Liu are AbbVie employees and may own AbbVie stock and/or options. The study was funded by AbbVie. AbbVie participated in the study design, data acquisition and interpretation, and in the writing, review, and approval of this abstract. Medical writing assistance was provided by Stephanie Parsons, PhD of Abbvie Inc.

References

1.Sandborn W.J. et al. Efficacy and Safety of Upadacitinib in a Randomized Trial of Patients with Crohn's Disease. Gastroenterology, 2020. [DOI] [PubMed] [Google Scholar]
2.Geert R., D'Haens J.P., Louis Edouard, Zhou Qian, Pugatch David, Liu John, Ana P. Lacerda, Edward V. Loftus Jr. Long-term efficacy and safety of upadacitinib treatment in patients with Crohn's disease: One-year results of the ongoing phase 2 CELEST open-label extension study. United European Gastroenterology Journal, 2019. 7(8 Suppl): p. 358–359.31019704 [Google Scholar]
3.Panes J. et al. P273 Efficacy and safety of upadacitinib maintenance treatment for moderate to severe Crohn's disease: Results from the CELEST study. Journal of Crohn's and Colitis, 2018. 12 (supplement 1): p. S238–S239. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.83

OP089 The Efficacy and Safety of Guselkumab Induction Therapy in Patients with Moderately To Severely Active Crohn'S Disease: Week 12 Interim Analyses From The Phase 2 Galaxi 1 Study

WJ Sandborn ^1,^✉, D Chan ², J Johanns ², G Lang ², OJ Adedokun ², A Afzali ³, JM Andrews ⁴, G D'Haens ⁵, S Danese ⁶, T Hisamatsu ⁷, R Panaccione ⁸, J Panés ⁹, W Reinisch ¹⁰, DT Rubin ¹¹, BE Sands ¹², BG Feagan, on behalf of the GALAXI 1 Investigators¹³

Introduction

Increasing evidence suggests a predominant role for IL-23 in Crohn's disease (CD). Guselkumab (GUS) is an IL-23 antagonist approved for the treatment of moderate to severe plaque psoriasis. The Phase 2 GALAXI 1 double-blind, placebo-controlled, multicenter study evaluates the efficacy and safety of GUS in patients with moderately to severely active CD with inadequate response/intolerance to conventional therapies (corticosteroid, immunosuppressive) and/or biologics (TNF antagonist, vedolizumab).

Aims & Methods

Patients were randomized 1:1:1:1:1 into 5 arms: GUS 200, 600, or 1200mg IV at Weeks (Wks) 0, 4, 8; ustekinumab (UST) ∼6mg/kg IV at Wk 0 and 90mg SC at Wk 8; or PBO IV. Interim analyses at Wk 12 evaluated the key outcomes of change in CDAI score from baseline, clinical remission and response, PRO-2 remission, clinical-biomarker response, endoscopic response, and safety in patients treated with GUS compared with PBO. UST was a reference arm.

Results

Two hundred fifty patients were evaluated; approximately 50% had failed biologic therapy. Baseline demographics and disease characteristics were generally similar among treatment groups (mean CD duration, 8.8 yr; mean CDAI, 306.6; median SES-CD, 11.0). Significantly greater reductions from baseline in CDAI were observed at Wk 12 in each GUS group vs PBO (LS means: -154.1, -144.3, -149.5 vs -36.0, respectively) and a higher proportion of patients assigned to each GUS dose achieved clinical remission (CDAI< 150): 54.0%, 56.0%, 50.0% vs 15.7%, respectively (Table 1). Similarly, at Wk 12, a higher proportion of patients treated with GUS achieved clinical response, PRO-2 remission, clinical-biomarker response, and endoscopic response vs patients treated with PBO. Among bio-failure patients, 45.5% (35/77) in the GUS combined group vs 12.5% (3/24) PBO achieved clinical remission at Wk 12; among conventional therapy failures, 61.6% (45/73) in the GUS combined group vs 18.5% (5/27) treated with PBO achieved clinical remission at Wk 12. Through Wk 12, overall rate of discontinuation was low (3.6%) and safety event rates were generally similar amongst treatment groups for AEs (40.0%, 52.0%, 46.0% and 56.9%), serious AEs (4.0%, 4.0%, 2.0% and 3.9%), infections (10.0%, 14.0%, 14.0% and 17.6%) and serious infections (2.0%, 0%, 0% and 0%) in the GUS 200, 600, 1200mg IV and PBO treatment groups, respectively. Through Wk 12, no cases of active TB, serious hypersensitivity reactions, or malignancies were reported.

Table 1.

[Efficacy analyses at Week 12]

	Placebo (Control)	Guselkumab 200 mg IV q4w	Guselkumab 600 mg IV q4w	Guselkumab 1200 mg IV q4w	Guselkumab Combined	Ustekinumab^a (Reference)
Primary efficacy analysis set	51	50	50	50	150	49
Change from baseline in CDAI score: N, Least Squares Mean (80% CI)^b,c	49, -36.0 (-53.3, -18.7)	48, -154.1 (-171.6, -136.6) p<0.001	49, -144.3 (-161.6, -126.9) p<0.001	47, -149.5 (-167.3, -131.7) p<0.001	144, -149.2 (-159.3, -139.2) p<0.001	49, -136.2 (-153.8, -118.7) p<0.001
Patients in clinical remission^c,d,e n (%)	8 (15.7%)	27 (54.0%) p<0.001	28 (56.0%) p<0.001	25 (50.0%) p<0.001	80 (53.3%) p<0.001	22 (44.9%)
Patients in clinical response ^c,e,f n (%)	12 (23.5%)	33 (66.0%) p<0.001	34 (68.0%) p<0.001	32 (64.0%) p<0.001	99 (66.0%) p<0.001	33 (67.3%)
Patients in PRO-2 remission^c,e,g n (%)	9 (17.6%)	20 (40.0%) p=0.014	27 (54.0%) p<0.001	19 (38.0%) p=0.022	66 (44.0%) p<0.001	19 (38.8%)
Patients in clinical-biomarker response^c,e,h n (%)	4 (7.8%)	27 (54.0%) p<0.001	24 (48.0%) p<0.001	21 (42.0%) p<0.001	72 (48.0%) p<0.001	25 (51.0%)
Patients in endoscopic response^c,e,i n(%)	6 (11.8%)	18 (36.0%) p=0.007	20 (40.0%) p=0.002	18 (36.0%) p=0.003	56 (37.3%) p<0.001	15 (30.6%)
Patients in endoscopic remission^c,e,j n (%)	2 (3.9%)	8 (16.0%) p=0.064	5 (10.0%) p=0.255	8 (16.0%) p=0.041	21 (14.0%) p=0.053	7 (14.3%)

Open in a new tab

^a

Ustekinumab ∼6 mg/kg (260 mg for weight ≤55 kg; 390 mg for weight >55 kg and ≤85 kg; 520 mg for weight >85 kg) IV -> 90 mg SC.

^b

Least Squares Mean based on a Mixed Effect Model Repeated Measures model.

^c

P-values compared guselkumab treatment group and ustekinumab treatment group with the placebo treatment group; p-values were not adjusted for multiplicity.

^d

Clinical remission defined as CDAI score <150.

^e

Participants who had insufficient data to determine remission/response status at Week 12 were considered not to be in remission/response.

^f

Clinical response defined as ≥100-point reduction from baseline in CDAI score or CDAI score <150.

^g

PRO-2 remission defined as an abdominal pain mean daily score ≤1 and stool frequency mean daily score ≤3.

^h

Clinical-biomarker response defined as clinical response and ≥50% reduction from baseline in CRP or fecal calprotectin.

ⁱ

Endoscopic response defined as ≥50% improvement from baseline in the Simple Endoscopic Score for Crohn's Disease (SES-CD) or SES-CD score ≤2.

^j

Endoscopic remission defined as SES-CD score ≤2.

Conclusion

GUS at all 3 doses (200, 600, and 1200mg IV) induced significantly greater improvements vs PBO across the key clinical and endoscopic outcome measures at Wk 12 in patients with moderately to severely active CD who had previously failed biologic or conventional therapy. No clear dose/exposure-response was apparent within the range of doses tested. Through Wk 12, GUS demonstrated a safety profile consistent with that established from clinical trials across investigational and approved indications.

Disclosure

This study was supported by Janssen Research & Development, LLC.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.84

OP090 Long-Term Treatment with Vedolizumab Sc in Ulcerative Colitis: Interim Results From Visible Ole

S Vermeire ^1,^✉, S Danese ², K Kisfalvi ³, W Zhang ³, S Adsul ⁴, S Bhatia ³, W Sandborn ⁵

Introduction

Vedolizumab (VDZ) is a gut-selective, humanised, a₄β₇ inte-grin monoclonal antibody currently available as an intravenous (IV) infusion and approved to treat moderately to severely active ulcerative colitis (UC) and Crohn's disease (CD). The efficacy and safety of a subcutaneous (SC) formulation was evaluated as maintenance therapy in patients with UC (VISIBLE 1) and CD (VISIBLE 2).^1,2

Aims & Methods

The ongoing VISIBLE open-label extension (OLE) is a multinational, multicentre, phase 3 study (NCT02620046/EudraCT 2015-000482-31) of patients with UC or CD previously enrolled in VISIBLE 1 or 2, respectively. We report interim results from the UC patient population of VISIBLE OLE. Patients who enrolled in VISIBLE OLE after completing 52 weeks of treatment in VISIBLE 1 (randomised completers) received VDZ 108 mg SC every 2 weeks (Q2W) when they entered VISIBLE OLE. Patients who responded by Week 14 in VISIBLE 1 after an additional third VDZ IV induction dose (nonrandomised Week 14 responders) received VDZ SC 108 mg Q2W in VISIBLE OLE. Safety is being evaluated as the VISIBLE OLE primary endpoint. Clinical efficacy outcomes included long-term clinical remission, defined as a partial Mayo score of ≤2 and no individual sub-score >1 point, and corticosteroid (CS)-free clinical remission.

[Table. Clinical Remission and Corticosteroid-Free Clinical Remission Over Time in Patients with Ulcerative Colitis]

VISIBLE 1 Randomised Completers, n/N, %^a
Outcome in VISIBLE OLE Patients by Prior Treatment in VISIBLE 1	Baseline^b	Week 6^b	Week 52^c	Week 76	Week 108
Clinical remission^d
Placebo	0/20 (0)	11/20 (55.0)	13/20 (65.0)	15/20 (75.0)	8/17 (47.1)
VDZSC	0/69 (0)	49/69 (71.0)	61/69 (88.4)	55/69 (79.7)	42/61 (68.9)
VDZIV	0/35 (0)	20/35 (57.1)	27/35 (77.1)	30/35 (85.7)	21/28 (75.0)
CS-free clinical remission^d
Placebo	0/8 (0)	1/8 (12.5)	5/8 (62.5)	5/8 (62.5)	3/7 (42.9)
VDZ SC	0/23 (0)	0/23 (0)	18/23 (78.3)	18/23 (78.3)	14/20 (70.0)
VDZ IV	0/14 (0)	0/14 (0)	10/14 (71.4)	10/14 (71.4)	10/12 (83.3)
Nonrandomised Week 14 Responders, n/N, %^a
Outcome in VISIBLE OLE Patients	Baseline^b	Week 14^b,c	Week 54	Week 78	Week 110
Clinical remission^d	0/107 (0)	67/107 (62.6)	47/107 (43.9)	44/107 (41.1)	31/93 (33.3)
CS-free clinical remission^d	0/49 (0)	1/49 (2.0)	12/49 (24.5)	13/49 (26.5)	11/44 (25.0)

Open in a new tab

CS, corticosteroid; IV, intravenous; OLE, open-label extension; SC, subcutaneous; VDZ, vedolizumab.

^a

Among evaluable patients at a particular visit, patients with missing data for determination of endpoint status were categorised as nonresponders.

^b

Analysis limited to VISIBLE 1 patients that rolled into VISIBLE OLE.

^c

Week 52 of this analysis combined Week 52 of VISIBLE 1 and Week 0 of VISIBLE OLE; Week 14 of this analysis combined Week 14 of VISIBLE 1 and Week 0 of VISIBLE OLE.

^d

Clinical remission defined as a partial Mayo score ≤2 and no individual subscore >1 point; CS-free clinical remission defined as patients on a baseline oral CS who discontinued oral CS use and were in clinical remission.

Results

VISIBLE OLE had enrolled 288 patients from VISIBLE 1 at the time of this interim analysis. During VISIBLE OLE, adverse events (AEs) occurred in 69% of patients with UC, with UC exacerbations (18%), nasopharyngitis (11%), upper respiratory tract infection (9%), and anaemia (7%) reported most frequently. Serious AEs occurred in 14% of patients with UC. Injection site reactions were reported in 4.5% of patients with UC and all were mild or moderate in severity. There were no cases of progressive multifocal leukoencephalopathy and no deaths. with VDZ SC treatment, clinical remission rates were maintained from Weeks 6 to 108 in VISIBLE 1 randomised completers and from Weeks 14 to 110 in the nonrandomised Week 14 responders (Table). Corticosteroid-free clinical remission rates were maintained in VISIBLE 1 randomised completers from Weeks 52 to 108 and in nonrandomised Week 14 responders from Weeks 54 to 110 (Table).

Conclusion

This VISIBLE OLE interim analysis supports the safety and efficacy of VDZ SC during longer-term maintenance treatment of patients with UC. These long-term safety findings are consistent with the known safety profile of VDZ. Clinical remission and CS-free clinical remission rates were maintained with up to 2 years of VDZ SC treatment.

Disclosure

SV: Financial support for research: AbbVie, Takeda, Pfizer, Johnson & Johnson; Lecture fee(s): Merck Sharp & Dohme Corp., AbbVie, Takeda, Ferring, Centocor, Hospira, Pfizer, Johnson & Johnson, Genen-tech/Roche, Tillotts. Consultancy: Merck Sharp & Dohme Corp., AbbVie, Takeda, Ferring, Centocor, Hospira, Pfizer, Johnson & Johnson, Genen-tech/Roche, Celgene, Mundipharma, Celltrion, SecondGenome, Prometheus, Gilead, Galapagos, ProDigest, Abivax, GSK, Tillotts. SD: Lecture fees: AbbVie, Ferring, Hospira, Johnson & Johnson, Merck, MSD, Takeda, Mundipharma, Pfizer Inc, Tigenix, UCB Pharma, Vifor, Biogen, Celgene, Allergan, Celltrion, Sandoz, Boehringer Ingelheim; Consultancy: AbbVie, Ferring, Hospira, Johnson & Johnson, Merck, MSD, Takeda, Mundipharma, Pfizer Inc, Tigenix, UCB Pharma, Vifor, Biogen, Celgene, Allergan, Celltrion, Sandoz, Boehringer Ingelheim. KK: Employee of Takeda; holds Takeda stock or stock options. WZ: Employee of Takeda; holds Takeda stock or stock options. SA: Employee of Takeda; holds Takeda stock or stock options. SB: Employee of Takeda; holds Takeda stock or stock options. WJS: Financial support for research: Atlantic Healthcare Limited, Amgen, Genentech, Gilead Sciences, AbbVie, Janssen, Takeda, Lilly, Celgene/Recep-tos, Pfizer, Prometheus Laboratories (now Prometheus Biosciences); Consulting fees from AbbVie, Allergan, Amgen, Arena Pharmaceuticals, Avex-egen Therapeutics, BeiGene, Boehringer Ingelheim, Celgene, Celltrion, Conatus, Cosmo, Escalier Biosciences, Ferring, Forbion, Genentech, Gil-ead Sciences, Gossamer Bio, Incyte, Janssen, Kyowa Kirin Pharmaceutical Research, Landos Biopharma, Lilly, Oppilan Pharma, Otsuka, Pfizer, Pro-genity, Prometheus Biosciences (merger of Precision IBD and Prometheus Laboratories), Reistone, Ritter Pharmaceuticals, Robarts Clinical Trials (owned by Health Academic Research Trust, HART), Series Therapeutics, Shire, Sienna Biopharmaceuticals, Sigmoid Biotechnologies, Sterna Bio-logicals, Sublimity Therapeutics, Takeda, Theravance Biopharma, Tigenix, Tillotts Pharma, UCB Pharma, Ventyx Biosciences, Vimalan Biosciences, Vivelix Pharmaceuticals; and Stock or stock options from BeiGene, Escalier Biosciences, Gossamer Bio, Oppilan Pharma, Prometheus Biosciences (merger of Precision IBD and Prometheus Laboratories), Progenity, Ritter Pharmaceuticals, Ventyx Biosciences, Vimalan Biosciences. Spouse: Opthotech - consultant, stock options; Progenity - consultant, stock; Op-pilan Pharma - employee, stock options; Escalier Biosciences - employee, stock options; Prometheus Biosciences (merger of Precision IBD and Prometheus Laboratories) - employee, stock options; Ventyx Biosciences -employee, stock options; Vimalan Biosciences - employee, stock options.

References

1.Sandborn W.J. et al. Gastroenterology. 2020; 158(3): 562–572. [DOI] [PubMed] [Google Scholar]
2.Vermeire S. et al. J Crohn's Colitis. 2020; 14(suppl 1): S020–S021. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.85

OP091 Size Matters: Is Ai Using Computer Vision Better Than Humans in Sizing Colonic Polyps?

M Abdelrahim ^1,^✉, H Saiga ², N Maeda ², E Hossain ¹, S Arndtz ³, P Bhandari ⁴

Introduction

Polyp size is an important biomarker that influences decisions on therapy and surveillance during colonoscopy. The 5 mm cut-off remains crucial for implementing resect and discard strategy. However current evidence shows significant inter observer variation between endoscopists in estimating polyp size, so this remains an unmet clinical need. We developed an automated system using computer vision technique for binary classification of polyp size in an experimental setting, and compared its performance to that of endoscopists of various levels of experience.

Aims & Methods

Artificially made pre measured polyps of different sizes were fixed into a pig colon model and used for this experiment. The pig models were examined with colonoscope, the procedure was recorded and the videos used for this experiment. For the system assisted sizing, videos were automatically annotated to mark polyps using specially designed software. We developed a computer vision system for size classification. in a triangulation based approach, distance between camera positions was calculated using structure from motion (see figure 1). Sizing was estimated using sequential frames, with 300 frames of each video to improve classification accuracy. For the endoscopist's sizing, we asked 10 endoscopists of varying experience to review the same 22 videos and categorise polyps as either ≤5mm or >6mm. Mean diagnostic accuracy was calculated and compared between the two groups using T test.

Results

Twenty two (22) polyps were video recorded and equally divided into two categories for polyp ≤5mm or >6mm. Overall, average diagnostic accuracy of the computer model was 85.2%, compared to 59.5% in the endoscopist group (P < 0.0001). in polyps’ ≤5mm, computer vision and endoscopists showed diagnostic accuracy of 81.2% and 66% respectively. Whereas, in polyps >6mm accuracy was 87.5% and 42.3% for computer vision model and endoscopists respectively. Table 1 summarizes results.

Conclusion

Our computer vision system is significantly better than endoscopists in binary classification of colonic polyps based on their size in an experimental setting. This model will be tested in real time human colonic examination to validate these results. To our knowledge, this is the first report examining computer assisted polyp sizing during endoscopy. This can form a basis for implementing resect & discard strategy in the AI world where lesion characterisation is also proving to be superior to humans.

Disclosure

Nothing to disclose

Table 1.

[shows average diagnostic accuracy of computer vision model and endoscopists in classifying polyps as either ≤5mm or ≥6mm.]

	polyps ≤5mm	polyps ≥6mm	Overall (all polyps)
Computer vision model	81.2%	87.5%	85.2%
Endoscopists	66%	42.3%	59.5%
P value	P < 0.0001	P < 0.0001	P < 0.0001

Open in a new tab

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.86

OP092 Machine Learning Models For The Prediction of Risk of Gastric Cancer After H. Pylori Eradication Therapy

WK Leung ^1,^✉, KS Cheung ¹, TK-L Lui ¹

Introduction

Gastric cancer is still the leading cause of cancer deaths in the world, particularly in East Asian countries. Although eradication of H. pylori (HP) could reduce the risk of gastric cancer development, individuals could still progress to gastric cancer.

Aims & Methods: Aim

To evaluate the performance of three different machine learning (ML) models on the prediction of gastric cancer risk in HP infected subjects after completion of HP eradication therapy.

Methods

We retrieved data from a territory-wide public health database which included all HP infected patients who had received their first course of clarithromycin-based triple therapy for HP between year 2003 and 2014. We excluded patients who developed gastric cancer within the first 12 months after HP therapy and those with past history of gastric cancer or surgery. Patients were divided into two cohorts with the first cohort (2003-2010) used as the training set (n=64,238) and the second cohort (2011-2014) as the validation set (n=25,330). The training set data was used to construct 3 machine learning (ML) models including Lasso regression (Lasso), multiple logistic regression (LG) and sparse neural networks (sNNR), to identify patients who developed gastric cancer (GC) within 5 years after the HP eradication therapy. A total of 26 clinical variables including patient's baseline characteristics, age of HP eradication, smoking and drinking habits, concurrent medical illness and medication uses (but not endoscopic and histologic parameters) were input into the ML models. The performances of the 3 ML models were determined by the area under receiver operating characteristic curve (AUC) analysis. The AUC was compared using Delong's test and the best cutoff point for each ML model was estimated by Youden's method.

Results

A total of 64,238 HP eradicated subjects were included in the training set and 135 (0.21%) developed GC with a mean follow-up of 4.7 years. The performance (AUC) of the sNNR in identifying GC in the independent validation set (54 GC, 0.21%) with mean follow-up 4.7 years was 0.95 (95%CI 0.94-0.96), which outperformed the two other ML models (Lasso: 0.94, p< 0.01 and LG: 0.90, p < 0.01). At the best cut-off, the sensitivity and specificity of the sNNR model was 96.3% (95%CI, 91.3-100%) and 92.4% (95%CI, 92.1-92.8%). The sNNR model was superior to the other two models in terms of sensitivity (96.3% Vs 92.6% [Lasso] or 75.9% [LG], p< 0.01) and positive predictive value (2.7% vs 1.6% [Lasso] or 2.1% [LG], p< 0.01). The negative predictive value (NPV) of sNNR model was 99.9% (95%CI, 99.9-100%).

Conclusion

Based on simple baseline patient's information, we have developed an accurate ML model that can predict the risk of gastric cancer after HP eradication with high sensitivity and NPV. This model is particularly useful in identifying low risk patients who may not require subsequent surveillance.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.87

OP094 Unexperienced Endoscopists Can Reach Expert Level in Detecting and Characterizing Colorectal Polyps By Using A Validated Polyp Detection and Characterization System

J Weigt ^1,^✉, H Neumann ², A Repici ³, C Hassan ⁴

Introduction

Despite great efforts to increase the detection of polyps still approximately twenty percent of the polyps are missed. Recent advances in image processing lead to the development of real time computer aided diagnosis systems in colonoscopy.

Aims & Methods

We aimed to evaluate a system that is able to perform polyp detection and polyp characterisation using artificial intelligence. Patients who underwent colonoscopy for any reason despite chronic or acute inflammatory changes or acute bleeding were eligible for inclusion. Images and video files for training and validation were collected in a multicenter study using the same endoscopic equipment and settings in all centres. Images and video files were annotated according to the investigators findings during colonoscopy and according to the histological reports of removed polyps.

After validation of the CAD System (CAD-Eye, Fujifilm, Jpn), we investigated the diagnostic yield in an image based experiment. in total 3 experts (>10.000 colonoscopies), and 3 non advanced (<100 colonoscopies) endoscopists participated in the experiment. Images were presented parallel to the participants from a single computer which showed the same images for 5 seconds to all participants. Images contained BLI, LCI and WLE images of different lesions.

Participants had to annotate the images as polyp included in the image or no polyp (n= 1634 images/ participant). in a second setting images had to be graded if the presented polyp was neoplastic or non-neoplastic (n= 402 images/ participant).

In total three experimental settings were carried out for each, detection and characterization: CAD Eye alone, Expert alone and non-Experts using CAD Eye.

The system interacts by using overlay information to the endoscopic image or video in real time and gives an acoustic signal.

Results

The analysis of training and validation of the system demonstrated homogeneous data distribution.

In the polyp detection experiment in WLE mode, the experts alone had a sensitivity of 95.1% and a specificity of 94.2% while the CAD-EYE alone had a sensitivity and specificity of 92.9% and 90.6 % respectively. The non experts together with the CAD Eye reached sensitivity of 94.8% and specificity of 76,4%.

Using LCI for detection, experts allone reached 95.8% and a specificity of 95.6% while the CAD-EYE alone had a sensitivity and specificity of 94.1% and 94,9 % respectively. The non experts together with the CAD Eye reached sensitivity of 96,4% and specificity of 79,7%. in the characterization experiment the accuracy using WLE was 75.2%, 84,2% and 77,7% for experts, CAD EYE and non-experts using CAD EYE. For BLI accuracy was 79.4%, 83.6% and 80.1% respectively. The negative predictive values (NPV) for the detection of polyps were 0.94, 0.76 and 0,91 for experts, non-experts and CAD-Eye alone using WLE. Using LCI the NPV increased in all subgroups to 0.96, 0.93 and 0.95 respectively.

Conclusion

Our initial study shows the beneficial use of the CAD-Eye system for polyp detection and classification and implicates the use in both, non-experts and experts. Clinical trials are needed to evaluate the impact on patient care.

Disclosure

Study spopnsored by Fujufilm Jpn.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.88

OP096 Colon Capsule Endoscopy (Cce) Is An Effective Filter Test For Colonic Polyp Surveillance

S Semenov ^1,^2,^✉, T Manoharan ², MS Ismail ^1,², S Sihag ^1,², K Hazel ^1,³, D Molloy ^1,², B Ryan ², N Breslin ², A O'Connor ², S O'Donnell ², D McNamara ^1,²

Introduction

Surveillance accounts for 30% of colonoscopy workload, the majority are normal. Identifying patients who require polypectomy would be advantageous.

Aims & Methods

We aimed to assess CCE and/or faecal immunochemical test (FIT) as a potential filter in surveillance.

Following ethical approval, patients due for polyp surveillance, aged 18-80 were identified from our waiting list, then invited by post for CCE and FIT. A CCE was considered positive if polyps or CRC was identified. CCE significant lesions (>3 or >6mm polyps), incomplete studies and positive FITs (>45 μg/g) were referred for endoscopy. CCE and endoscopy results were correlated. FIT accuracy was compared to CCE.

Results

To date, 300 surveillance patients have been invited, the uptake rate was 46% (138/300).

Of 114 analysed CCEs (mean age 65 (31-80), 62 (54%) males), 70% (80/114) were complete. Image quality was adequate in 103 (90%). Overall, CCE positivity was 70% (80/114) with 54% (43/80) having significant polyps. PPV of CCE was 94% (30/32 completed endoscopies). Significant extra-colonic findings, requiring investigation, were reported in 1.8% (2/114). There were no complications.

The only characteristic associated with a positive CCE was older age >70 (OR 2.7, p=0.04, 95%CI 1.0547 to 6.9537).

2/98 (2%) returned FITs were positive, range 0-279 μg/g and mean 9 μg/g. (+)FIT and CCE concordance was 100%, but FIT sensitivity was inadequate (3%) with an NPV of 30%. ROC analysis gave a sensitivity and specificity of 17% and 93%, respectively, for a FIT of >10 μg/g.

In all, 70/114 (61%) were referred for endoscopy, 18 (16%) sigmoidoscopy and 52 (46%) colonoscopy, 39 (34%) for an inadequate CCE. in our surveillance cohort, CCE reduced the need for any endoscopy by almost 40% (44/114) and 54% (62/114) were spared a colonoscopy.

Conclusion

Unlike FIT, CCE is useful in selecting patients for polypectomy in polyp surveillance and avoiding unnecessary colonoscopy.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.89

OP098 Efficacy of A Thin- Versus Thick-Wire Diameter Snare For Cold Snare Polypectomy - A Multi-Center Randomised Controlled Trial

M Sidhu ^1,^2,^✉, N Forbes ³, DJ Tate ⁴, L Desomer ⁵, A van Hattem ¹, S Vosko ⁵, N Shahidi ¹, NG Burgess ⁷, E Cheng ⁸, A Schell ⁹, R Hilsden ³, SJ Heitman ¹⁰, MJ Bourke ^11,²

Introduction

Cold snare polypectomy (CSP) has been shown to be effective and safe for the removal of small adenomas (< 10mm), resulting in less delayed post-polypectomy bleeding compared to snare polypectomy with cautery. Despite the development of specialised thin-wire snares, data comparing these devices for CSP with traditional thick-wire snares is scarce.

Aims & Methods

We sought to assess the efficacy of a thin-wire snare compared to a thick-wire snare for CSP. A prospective, international, multi-centre randomised trial was performed (NCT02581254). The primary endpoint was the incomplete resection rate (IRR). All patients referred for colonoscopy were eligible. Exclusion criteria included lesions with evidence of submucosal invasion at colonoscopy. Upon detection of a suitable polyp (< 10 mm and beyond the rectosigmoid), we performed 1:1 randomisation to use of a thin-wire (0.30 mm) or thick-wire (0.47 mm) snare for CSP. Only one polyp per patient was randomized. For patients with multiple polyps, the study polyp was the first polyp encountered. After complete endoscopic resection was achieved, two biopsies were performed from either side of the defect margin and sent to histopathology for analysis as separate specimens. All patients were contacted at 30 days to assess for adverse outcomes. Binary variables were assessed for significance using the chi-squared statistic (p< 0.05).

Results

Over 50 months to November 2019, 1195 patients were enrolled. 578 patients had no polyp detected. 617 patients (57.5% male) with polyps (61.8% - adenomas, 9% - serrated) were then randomised to CSP with either thin-wire (n=316) or thick-wire (n=301) snare. Patient and procedural characteristics were similar between the groups (Table 1). Overall complete endoscopic resection was achieved in 99.8% (616/617). Marginal biopsies of the polypectomy defect were successfully performed in 609/617 cases (98.9%). The IRR for thin- and thick-wire snares was comparable at 1.4% versus 2.7% [RR - 0.52,95% CI (0.13-0.89),p = 0.21]. Only one case of clinically significant post endoscopic bleeding was observed in the thin-wire snare arm. No other adverse events occurred in the cohort.

Table 1.

[Lesion characterstics and procedure outcomes; IRR - Incomplete Resection Rate (after exclusion of normal colonic mucosa at histology).]

	Thin n=316	Thick n=301	Total n=617	p
Mean Age, years (SD)	60.3 (8.9)	59.6 (10.4)	60 (9.7)	0.68
Male Sex, n (%)	184 (58.2)	171 (57.5)	355 (57.5)	0.72
Mean Bowel Preparation Score - BBPS (SD)	7.6 (1.3)	7.5 (1.4)	7.6 (1.4)	0.46
Right Colon, n (%)	93 (29.4)	96 (31.9)	189 (30.6)	0.16
Defect Protrusion, n (%)	83 (26.3)	101 (33.7)	184 (29.9)	0.05
IRR*, n (%)	4/282 (1.4%)	7/257 (2.7%)	11/539 (2)	0.21
Histology - TA/TVA, n (%)	193 (68.4) / 7 (2.5)	173 (67.3) / 4 (1.6)	366 (67.9) / 11 (2)	0.41
Major DMI [1], n (%)	0 (0)	0 (0)	0	-
Delayed Bleeding, n (%)	1 (0.3)	0 (0)	1 (0.2)	0.40

Open in a new tab

Conclusion

In this study, we demonstrated comparable efficacies of thin-and thick-wire snares when used for CSP of small polyps with minimal adverse events. Incomplete resection rates were lower than previously reported for both study arms. Therefore, optimal operator technique, is more important than snare design alone in CSP.

Disclosure

Bourke, MJ - Research support from Olympus, Boston, Cook

References

1.Burgess et al. Gut, 2016 [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.90

OP100 Comparison of Cold Biopsy Forceps Vs Cold Snare For Diminutive Colorectal Polyp Removal: A Multicenter Randomized Controlled Trial

G Perrod ^1,^✉, E Perez-Cuadrado-Robles ¹, E Coron ², M Pioche ³, N Etchepare ⁴, D Danan ¹, A Becq ⁵, N Musquer ⁶, X Dray ⁷, A Laquière ⁸, B Jais ⁹, C Broudin ¹⁰, H Benosman ¹, C Cellier ¹¹, G Rahmi ¹¹

Introduction

European guidelines recommends the use of cold snare polypectomy (CSP) for removal of diminutive colorectal polyps (≤ 5 mm) (DCP) (1). However, for DCP < 4 mm cold biopsy forceps (CBF) may be optional, especially when CSP is technically difficult. in this study, we aimed to compare the efficacy of CSP with CBF for removal of DCP in routine colo-noscopy.

Aims & Methods

From January 2017 to January 2019, we conducted a multicenter randomized controlled trial in 8 French tertiary care centers. of the 138 patients screened, 123 were prospectively included and 180 DCPs were removed by either CBF or CSP after randomization (1:1). The primary end-point was the histological complete resection rate defined by negative additional biopsies taken from the edge of the polypectomy site. The secondary end-points were the procedure time and the rates of tissue retrieval and adverse event.

Results

Among the 180 DCPs, 121 (67.2%) adenomas or sessile serrated

lesions were considered for the analysis. The median size was 4 (1-5) mm. Most of lesions 55.4% (67/121) were flat (Paris classification 0-IIa or 0-IIb) and located in the proximal colon 44.6% (54/121). Characteristics of DCPs are summarized in table 1. The overall complete resection rate was 91.5% (108/121). En bloc resection rate was superior in the CSP group (91.7%) as compared to the CBF group (42.6%); (p< 0.001). However, the complete resection rates were similar between the CSP and CBF groups (93.3% vs 90.2%; p=0.527), including for polyps < 4 mm (94.4% vs 89.2% p=0.234). Moreover, there was no difference according to histological type, Paris classification or localization in terms of complete resection rate. All specimens were retrieved and there was no difference between the 2 groups in terms of procedure times and adverse events. Finally, Univariate analysis did not identify any potential factor associated with complete resection rate.

Table 1.

[Polyps characteristics of patients presenting with at least one polyp<6mm and a final diagnosis of adenoma or sessile serrated lesion (SSL).]

	Overall n=121	Cold biopsy forceps n=61	Cold snare polypectomy n=60	p value
Size, mm, median (range)	4 (1-5)	4 (2-5)	4 (1-5)	0.279
Size, mm, n (%)				0.943
- 4-5 mm	46 (38.02%)	23 (37.7%)	23 (38.33%)
- 1-3 mm	75 (61.98%)	38 (62.3%)	37 (61.67%)
Paris classification, n (%)				0.776
- 0-Is	54 (44.63%)	28 (45.9%)	26 (43.33%)
- 0-II	67 (55.37%)	33 (54.1%)	34 (56.67%)
Localization, n (%)				-
- Caecum/ Ascending colon	54 (44.63%)	29 (47.54%)	25 (41.67%)
- Transverse colon	28 (23.14%)	14 (22.95%)	14 (23.33%)
- Descending colon	15 (12.4%)	6 (9.84%)	9 (15%)
- Sigmoid colon	19 (15.7%)	11 (18.03%)	8 (13.33%)
- Rectum	5 (4.13%)	1 (1.64%)	4 (6.67%)
Histology, n (%)				0.371
- Adenoma	91 (75.21%)	48 (78.69%)	43 (71.67%)
- SSL	30 (24.79%)	13 (21.31%)	17 (28.33%)
Tissue retrieval failure, n (%)0	0 (0)	0 (0)	0 (0)	-

Open in a new tab

Conclusion

In this multicenter randomized controlled trial, complete histological resection was achieved equally by CBF and CSP in diminutive colorectal polyps. Therefore, CBF should be considered for DCP removal in routine practice as an alternative to CSP.

Disclosure

G. Perrod, C. Broudin, D. Danan, H. Benosman, E. Perez, N. Etchepare, B. Jais, A. Becq. A. Laquiere: none G. Rahmi: personal fees from FUJIFILM, MEDTRONIC, grants from NORGINE

References

1.Ferlitsch M., Moss A., Hassan C. et al. Colorectal polypec-tomy and endoscopic mucosal resection (EMR): European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline. Endoscopy 2017; 49: 270–297. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.91

OP101 Quality Measures For Screening Colonoscopy and Colorectal Cancer Risk

P Wieszczy ^1,^✉, M Bugajski ², W Januszewicz ¹, MJ Rupinska ¹, R Franczyk ², J Szlak ², M Pisera ¹, M Turkot ¹, M Rupinski ², U Wojciechowska ³, J Didkowska ³, J Regula ^1,², MF Kaminski ^1,⁴

Introduction

Post-colonoscopy colorectal cancer (PCCRC), i.e. cancer diagnosed between colonoscopy and recommended surveillance is considered as one of the major failures of colorectal cancer screening. Several quality measures were suggested as possible mediators of low PCCRC risk, including adenoma detection rate (ADR), polyp detection rate (PDR), advanced adenoma (>10 mm/villous component/ high-grade dysplasia) detection rate (AADR), number of adenomas per colonoscopy (nADR) and number of adenomas per colonoscopy with >1 adenoma (nADR+). So far, only ADR was reported to be associated with PCCRC risk (1, 2).

Aims & Methods

The aim of the study was to estimate association between ADR, PDR, AADR, nADR, nADR+ and PCCRC risk. We used data from the Polish CRC Screening Program from 2000-2011. We calculated yearly quality measures for all endoscopists with >100 colonoscopies. We followed their patients through the National Cancer Registry until Dec 31st, 2017. Follow-up time was censored at recommended surveillance - 3, 5 or 10 years in high, low and no-adenoma group, respectively (3). We estimated hazard ratios (HR) for PCCRC and 95% confidence intervals (CI) using Cox proportional-hazard models for quality measures (categorized into quintiles). HRs were adjusted for patients’ age, sex, family history of CRC and colonoscopy year. We calculated the risk of missed PCCRC as a number of PCCRCs per 100,000 person-years (p-y) of follow-up.

Results

Data of 192,067 patients and 262 endoscopists was used to calculate quality measures. For the analysis of PCCRC risk we excluded 18,780 (10%) patients because of colorectal cancer diagnosed at screening (1%), incomplete colonoscopy (4%) or poor bowel preparation (5%). Among remaining 173,287 patients median age was 56 years (range 40-66) and 38% were males. During the 1,407,416 p-y of follow-up we identified 347 PC-CRCs (25 per 100,000 p-y).

Median ADR was 17.9% (quintiles, 12.1, 15.9, 19.9, 24.8), median PDR was 34.6% (quintiles 19.9, 27.5, 34.4, 42.6), median AADR was 6.3% (quintiles 4.1, 5.4, 6.8, 9.0), median nADR was 0.24 (quintiles 0.15, 0.20, 0.27, 0.36) and median nADR+ was 1.33 (quintiles 1.19, 1.27, 1.39, 1.53).

All quality measures were associated with PCCRC risk. Reaching 4th or 5th quintile of ADR, PDR and nADR reduced the PCCRC risk by 50 to 70% (see Table).

For high-quality defined as >4th quintile, missed PCCRC risk per 100,000 p-y (95% CI) was similar for ADR (15; 12-19), PDR (17; 14-22), nADR (15; 12-19) and nADR+ (18; 15-23). For AADR >4th quintile the risk was significantly higher than for ADR and nADR (23; 20-28).

Conclusion

Both relative and absolute reduction of PCCRC risk was observed for ADR, PDR, nADR and nADR+. PDR should be considered an alternative to ADR if histology results are unavailable. Despite clinical relevance,

nADR and nADR+ were not associated with better discrimination of PCCRC risk than ADR. Although significant relative reduction in PCCRC risk for high AADR was observed, high absolute risk of missed PCCRC indicates poor discrimination abilities of this parameter.

[Number of PCCRCs and HRs for quintiles of quality measures]

Quintile		ADR	PDR	AADR	nADR	nADR+
1st (ref.)	PCCRC	103	115	84	107	106
2nd	PCCRC	86	83	77	90	92
	HR (95% CI)	0.84 (0.64-1.11)	0.75 (0.57-0.98)	0.83 (0.59-1.17)	0.81 (0.61-1.06)	0.76 (0.58-0.99)
3rd	PCCRC	85	64	55	74	62
	HR (95% CI)	0.82 (0.63-1.06)	0.62 (0.46-0.85)	0.67 (0.47-0.97)	0.74 (0.55-0.99)	0.58 (0.42-0.80)
4th	PCCRC	42	56	82	52	40
	HR (95% CI)	0.48 (0.33-0.69)	0.57 (0.40-0.82)	0.86 (0.60-1.21)	0.56 (0.40-0.79)	0.51 (0.33-0.78)
5th	PCCRC	31	29	49	24	47
	HR (95% CI)	0.39 (0.24-0.63)	0.35 (0.24-0.52)	0.61 (0.43-0.86)	0.30 (0.18-0.50)	0.57 (0.36-0.91)

Open in a new tab

Disclosure

MFK: speaking/consultation fee from Olympus, Fujifilm, Boston Scientific, Medtronic, Alfa Sigma, Norgine, ERBE

References

1.Kaminski M.F., Regula J., Kraszewska E., Polkowski M., Wojciechowska U., Didkowska J. et al. Quality indicators for colonoscopy and the risk of interval cancer. N Engl J Med. 2010; 362(19): 1795–803. [DOI] [PubMed] [Google Scholar]
2.Corley D.A., Jensen C.D., Marks A.R., Zhao W.K., Lee J.K., Doubeni C.A. et al. Adenoma Detection Rate and Risk of Colorectal Cancer and Death. N Engl J Med. 2014; 370(14): 1298–306. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Winawer S.J., Zauber A.G., Fletcher R.H., Stillman J.S., O'Brien M.J., Levin B. et al. Guidelines for colonoscopy surveillance after polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer and the American Cancer Society. Gastroenterology. 2006; 130(6): 1872–85. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.92

OP102 Association Between Colonoscopy Quality Indicators, Diagnostic Yield, and Post Colonoscopy Colorectal Cancer Incidence in Lynch Syndrome

A Sánchez ^1,^✉, VH Ross ², M Navarro ³, MP Pineda ³, B Caballol Oliva ¹, L Moreno ¹, S Carballal ¹, L Rodríguez-Alonso ⁴, RYC Teresa ⁵, G Llort ⁶, C Romero ⁶, V Piñol ⁷, A Lopez ⁸, I Salces ⁹, MD Pico Sala ¹⁰, L Rivas ¹¹, L Bujanda Fernández de Piérola ¹², M Garzon ¹³, Moreno E Pizarro ¹³, E Martinez de Castro ¹⁴, MJ López-Arias ¹⁴, C Poves ¹⁵, C Garau Colom ¹⁶, D Rodríguez-Alcalde ¹⁷, M Herraiz ¹⁸, C Alvarez Urturi ¹⁹, X Bessa ¹⁹, A Dacal Rivas ²⁰, M Carrillo-Palau ²¹, L Cid-Gómez ²², M Ponce ²³, A Suárez González ²⁴, E Saperas Franch ²⁵, E Aguirre ²⁶, B Morales-Romero ¹, T Ocaña ¹, L Rivero Sánchez ¹, G Jung ¹, J Cubiella ¹¹, R Jover ¹⁰, F Rodriguez-Moranta ⁴, J Balmana ⁸, A Castells ¹, G Capellà ³, M Serra-Burriel ²⁷, L Moreira Ruiz ¹, M Pellisé Urquiza ¹, F Balaguer ¹

Introduction

Lynch syndrome(LS) is the most common inherited cause of colorectal cancer(CRC). Although colonoscopy has shown to reduce CRC incidence and mortality, high incidence of post-colonoscopy CRC(PCCRC) has been recently described. Variation in quality has emerged as an important issue for colonoscopy. Quality of surveillance colonoscopy in LS has not been properly evaluated.

Aims & Methods

Evaluate the impact of colonoscopy quality indicators on adenoma detection and PCCRC prevention in LS. We conducted a multicenter retrospective study(2016-2019) including LS carriers from 26 Spanish high-risk clinics and 1 from The Netherlands. The current study focused on LS healthy carriers with at least 2 colonoscopies, excluding patients with a CRC diagnosed prior or in the first colonoscopy. We considered colonoscopy quality indicators:cecal intubation, bowel preparation, type of scope definition, and time-interval between colonoscopies. The use of pan-chromoendoscopy was evaluated as an enhancement technique.

We emulated a target trial comparing the results from the first and second surveillance colonoscopies on adenoma and PCCR diagnosis between each quality indicator. Risk analyses were conducted using a multi-variable logistic regression model. Supplementary analysis to assess factors associated with PCCRC risk were performed. By logistic regression we compared characteristics between carries who did and did not developed PCCRC; and between carriers who underwent all their surveillance colo-noscopies accomplishing high-quality standards with those who did not.

Results

We included 893 LS healthy carriers under surveillance. Ten-year cumulative incidence of adenoma and PCCRC was 60.6%(95%CI; 55.5-65.2%) and 7.9%(95%CI; 5.2-10.6%), respectively. Adequate bowel preparation[OR=2.07(95%CI 1.06-4.3)] and ceccal intubation(20%vs0%;p=.01) were significantly associated with improved adenoma detection, as well as chromoendoscopy use[OR=2.14;95%CI 1.15-3.95]. Non-advanced adenomas and < 1cm were better detected with all quality indicators, including the use of high-definition scopes [OR=2.18(95%CI 1.13-4.2) and OR=1.99(95%CI 1.05-3.76); respectively]. Flat lesions were more frequently detected with high-definition scopes[OR=3.04(95%CI 1.13-8.65)] and chromoendoscopy[OR=6.44;95%CI 2.29-19.59]. Conversely, no association was found for advanced adenomas or adenomas ≥1cm. Importantly, PCCRC risk was significantly lower when colonoscopies were performed in less than 3-year interval[OR=0.35;95%CI 0.14-0.97]. We observed a consistent but no significant reduction in PCCRC risk for previous complete examination[OR=0.16;95%CI 0.03-1.28]; previous adequately prepared examination[OR=0.64;95%CI 0.17-3.24] and previous use of a high-definition scope [OR=0.37(95%CI 0.02-2.33[FB1])]. Moreover, MLH1 or MSH2carriers[OR=2.45;95%CI 1.06-5.62;p=.04)] and an advanced adenoma at any colonoscopy[OR=2.42;95%CI 1.22-4.8;p=.01)] were indepent factors associated to PCCRC. Performing all colonoscopies with a >3-year interval[OR 5.17;95%CI 2.18-12.25;p< .001]and the use of standard definition scopes[OR=2.77;95%CI 1.44-5.31;p=.002] in all procedures were associated with PCCRC development; whereas performing all examinations complete was a protective factor[OR=0.24;95%CI 0.11-0.5;p< .001]. Finally, the PCCRC risk was consistently lower in carriers under high-quality surveillance(2.8vs 7.1%, p< .001 all colonoscopies complete, adequately prepared and < 3-years interval).

Conclusion

Colonoscopy quality indicators in LS influences CRC prevention. High-quality colonoscopy is of outmost importance in this high-risk condition.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.93

OPT03 Quality of Colonoscopy Per Indication in A Large Nationwide Dutch Colonoscopy Cohort

KJ Nass ^1,^2,^✉, M van der Vlugt ¹, A van der Beek ³, AAJ van Esch ⁴, T Hummel ⁵, MM Lacle ⁶, M Ledeboer ⁷, ME van Leerdam ^8,⁹, RJT Ouwendijk ¹⁰, PJ van der Schaar ¹¹, MCW Spaander ¹², M Wouters ^1,^2,³, P Fockens ¹, E Dekker ¹

Introduction

Over the past decade, several quality indicators have been developed to optimize the performance of colonoscopies. These indicators are usually monitored in a heterogeneous population without distinction between different indications for colonoscopy, or in a specific subgroup such as a screening population. This study aims to assess the variation in quality of colonoscopy and complication rates according to different indications, using prospective data from the national colonoscopy registry of endoscopy services and linkage to the national complication registry of endoscopies in the Netherlands.

Aims & Methods

This retrospective study was conducted with prospec-tively collected data of the Dutch Gastrointestinal Endoscopy Audit (DGEA) and the Dutch Registration of Complications in Endoscopy (DRCE). Patient and endoscopy characteristics of all colonoscopies performed in the Netherlands are automatically extracted from the endoscopy reporting system in participating endoscopy services and recorded in the DGEA dataset. The DRCE is a prospective web-based national complication registry for endoscopy. Complications occurring within 30 days after colonoscopy are manually recorded by endoscopists in the DRCE. DGEA and DRCE data between 01-01-2016 and 31-12-2018 were analyzed. Colonoscopy data was used from endoscopy services which were simultaneously participating in both registries. in order to calculate complication rates, data were linked to the level of endoscopy service and indication. Indications for colonoscopy were categorized in four different groups: diagnostic, therapeutic, FIT-positive screening participants and surveillance colonoscopies. Adequate bowel preparation was defined as a Boston Bowel Preparation Scale of > 6, with a minimum score of 2 per segment.

Results

During the 3-year study period, 266.981 colonoscopies were recorded in the DGEA by endoscopy services which were simultaneously participating in the DGEA and DRCE (median age at colonoscopy: 64 years, IQR: 55 - 71 years; 50.3% male). Outcomes of quality indicators and complication rates per indication are described in Table 1.

Table 1.

[Outcomes of quality indicators and complication rates related to the indication for colonoscopy.]

Indication for colonoscopy	Overall n = 266.981	Diagnostic n = 133.462	Therapeutic n = 9990	Screening n = 55.904	Surveillance n = 67.625	p value
Successful cecal intubation, n (%)	244.701 (92)	120.489 (90)	7353 (74)	54.085 (97)	62.774 (93)	<0.001
Adequate bowel preparation, n (%)	240.702 (90)	118.687 (89)	7152 (72)	53.160 (95)	61.703 (91)	<0.001
Overall complications, n (%)	1540 (0.58)	515 (0.39)	273 (2.73)	532 (0.95)	220 (0.33)	<0.001
Perforation, n (%)	173 (0.06)	60 (0.04)	51 (0.51)	32 (0.06)	30 (0.04)	<0.001
Bleeding, n (%)	939 (0.35)	257 (0.19)	156 (1.56)	401 (0.71)	125 (0.18)	<0.001

Open in a new tab

Conclusion

This study describes the first results of a large cohort from the national colonoscopy registry in the Netherlands, which is generated without any additional effort for the endoscopists and successfully linked with data of the national complication registry of endoscopies. Overall quality

of colonoscopies performed in the Netherlands was high and had a low risk of complications. As expected, in therapeutic and screening colonoscopies complication rates were higher than in diagnostic and surveillance colonoscopies. Lower rates for successful cecal intubation and adequate bowel preparation in therapeutic colonoscopies might be caused by the fact that for therapeutic colonoscopies cecal intubation is not always required.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.94

OP104 Colonoscopy Quality Assurance in An Organized Fit-Based Colorectal Cancer Screening Program

MCW Spaander ¹, PHA Wisse ^2,^✉, S De Boer ³, B den Hartog ⁴, M Oudkerk Pool ⁵, J Terhaar sive Droste ⁶, C Verveer ⁷, FG van Maaren-Meijer ⁸, E Dekker ⁹

Introduction

In the Netherlands, average risk individuals aged 55-75 years are invited for colorectal cancer (CRC) screening by biennial fecal immunochemical test (FIT). Positive FIT is followed by colonoscopy. High quality of colonoscopy is essential for optimal performance of a screening program. Therefore, endoscopists performing within the Dutch CRC screening program have to receive accreditation and fulfill minimum standards. in this study we assessed the quality of colonoscopies performed by the certified endoscopists.

Aims & Methods

Structured data on endoscopy and pathology is collected within the Dutch CRC screening program and stored in a national database. For this study, data were obtained for all first colonoscopies after a positive FIT within the first five years of the national program (2014 - 2018). Quality indicator performance was assessed for each endoscopist with >50 procedures.

[Table. Overview of quality indicators and minimum standards. Only adverse events registered in 2017 and 2018 are included.]

Quality criteria	Description	Minimum standard
(Unadjusted) cecal intubation rate	Percentage of procedures which reached cecum	≥95%
Bowel preparation	Percentage of procedures with a bowel sufficient clean for inspection (Boston Bowel Preparation Scale ≥6)	≥90%
Withdrawal time	Percentage of procedures with inspection time ≥6 min (only procedures without detection of lesions are included)	≥90%
Cancer detection rate	Percentage of procedures with detection of colorectal cancer	Monitoring
Adenoma detection rate	Percentage of procedures with detection of adenoma	≥30%
MAP	Mean number of adenomas per procedure	Monitoring
MAP+	Mean number of adenomas per positive procedure (only procedures with at least detection of one adenoma are included)	Monitoring
Polyp removal rate	Percentage of procedures without second colonoscopy to remove a detected lesion	≥90%
Comfort score	Percentage of procedures with moderate/severe discomfort for the patient	Monitoring
Adverse events	Rate of bleeding, perforation or other event leading to prolonged hospital stay (categorized as mild, moderate, severe or fatal event)	Monitoring

Open in a new tab

We determined quality indicators regarding completeness of colon visualization (cecal intubation rate, bowel preparation, and withdrawal time), detection rates (CRC detection rate, adenoma detection rate (ADR), mean number of adenomas per procedure, and mean number of adenomas per positive procedure), removal rates (percentage of patients that had to undergo second colonoscopy for removal of a polyp), patient satisfaction (Gloucester comfort score) and adverse events. in colonoscopies in which no lesions were detected, withdrawal time was assessed.

Results

In total 431 endoscopists performed 237.092 first colonoscopies. in these colonoscopies, cecal intubation rate was 97.0%, bowel preparation was at least sufficient in 97.5% and withdrawal time was >6 minutes in 96.7%. CRC detection rate was 7.3% and ADR was 64.2%. Mean number of adenomas was 1.7 per procedure and 2.5 per positive procedure. in 96.4% of colonoscopies all polyps were removed. in 4.2% of procedures the patient experienced moderate or severe discomfort. Post-procedure complications occurred in 0.69% of procedures, in 75% due to postcolonoscopy bleeding. Severity of complications was mild or moderate in 92%. of all endoscopists, 401/431 (93.0%) performed >50 colonoscopies. in total 365 (365/401, 91.0%) endoscopists had an unadjusted cecal intubation rate >95%. Sufficient bowel preparation was achieved in >90% of patients for 390 endoscopists (390/401, 97.3%). Withdrawal time was >6 minutes in >90% of colonoscopies for 366 endoscopists (366/401, 91.3%). ADR was >30% for all endoscopists (401/401, 100%), moreover ADR was >40% for 399 endoscopists (399/401, 99.5%) and >50% for 396 endoscopists (396/401, 98.8%). in total 388 endoscopists (388/401, 96.8%) removed all polyps in >90% of colonoscopies.

Conclusion

Colonoscopies, performed after a positive FIT in the Dutch CRC screening program, are of high quality. All minimum standards are met by over 90% of endoscopists. ADR is much higher than the current minimum standard, which suggests to increase the minimum standard for ADR to at least >40%, but maybe to >50%, for optimal quality assurance in FIT-based CRC screening programs.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.95

OP105 Quality Indicators in Colonoscopies For Positive Fecal Immunochemical Tests (Fit) in Patients with Risk Factors For Colorectal Cancer (Crc) Before and After The Age of 50

R Kuvaev ^1,^2,^3,^✉, E Melnikova ⁴, I Maev ⁵, K Kryukov ⁴, T Kryukova ⁴, S Kashin ⁶, D Zavyalov ^1,³

Introduction

CRC represents the second commonest cause of cancer morbidity and mortality in Europe [1]. For CRC, it is recommended to implement the organized FIT-based screening programs for average-risk populations, targeting individuals aged between 50 and 75 years [2]. However, in high risk population the target group could be expanded to include individuals before 50.

The long-term efficacy in preventing CRC has been strongly associated with the quality of the colonoscopy [3]. However, little is known about the quality indicators of the FIT-based CRC screening program in different age patients with risk factors for CRC.

Aims & Methods

The aim of this study was to investigate the quality indicators of the FIT-based CRC screening program in different age groups of population with occupational exposures. A total of 1105 employees of oil refinery (age >37 years) were screened with FIT (ColonView, Biohit HealthCare, and Nadal, nal von minden GmbH).

Of 258 (23.3%) FIT-positive persons, 136 (53.3%) individuals (mean age 51.7± 7.2 years, 64 (47.1%) males, 72 (52.9%) females,) underwent colo-noscopy. We also reviewed control group of 204 colonoscopies (mean age 52.3 ± 7.4 years, 101 (49.5%) males, 103 (51.5%) females) performed for opportunistic CRC screening during the same period of time. Quality indicators (adenoma detection rate (ADR), advanced adenoma detection rate (AADR), sessile serrated adenoma detection rate (SSADR), polyp detection rate (PDR), adenoma per colonoscopy (APC), adenoma per positive participant (APPP), CRC detection rate) in FIT-positive group and control group were calculated.

Results

The results of the analysis are summarized in table 1. Overall ADR (32.4% vs 18.1%, p< 0.05), ADR in persons < 50 years (25.0% vs 11.1%, p< 0.05), overall PDR (50.7% vs 37.3%, p< 0.05), overall AADR (27.9% vs 18.6%, p< 0.05) in FIT positive group were significantly higher than the corresponding values for control group. in FIT positive group AADR in persons < 50 years was significantly lower than AADR in persons >50 years (17.3% vs 34.5%, p< 0.05). Other quality indicators had no significant differences in subgroups of FIT positive persons.

[Colonoscopy quality indicators in FIT-positive group and control group]

	FIT-positive group			Control group
Quality indicator	< 50 years (n=52)	≥ 50 years (n= 84)	Total (n=136)	< 50 years (n=108)	≥ 50 years (n=96)	Total (n=204)
ADR	25.0%	37.0%	32.4%	11.1%	26.0%	18.1%
AADR	17.3%	34.5%	27.9%	11.1%	27.0%	18.6%
SSADR	21.2%	27.4%	25.0%	22.2%	20.8%	21.6%
PDR	48.1%	52.4%	50.7%	37%	37.5%	37.3%
APC	0.27	0.56	0.45	0.13	0.31	0.22
APPP	1.08	1.52	1.39	1.17	1.20	1.18
CRC detection rate	1.9%	4.8%	3.7%	0%	1%	0.5%

Open in a new tab

Conclusion

FIT-based CRC screening for population with risk factors for CRC is effective in individuals of all age groups (including persons before the age of 50) and allows to reduce the number of unreasonably performed colonoscopies. Higher target standards of ADR, AADR and PDR are justified as quality indicators for colonoscopies in persons with positive FIT results.

Disclosure

Nothing to disclose

References

1.IARC. Available at: https://gco.iarc.fr/today/data/factsheets/cancers/10_8_9-Colorectum-fact-sheet.pdf
2.Sărftoiu A., Hassan C., Areia M. et al. Role of gastrointestinal endoscopy in the screening of digestive tract cancers in Europe: European Society of Gastrointestinal Endoscopy (ESGE) Position Statement. Endoscopy. 2020. Apr; 52(4): 293–304. doi: 10.1055/a-1104-5245.Epub2020Feb12. [DOI] [PubMed] [Google Scholar]
3.Kaminski M.F., Thomas-Gibson S., Bugajski M. et al. Performance measures for lower gastrointestinal endoscopy: a European Society of Gastrointestinal Endoscopy (ESGE) Quality Improvement Initiative. Endoscopy 2017; 49: 378–397 [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.96

OP106 Tofacitinib For The Treatment of Ulcerative Colitis: An Update On The Analysis of Malignancy Rates From The Ulcerative Colitis Clinical Programme As of May 2019

GR Lichtenstein ¹, MA Ciorba ², G Rogler ³, AI Sharara ⁴, N Sunna ⁵, R Mundayat ⁶, DT Judd ⁷, N Lawendy ⁷, J Panés ^8,^✉

Introduction

Tofacitinib is an oral, small molecule JAK inhibitor for the treatment of ulcerative colitis (UC). We present an updated [1] analysis of adjudicated malignancies in the tofacitinib UC clinical programme as of May 2019.

Aims & Methods

Malignancies were evaluated from 3 randomised, placebo-controlled studies (2 Phase [P]3 induction studies [NCT01465763; NCT01458951] and 1 maintenance P3 study [NCT01458574]), and an ongoing, open-label, long-term extension (OLE) study (NCT01470612). Three cohorts were analysed: Induction (P3 induction studies), Maintenance (P3 maintenance study) and Overall (patients [pts] receiving tofacitinib 5 or 10 mg twice daily [BID] in P3 or OLE studies). Analysis was by predominant dose (PD) 5 or 10 mg BID, based on average daily dose < 15 mg or ≥ 15

mg, respectively (82.4% of pts assigned PD 10 mg BID). Proportions and incidence rates (IRs; unique pts with events per 100 pt-years [PY] of exposure) were evaluated for malignancies (excl. non-melanoma skin cancer [NMSC]) and NMSC. An independent adjudication committee reviewed potential malignancies.

Results

1124 pts were evaluated for malignancies (2576.4 PY of tofaci-tinib exposure; ≤ 6.8 years of treatment). No malignancies (excl. NMSC) occurred in Induction Cohort pts. Malignancies (excl. NMSC) occurred in 1 placebo-treated Maintenance Cohort pt, and in 20 Overall Cohort pts (IR 0.75 [95% confidence interval (CI) 0.46,1.16]; PD tofacitinib 5 mg BID, n=3, IR 0.44 [0.09,1.29]; PD tofacitinib 10 mg BID, n=17, IR 0.86 [0.50,1.38]) (Table); 3 new cases since Sep 2018 [1]. NMSC events in the Induction and Maintenance Cohorts were previously reported (Table) [1]. NMSC occurred in 19 Overall Cohort pts (IR 0.73 [95% CI 0.44,1.13]; PD tofacitinib 5 mg BID, n=3, IR 0.45 [0.09,1.30]; PD tofacitinib 10 mg BID, n=16, IR 0.82 [0.47,1.34]) (Table); no new cases since Sep 2018 [1].

Conclusion

Malignancies (excl. NMSC) and NMSC occurred infrequently with tofacitinib treatment. There was no apparent clustering of types of malignancy (excl. NMSC). The IRs of malignancies (excl. NMSC) and NMSC were comparable to those previously reported in the tofacitinib UC clinical programme [1], and in pts with UC treated with biologics [2,3].

[Table. Demographics and summary of incidence of malignancies (all causality) among pts in the Induction,Maintenance and Overall Cohorts]

	Induction Cohort (8 weeks)^a		Maintenance Cohort (52 weeks)^a			Overall Cohort (≤6.8 years)
	Placebo (N=234; 38.2 PY)	Tofacitinib 10 mg BID (N=905; 151.2 PY)	Placebo (N=198; 100.4 PY)	Tofacitinib 5 mg BID (N=198; 146.2 PY)	Tofacitinib 10 mg BID (N=196; 154.3 PY)	Tofacitinib All (N=1124; 2576.4 PY)
Age (years), mean (SD)	41.1 (14.4)	41.2 (13.8)	43.4 (14.0)	41.9 (13.7)	43.0 (14.4)	41.2 (13.9)
Prior TNFi failure, n (%)	124 (53.0)	465 (51.4)	89 (44.9)	83 (41.9)	92 (46.9)	583 (51.9)
Prior immuno-suppressant treatment, n (%)	160 (68.4)	683 (75.5)	134 (67.7)	149 (75.3)	144 (73.5)	838 (74.6)
Prior immuno-suppressant failure, n (%)	158 (67.5)	661 (73.0)	129 (65.2)	143 (72.2)	141 (71.9)	813 (72.3)
Disease duration (years), mean (SD)	8.1 (7.0)	8.1 (7.0)	8.8 (7.5)	8.3 (7.2)	8.7 (7.0)	8.2 (7.0)
Malignancies (excl. NMSC), n (%), IR [95% CI]^b	0 (0.0), 0.00 [0.00,9.13]	0 (0.0), 0.00 [0.00,2.33]	1 (0.5), 0.97 [0.02,5.39]	0 (0.0), 0.00 [0.00,2.48]	0 (0.0), 0.00 [0.00,2.35]	20 (1.8), 0.75 [0.46,1.16]^c
NMSC, n (%), IR [95% CI]	0 (0.0), 0.00 [0.00,9.13]	2 (0.2), 1.26 [0.15,4.56]	1 (0.5), 0.97 [0.02,5.40]	0 (0.0), 0.00 [0.00,2.48]	3 (1.5), 1.91 [0.39,5.59]	19 (1.7), 0.73 [0.44,1.13]^d

Open in a new tab

Overall Cohort data as of May 2019, database not locked

^a

As per the SCS analysis convention, only events occurring within 28 days after the last dose are included in this table for calculation of proportion and IR

^b

In the Maintenance Cohort (placebo arm), 1 case of breast cancer; in the Overall Cohort, 20 malignancy events (excl. NMSC), incl. 2 cases of breast cancer, cervical cancer, malignant melanoma and adenocarcinoma of the colon, and 1 case each of oesophageal adenocarcinoma, hepatic angiosarcoma, cholangiocarcinoma, leiomyosarcoma, diffuse large B-cell lymphoma, Epstein-Barr virus-associated lymphoma, renal cell carcinoma, lung cancer, invasive ductal breast carcinoma, essential thrombocythemia, acute myeloid leukaemia, xerosis, plus secondary malignancies in the liver and peritoneum

^c

Of the 20 pts with malignancy events (excl. NMSC), 3 had received PD tofacitinib 5 mg BID (average total daily dose <15 mg; IR 0.44; 95% CI 0.09, 1.29) and 17 had received PD tofacitinib 10 mg BID (average total daily dose >15 mg; IR 0.86; 95% CI 0.50, 1.38). in the Overall Cohort, the majority (85.0%) of pts with malignancy events (excl. NMSC) were receiving the PD 10 mg BID dose; however, due to the design of the OCTAVE studies, the majority (926/1124; 82.4%) of pts received PD tofacitinib 10 mg BID

^d

Of the 19 Overall Cohort pts with NMSC, most (17/19) had prior immunosuppressant failure, most (15/19) had prior treatment failure with TNFi, and 7 had prior NMSC

BID, twice daily; CI, confidence interval; IR, incidence rate (unique pts with events per 100 PY of exposure); N, number of pts treated in the treatment group; n, number of pts in a specified category; NMSC, non-melanoma skin cancer; PD, predominant dose; pts, patients; PY, pt-years; SCS, summary of clinical safety; SD, standard deviation; TNFi, tumour necrosis factor inhibitor

Disclosure

GR Lichtenstein has received personal fees from AbbVie, American Regent, Celgene, Cellceutix, Eli Lily, Endo, Ferring, Gilead, Janssen Biotech, Merck, Morphic Therapeutic, Pfizer Inc, Prometheus Laboratories, Romark, Salix/Valeant, Shire Pharmaceuticals, Takeda and UCB; other fees from Am J Gastroenterol, American Regent, Gastroenterol Hepatol, Merck, Romark, SLACK, Springer Science and Business Media and Up-To-Date; and other fees (paid to the University of Pennsylvania) from Janssen Orthobiotech, Pfizer Inc and Takeda. MA Ciorba has received lecture and consultancy fees from Pfizer Inc and Takeda, and lecture fees from AbbVie and UCB. G Rogler has received research support and lecture and consultancy fees from AbbVie, Dr. Falk Pharma GMbH, MSD, Pfizer Inc, Takeda, Tillotts, UCB and Zeller; lecture and consultancy fees from Janssen, Phadia and Vifor Pharma; research support and consultancy fees from Augurix, Calypso, Ferring, Novartis and Roche; research support from Ardeypharm and Flamentera; lecture fees from AstraZeneca; consultancy fees from Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Fisher, Genentech, Gilead and Vital Solutions; is a shareholder of Nestlé, Novartis, PharmaBiome and Roche; and holds a directorship at PharmaBiome. AI Sharara has received grants, speaker fees and advisory board fees from AbbVie, Janssen and Takeda; speaker fees from Dr. Falk Pharma GMbH, Ferring and Tillotts; and advisory board fees from Pfizer Inc. N Sunna, R Mundayat, DT Judd and N Lawendy are employees and shareholders of Pfizer Inc. J Panés has received grants and personal fees from AbbVie and MSD, and personal fees from Arena Pharmaceuticals, Boehringer Ingelheim, Celgene, Genentech/Roche, Gilead, GoodGut, GSK, Immunic, Janssen, Nestlé, Oppilan, Pfizer Inc, Progenity, Robarts Clinical Trials, Takeda, Theravance Biopharma and TiGenix.

References

1.Lichtenstein et al. UEG Journal 2019; 7(Suppl): P0393. [Google Scholar]
2.Sandborn et al. Gastroenterology 2012; 142: 257–65. [DOI] [PubMed] [Google Scholar]
3.FDA Advisory Committee Meeting sNDA 203214. FDA 2018: S18. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.97

OP107 Lower Risk of Recurrence with A Higher Induction Dose of Mesalazine and Longer Duration of Treatment in Ulcerative Colitis: Results From The Dutch, Non-Interventional, Impact Study

R West ^1,^✉, J Kuijvenhoven ², M Russel ³, A Bodelier ⁴, KF Bruin ⁵, P van Winkel ⁶

Introduction

Mesalazine is the mainstay of treatment for mild-to-moderate ulcerative colitis (UC) and is effective for the induction and maintenance of remission.^1,2 The dose and duration of induction and maintenance treatment is a potentially important determinant of effectiveness, with evidence suggesting continuing the induction dose for 6-12 months may result in better outcomes, such as clinical remission.^2-5 However, the dose and duration of mesalazine treatment is known to vary in clinical practice.

Aims & Methods

IMPACT was a non-interventional, observational, prospective study that assessed how mesalazine is used in Dutch clinical practice; at which doses and for how long, and how any differences affect disease outcomes. Adult subjects with mild-to-moderate UC that extended beyond the rectum and who received de novo prolonged-release mesalazine (as sachets of granules or tablets) were followed for 12 months,

with those who received additional treatments (local/systemic steroids, immunosuppressants, biologics) excluded. Disease activity (recurrence defined as any of the following after Month 3: increase in dose; addition of enema; or initiation of another therapy) and dose of mesalazine and other treatments were assessed at each clinic visit. The study was performed in accordance with the Declaration of Helsinki and registered as NCT02261636.

Results

In total, 151 patients (median age 46 years; UCDAI score: median 5.0, mean 5.4) were enrolled of whom 71.5% (108) were newly diagnosed with mild-to-moderate UC (Table). The majority (120; 79.4%) of patients received a dose of >4g/day, mostly as a 4g sachet (115; 76.2%). Approximately two-thirds (97; 64.2%) did not receive rectal treatment. Nearly a third (48; 31.8%) of patients had a dose reduction, with the mean time to dose reduction (TDR; primary outcome) being 8.3 months. Disease extent and endoscopic appearance did not appear to influence duration of induction therapy, while TDR increased with higher UCDAI scores at baseline. TDR was longer in patients without recurrence (mean 8.8 months) compared to those with recurrence (mean 4.1 months), although this was not significant (p=0.0917). Patients on >4g/day had a significantly lower chance of recurrence versus those on 2-4g/day (26.6% vs 62.5%; p=0.0424). A longer duration of treatment was associated with a significantly reduced risk of recurrence (hazard ratio [HR] >6 months vs 3-6 months: 0.19, 95% CI 0.08-0.46), which was particularly apparent for those on >4g/day (HR: 0.15, 95% CI 0.06-0.40 vs HR: 0.26, 95% CI 0.01-11.9 for 2-4g/day).

Conclusion

A higher induction dose of mesalazine (>4g/day) and longer duration of treatment (>6 months) was associated with a lower risk of recurrence. Results from previous studies in UC suggesting better outcomes with a longer induction period are therefore - for the first time - confirmed in a non-interventional, real world evidence study.

[Patient baseline characteristics]

Baseline characteristics		Enrolled subjects (N=151)
Male gender, n (%)		78 (51.7%)
Age, median (range)		46.0 (18-43)
UCDAI, mean (standard deviation)		5.4 (2.47)
Newly diagnosed UC, n (%)		108 (71.5%)
Disease Extent	Left-sided UC, n (%)	61 (40.4%)
Disease Extent	Proctosigmoiditis, n (%)	49 (32.5%)
Disease Extent	Extensive UC, n (%)	39 (25.8%)
Disease Extent	Unknown as endoscopy not completed, n (%)	2 (1.3%)

Open in a new tab

Disclosure

This study was sponsored by Ferring BV. Strategen provided medical writing services, which were funded by Ferring. R.L. West has participated in advisory boards or as a speaker or consultant for the following companies: AbbVie, Janssen. P. van Winkel is a full-time employee of Ferring. All other authors have no relevant disclosures.

References

1.Harbord M., Eliakim R., Bettenworth D. et al. Third European Evidence-based Consensus on Diagnosis and Management of Ulcerative Colitis. Part 2: Current Management. J Crohns Colitis. 2017; 11: 769–84. [DOI] [PubMed] [Google Scholar]
2.CBO Richtlijn Inflammatoire darmziekten bij volwassenen, NVMDL 2008. (Dutch IBD treatment guidelines). https://icc-ibd.com/richtlijn [Google Scholar]
3.Takeshi-ma F., Matsumura M., Makiyama K. et al. Efficacy of long-term 4.0 g/day mesalazine (Pentasa) for maintenance therapy in ulcerative colitis. Med Sci Monit. 2014; 20: 1314–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Takahashi F., Tominaga K., Kanamori A. et al. Timing for dose-down of 5-ASA depends on mucosal status with ulcerative colitis. Scand J Gastroenterol. 2016; 51: 827–34. [DOI] [PubMed] [Google Scholar]
5.Pica R., Cassieri C., Cocco A. et al. A randomized trial comparing 4.8 vs. 2.4g/day of oral mesalazine for maintenance of remission in ulcerative colitis. Digest Liv Dis. 2015; 47: 933–7. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.98

OP108 Efficacy and Safety of Mirikizumab After 52-Weeks Maintenance Treatment in Patients with Moderate-To-Severe Crohn'S Disease

BE Sands ^1,^✉, WJ Sandborn ², L Peyrin-Biroulet ³, P Higgins ⁴, F Hirai ⁵, V Jairath ⁶, G D'Haens ⁷, MT Abreu ⁸, R Belin ⁹, E Gomez Valderas ⁹, D Miller ⁹, A Naegeli ¹⁰, V Arora ⁹, P Pollack ⁹, J Tuttle ⁹, T Hibi ¹¹

Introduction

Mirikizumab (miri) is a humanized, IgG4 monoclonal antibody targeting the p19 subunit of IL-23 that has demonstrated clinical efficacy in Phase 2 trials in psoriasis and ulcerative colitis. Previously, we reported mirikizumab was more effective than placebo (PBO) for inducing clinical and endoscopic remission and response at 12 weeks in patients with moderate-to-severe Crohn's disease. Here we report results of maintenance treatment at Week 52. (Phase 2 SERENITY; NCT02891226).

Aims & Methods

Our aims were to evaluate the long-term efficacy and safety of miri over 52 weeks. Patients with moderate-to-severe CD were randomized 2:1:1:2 across 4 treatment arms (PBO, 200, 600, 1000mg miri, administered intravenously (IV) every four weeks (Q4W) at Weeks 0, 4, and 8). Patients who received miri and achieved >1 point improvement at Week 12 in Simple Endoscopic Score for Crohn's Disease (SES-CD) were re-randomized 1:1 into double-blind maintenance to continue IV treatment assignment Q4W (IV-C; N=41) or to 300mg miri SC Q4W (SC; N=46). Due to small sample sizes and lack of an apparent trend across doses at Week 52, all IV and all SC arms were pooled. The primary endpoint (endoscopic response at Week 12) has been previously reported. Clinical and endoscopic endpoints at Week 52 are reported as defined in Table 1. Missing data were imputed as nonresponse.

[Baseline demographics and Week 12 and Week 52 endoscopic and clinical outcomes in patients with an improvement of SES-CD >1 point at Week 12]

Baseline Demographics			Week 12			Week 52^*
	miri IV-C N=41	miri SC N=46		miri IV-C N=41	miri SC N=46		miri IV-C N=41	miri SC N=46
Age, Mean (SD)	40.0 (13.2)	37.8 (11.9)	endoscopic response^a, n (%)	23 (56.1)	24 (52.2)	endoscopic response^a, n (%)	24 (58.5)	27 (58.7)
Gender, male (n, %)	21 (51.2)	23 (50.0)	endoscopic remission^b, n (%)	6 (14.6)	14 (30.4)	endoscopic remission^b, n (%) in W12 remitters n/N (%)	8 (19.5) 3/6 (50.0)	15 (32.6) 9/14 (64.3)
Concomitant CD Medications (n, %) Corticosteroids Azathioprine Methotrexate	8 (19.5) 0 0	11 (23.9) 1 (2.2) 0	PRO remission^c, n (%)	14 (34.1)	9 (19.6)	PRO remission^c, n (%) in W12 remitters, n/N (%)	19 (46.3) 10/14 (71.4)	21 (45.7) 6/9 (66.7)
Prior Biologic Exposure (n, %)	25 (61.0)	27 (58.7)	CDAI remission^e, n (%)	13 (31.7)	15 (32.6)	CDAI remission^e, n (%) in W12 remitters n/N (%)	16 (39.0) 9/13 (69.2)	26 (56.5) 13/15 (86.7)
CRP (median, Q1-Q3)	6.0 (2.1-20.6)	6.3 (2.4-12.9)				TEAE, n (%)	31 (75.6)	35 (76.1)
						SAE^h, n (%)	0 (0)	2 (4.3)

Open in a new tab

^*

for all efficacy endpoints, confidence intervals overlap

^a

endoscopic response: 50% reduction from baseline in SES-CD

^b

endoscopic remission: SES-CD score of <4 for ileal-colonic disease or <2 for isolated ileal disease, and no subscore >1

^c

PRO remission: SF í 2.5 and AP í 1 and no worse than baseline

^d

CDAI response: decrease from baseline in CDAI Score of 100 points or more or a CDAI score < 150

^e

CDAI remission: A CDAI score of <150 points; fABL: as observed; change from baseline

^g

Endoscopic improvement: ≥1 point improvement in SES-CD at Week 12

^h

One patient had a CD flare with concurrent pyelonephritis, hypokalaemia, and dehydration; the second patient had a perforated distal ileum and peritonitis. Neither were determined to be related to study treatment.

Results

Baseline demographics and disease characteristics as well as Week 12 and 52 endoscopic and clinical outcomes are reported (Table 1). Endoscopic (SES-CD) response rates at Week 52 were 58.5% (24/41) and 58.7% (27/46) in the IV-C and SC groups, respectively. PRO remission rates were 46.3% (19/41) and 45.7% (21/46) in the IV-C and SC groups, respectively. Among those with endoscopic response (50% reduction from baseline in SES-CD) at Week 12, 69.6% (16/23) and 66.7% (16/24) in the IV-C and SC groups, respectively, also had endoscopic response at Week 52. Among those with endoscopic remission at Week 12, 50.0% (3/6) and 64.3% (9/14) in the IV-C and SC groups, respectively, also had endoscopic remission at Week 52. One patient in each group discontinued due to an adverse event (AE), and similar frequencies of treatment-emergent AEs and serious AEs were reported in IV-C and SC groups.

Conclusion

Miri demonstrated sustained efficacy to 52 weeks by multiple measures, with few discontinuations due to AEs during the maintenance period. These Phase 2 data support continued characterization of miri efficacy and safety in Crohn's disease in the ongoing VIVID Phase 3 program (NCT03926130).

Disclosure

B. E. Sands has received consultancy fees from from 4D Pharma, Abbvie, Allergan Sales, Amgen, Arena Pharmaceuticals, Boehringer Ingelheim, Capella Biosciences, Celgene, Eli Lilly and Company, EnGene, Ferring, Gilead, Janssen, Lyndra, MedImmune, Oppilan Pharma, Otsuka, Palatin Technologies, Pfizer, Progenity, Rheos Medicines, Seres Therapeutics, Synergy Pharmaceuticals, Takeda, Target PharmaSolutions, Theravance Biopharma R&D, TiGenix, Vivelix Pharmaceuticals, WebMD, and research funding from Celgene, Janssen, Pfizer, and Takeda; W. J. Sandborn has received consultancy fees for Abbvie, Allergan, Amgen, Boehringer Ingelheim, Celgene, Conatus, Cosmo, Eli Lilly and Company, Escalier Biosciences, Ferring, Genentech, Gilead, Janssen, Miraca Life Sciences, Nivalis Therapeutics, Novartis Nutrition Science Partners, Op-pilan Pharma, Otsuka, Paul Hastings, Pfizer, Precision IBD, Progenity, Prometheus Laboratories, Ritter Pharmaceuticals, Robarts Clinical Trials (owned by Health Academic Research Trust or HART), Salix, Shire, Seres Therapeutics, Sigmoid Biotechnologies, Takeda, Tigenix, Tillotts Pharma, UCB Pharma, and Vivelix, research grants from Abbvie, Amgen, Atlantic Healthcare Limited, Celgene/Receptos, Eli Lilly and Company, Genentech, Gilead Sciences, Janssen, and Takeda, and owns stocks/shares in Escalier Biosciences, Ritter Pharmaceuticals, Oppilan Pharma, Precision IBD, and Progenity; L. Peyrin-Biroulet has received honoraria from AbbVie, Allergan, Alma, Amgen, Arena, Sterna, Biogen, Boerhinger Ingelheim, Celgene, Celltrion, Enterome, Ferring, Genentech, Gilead, Hikma, Index Pharmaceuticals, Janssen, MSD, Nestle, Pfizer, Pharmacosmos, Roche, Samsung Bioepis, Sandoz, Takeda, Tillots, grants from Abbvie, MSD, Takeda, and stock options from CTMA; P. D. R. Higgins has received consultancy and/ or advisory board fees from Abbvie, Eli Lilly and Company, and Takeda, and honoraria from Takeda; F. Hirai has nothing to disclose; V. Jairath has received consulting fees from AbbVie, Arena pharmaceuticals, Celltrion, Eli Lilly and Company, GlaxoSmithKline, Genentech, Janssen, Merck, Pendopharm, Robarts Clinical Trials, Sandoz, Takeda, and Topivert, and is on the speakers bureau of Abbvie, Ferring, Janssen, Pfizer, Shire, and Takeda; G. D'Haens has served as an advisor for: AbbVie, Ablynx, Allergan, Alpha-biomics, Amakem, Amgen, AM Pharma, Arena Pharmaceuticals, Astra-Zeneca, Avaxia, Biogen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene/Receptos, Celltrion, Cosmo, Dr Falk Pharma, Echo Pharmaceuticals, Eli Lilly and Company, Engene, Ferring, Galapagos, Genentech/Roche, Gilead, GlaxoSmithKline, Gossamerbio, Hospira/Pfizer, Immunic, Johnson and Johnson, Kintai Therapeutics, Lycera, Medimetrics, Medtronics, Merck Sharp Dome, Millenium/Takeda, Mitsubishi Pharma, Mundipharma, Nextbiotics, Novo Nordisk, Otsuka, Pfizer/Hospira, Photopill, Prodigest, Progenity, Prometheus Laboratories/Nestle, Protagonist, RedHill; Robarts Clinical Trials, Salix, Samsung Bioepis, Sandoz, Seres/Nestle, Setpoint, Shire, Teva, Tigenix, Tillotts, Topivert, Versant, and Vifor; T. Hibi has received advisory/consultancy fees from: AbbVie, Bristol-Myers Squibb, Celltrion, EA Pharma, Eli Lilly and Company, Gilead Sciences, Janssen, Kyo-rin, Mitsubishi-Tanabe Pharma, Nichi-Iko Pharmaceutical, Pfizer, Takeda Pharmaceutical, Zeria Pharmaceutical, and research grants from: AbbVie, EA Pharma, JIMRO, Otuska Holdings, and Zeria Pharmaceuticals; M. Abreu has received consultancy fees from: Boehringer Ingelheim, Eli Lilly and Company, Focus Medical Communications, Gilead, Landos Biopharma, research funding from: Pfizer, Prometheus and Takeda; R Belin, EG Val-deras, D Miller, AN Naegeli, V Arora, P Pollack, and J Tuttle are current employees and shareholders of Eli Lilly and Company; T. Hibi has received Advisory/consultancy fees from Abbvie, Bristol-Myaers Squibb K.K, Cell-trion, EA Pharma, Eli Lilly and Company, Gilead Sciences, Janssen, Kyo-rin, Mitsubishi-Tanabe Pharma., Nichi-Iko Pharmaceutical, Pfizer, Takeda Pharmaceutical, Zeria Pharmaceutical and research grants from Abbvie, EA Pharma, JIMRO, Otuska Holdings, and Zeria Pharmaceuticals

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.99

OP109 Effect of Ustekinumab Maintenance Therapy On Stool Frequency and Rectal Bleeding Through 2 Years in The Unifi Phase 3 Study in Ulcerative Colitis

L Peyrin-Biroulet ^1,^✉, MT Abreu ², S Danese ³, C Marano ⁴, Y Zhou ⁴, H Zhang ⁴, RW Leong ⁵, D Rowbotham ⁶, R Panaccione ⁷, WJ Sandborn, on behalf of the UNIFI Investigators⁸

Introduction

The UNIFI randomized-withdrawal maintenance study evaluated the safety and efficacy of SC ustekinumab (UST) in patients (pts) with moderately to severely active ulcerative colitis (UC) who had responded to IV UST induction. Pts who completed the maintenance study could enter a long-term extension (LTE) through 220 weeks (wks). We evaluated the effect of UST maintenance therapy on stool frequency and rectal bleeding through 92wks in pts who were treated in the LTE.

Aims & Methods

All pts who completed Wk44 were eligible to enter and continue in the LTE at the investigator's discretion. PBO pts were discontinued from the LTE after maintenance study unblinding. During the LTE, pts were eligible to receive dose adjustment (PBO to q8w or q12w to q8w or q8w to q8w [sham dose adjustment]) starting at Wk56 based on investigator assessment of pts’ UC disease activity. Dose adjustment was considered part of the treatment experience and not considered a treatment failure for these analyses. Pts recorded the number of stools & rectal bleeding symptoms for 7 days before each visit. The proportions of pts with Mayo

stool frequency subscores of 0 (normal number of stools) or 1 (1 to 2 stools more than normal) or rectal bleeding subscores of 0 (no blood seen) were evaluated. Absolute stool number was also summarized.

Results

Pts who were randomized and treated in the LTE were included in these analyses (UST 90mg q12w: n=141; UST 90mg q8w: n=143). Absolute stool numbers in the UST q12w and q8w groups were 6.6 & 6.5 stools per day, respectively, at induction baseline (BL) and decreased to 2.8 & 2.7, respectively, by maintenance BL. Reductions achieved at maintenance BL were maintained at Wk92 (2.7 & 2.2, respectively). At induction BL, only 12.8% & 18.2% of pts, respectively, had Mayo stool frequency subscores of 0 or 1. By maintenance BL, 80.9% & 80.4%, respectively, had Mayo stool frequency subscores of 0 or 1, & these percentages were maintained through Wk92 (79.4% & 86.7% at Wk92, respectively). At induction BL, 9.2% & 7.0%, respectively, had no blood seen in the stool (Mayo rectal bleeding subscore of 0). By maintenance BL, 87.2% & 84.6%, respectively, had no blood seen in the stool, & these percentages were maintained through Wk92 (86.5% & 88.8%).

Table 1.

[Stool frequency and rectal bleeding symptoms through Wk92 in randomized pts who entered the LTE]

	UST 90 mg q12w (N=141)	UST 90 mg q8w (N=143)
Absolute Stool Numbers, mean (SD)^a
Induction Wk 0	6.6 (3.22)	6.5 (3.16)
Maintenance Wk 0	2.8 (1.62)	2.7 (1.53)
Maintenance Wk 44^b,c	2.4 (1.49)	2.3 (1.34)
Wk 92 of LTE^b,c	2.7 (2.48)	2.2 (1.60)
Absolute Stool Number ≤3, % (n)^a
Induction Wk 0	6.4 (9)	13.3 (19)
Maintenance Wk 0	68.8 (97)	63.6 (91)
Maintenance Wk 44^b,c	78.0 (110)	80.4 (115)
Wk 92 of LTE^b,c	72.3 (102)	75.5 (108)
Stool Frequency Subscores of 0 or 1, % (n)^a
Induction Wk 0	12.8 (18)	18.2 (26)
Maintenance Wk 0	80.9 (114)	80.4 (115)
Maintenance Wk 44^b,c	83.7 (118)	90.2 (129)
Wk 92 of LTE^b,c	79.4 (112)	86.7 (124)
Rectal Bleeding Subscores of 0, % (n)^a
Induction Wk 0	9.2 (13)	7.0 (10)
Maintenance Wk 0	87.2 (123)	84.6 (121)
Maintenance Wk 44^b,c	97.2 (137)	90.2 (129)
Wk 92 of LTE^b,c	86.5 (122)	88.8 (127)

Open in a new tab

^a

Pts who underwent dose adjustment were not considered treatment failures.

^b

Pts who had an ostomy or colectomy, or discontinued study agent due to lack of therapeutic effect or due to an AE of worsening UC prior to wk92 visit were considered not to have a Mayo subscore indicating inactive or mild disease from the time of event onward.

^c

Pts who had a missing Mayo subscore at a timepoint were considered not to have a subscore indicating inactive or mild disease for that Mayo subscore.

Conclusion

Among pts who were treated in the LTE, reductions in stool frequency and rectal bleeding that had been achieved after IV UST induction were maintained through 2 yrs of UST SC maintenance. These data support the durability of response to UST in UC.

Disclosure

This study was supported by Janssen Research & Development, LLC, Spring House, PA, USA

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.100

OP110 Efficacy and Safety of Ustekinumab For Crohn'S Disease Through 5 Years: Final Results From The Im-Uniti Long-Term Extension

W Sandborn ^1,^✉, P Rutgeerts ², J Lynch ³, R Rebuck ³, Y Wang ³, B Zou ³, C Gasink ⁴, OJ Adedokun ³, BE Sands ⁵, SB Hanauer ⁶, S Targan ⁷, S Ghosh ⁸, W de Villiers ⁹, J-F Colombel ⁵, BG Feagan ¹⁰, on behalf of the IM-UNITI investigators

Introduction

Ustekinumab (UST) is approved for treatment of moder-ate-to-severely active Crohn's disease (CD) and ulcerative colitis. Here we report final results of the IM-UNITI long-term extension (LTE) study, with efficacy and safety through 5 years (yrs) of subcutaneous (SC) UST treatment.

Aims & Methods

1281 patients (pts) entered the maintenance study, including 397 IV UST induction responders, randomized to either SC placebo [PBO], n=133; UST 90mg q12w, n=132; or UST 90mg q8w, n=132. Additional nonrandomized pts were assigned to receive UST 90mg q12w (PBO induction non-responder), UST 90mg q8w (delayed Wk16 responder to UST), or PBO (PBO induction responder). All pts who entered the LTE at Wk44 continued the same treatment regimen they were receiving at Wk44, including 567 UST-treated pts (237 randomized). 151 PBO-treated pts terminated study participation after study unblinding and analysis of the Wk44 data.

Results

Among all pts originally randomized in the maintenance study (ITT analysis), 28.7% and 34.4% of pts were in clinical remission 5 yrs later, at Wk252, on 90 mg UST q12w and q8w, respectively. Approximately half (51.1%) of the pts who entered the LTE completed their final dosing visit. Key results after 5 yrs among pts who entered the LTE, including response and remission overall and in anti-tumor necrosis factor alpha (anti-TNF) therapy history subsets, are summarized in Table 1. of the pts who were in clinical remission at wk252, 89.5% who received continuous q12w (34/38) and 93.3% who received continuous q8w (42/45) were not receiving steroids. Antibody to UST rates through Wk272 remained low, occurring in 5.8% (31/532) of randomized pts continuously receiving UST in the LTE (excluding randomized PBO dose adjusters). Safety events (per hundred pt-yrs) were not higher among all UST-treated pts entering the LTE compared with PBO from Wk0 through Wk272, with 318.5 pt-yrs of follow-up among PBO pts and 2254.6 pt-yrs of follow-up among UST-treated pts. The average duration of follow-up was 89 wks among PBO pts and 206.8 wks among UST-treated pts. Safety events per hundred pt-yrs of follow-up through Wk272 for PBO vs combined UST were AEs: 452.1 vs 325.2, SAEs: 20.4 vs 17.3, infections: 103.0 vs 93.4, and serious infections: 3.8 vs 3.5. No new deaths were reported after Wk156 (6 reported previously). Six pts had malignancies (excluding NMSC) between Wks156 and 272 (q12w: intraocular melanoma, renal cell carcinoma; q8w: endometrial adenocarcinoma, lentigo maligna melanoma, lobular breast cancer in situ, pancreatic carcinoma).

Table 1.

[IM-UNITI Efficacy Assessments at Wk 252 Among Randomized Pts]

From Wk 0 of Maintenance	90 mg UST q12w (n=129)	90 mg UST q8w (n=128)
Clinical Remission (%)^a,b	28.7 (37/129)	34.4 (44/128)
Clinical Remission in pts naïve to TNF-antagonists, % (n/N)^a,b	28.3 (15/53)	44.2 (23/52)
Clinical Remission in TNF-failure pts, % (n/N)^a,b	22.8 (13/57)	21.4 (12/56)
Upon entry into LTE	90mg UST q12w (n=84)^c	90mg UST q8w (n=82)^c	Pts with prior dose adjustment (n=71)^d
Clinical Remission % (n/N)^b,e	45.2 (38/84)	54.9 (45/82)	43.7 (31/71)
Clinical Remission in pts naïve to TNF-antagonists, % (n/N)^b,e	39.5 (15/38)	59.0 (23/39)	57.1 (16/28)
Clinical Remission in TNF-failure pts, % (n/N)^b,e	40.6 (13/32)	44.4 (12/27)	34.4 (11/32)
Clinical Remission and not receiving corticosteroids at Wk 252, % (n/N)^b,e	40.5 (34/84)	51.2 (42/82)	35.2 (25/71)
Clinical remission in pts in remission at maintenance baseline, % (n/N) ^b,e	50.0 (27/54)	59.3 (32/54)	44.9 (22/49)
Modified observed case analysis: Clinical remission with treatment failure rules at Wk 252, excluding pts with missing data, % (n/N)^e,f	61.3 (38/62)	76.3 (45/59)	63.3 (31/49)
Clinical response, % (n/N)^b,e	53.6 (45/84)	57.3 (47/82)	46.5 (33/71)

Open in a new tab

^a

Pts who had a prohibited CD-related surgery, had a loss of response, or discontinued study agent due to lack of efficacy or due to an AE of worsening CD after Wk 44 and prior to the designated analysis timepoint were considered not to be in clinical remission, regardless of their CDAI score.

^b

Pts who had insufficient data to calculate the CDAI score at the designated analysis timepoint were considered not to be in clinical remission.

^c

Pts who were in clinical response to UST IV induction dosing, were randomized to receive study drug on entry into the maintenance study, and did not meet loss of response criteria from Wk 8 through Wk 32.

^d

Pts who were in clinical response to UST induction dosing, were randomized, met loss of clinical response criteria from Wk 8 through Wk 32, and initiated UST 90 mg SC q8w (for pts randomized to receive PBO SC or UST 90 mg SC q12w on entry into this maintenance study) or continue UST 90 mg SC q8w (for pts randomized to receive UST 90 mg SC q8w on entry into this maintenance study).

^e

Pts who had a prohibited CD-related surgery or discontinued study agent due to lack of efficacy or due to an adverse event of worsening CD prior to the designated analysis timepoint were considered not to be in clinical remission, regardless of their CDAI score.

^f

Pts who had insufficient data to calculate the CDAI score at the designated timepoint are excluded, pts who had a treatment failure prior to the designated timepoint are considered not to be in clinical remission, regardless of whether they had insufficient data.

Conclusion

Pts receiving SC UST generally maintained clinical response and remission through 5 yrs. Nearly half of anti-TNF-naive pts who responded to UST IV induction were in remission after 5 yrs of SC q8w maintenance treatment. No new safety signals were observed.

Disclosure

This study was supported by Janssen Research & Development, LLC, Spring House, PA, USA

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.101

OP111 Expression of Micrornas Is Closely Related To Direct Markers of Liver Fibrosis: A Study On Alcoholic Liver Cirrhosis Patients

H Cichoz-Lach ¹, A Michalak ^1,^✉, M Guz ², J Kozicka ¹, M Cybulski ², W Jeleniewicz ², A Stepulak ², B Kasztelan-Szczerbinska ¹

Introduction

MicroRNAs (miRNAs) are well known factors participating in numerous liver disorders. The expression of hepatic microRNAs (miRNAs) might be affected by chronic alcohol consumption and related to alcohol-induced oxidative stress, liver injury, and inflammation. The majority of available studies were conducted to evaluate a potential role of miRNAs as indicators of liver injury or a possible target for the treatment. However, the data on a possible relationship between miRNAs and direct markers of liver fibrosis seem to be strictly limited.

Aims & Methods

We decided to evaluate the expression of selected miRNAs in alcoholic liver cirrhosis (ALC) patients and to assess their relationships with direct markers of liver fibrosis. Two hundred thirty nine persons were enrolled to the study: 139 with ALC and 100 healthy volunteers in control group. Serological expression of miR-126-3p, miR-197-3p and miR-1-3p was evaluated in all examined patients. Direct serological markers of liver fibrosis were also assessed: procollagen I carboxyterminal pro-peptide (PICP), procollagen III aminoterminal propeptide (PIIINP), platelet-derived growth factor AB (PDGF-AB), transforming growth factor-a (TGF-a) and laminin. Then a correlation between evaluated indices was performed. Diagnostic value of assessed miRNAs together with proposed cut-off in research group were measured with area under the curve (AUC).

Results

The concentration of miR-197-3p was significantly lower in ALC group compared to controls (p< 0.0001). We noticed several correlations between examined miRNAs and indirect markers of liver fibrosis: a positive one between miR-197-3p and PDGF-AB (p< 0.05) and negative dependences between miR-1-3p and both: TGF-a and laminin (p< 0.005). Additionally, miR-197-3p was found to correlate negatively with TGF-a, too (p< 0.001). MiR-197-3p turned out to be the most powerful diagnostic marker among assessed miRNAs with AUC of 0.691 for the cut-off < 1.01 amol/μl (p< 0.0001).

Conclusion

To the best of our knowledge, serological expression of miR-126-3p, miR-197-3p and miR-1-3p was mostly explored in patients with hepatocellular injury and hepatocellular carcinoma. Our investigation seems to be the first one that shows a direct linkage between miRNAs and direct markers of liver cirrhosis. MiRNAs could be potentially perceived as probable noninvasive indices of the rebuilt within the extracellular matrix in ALC patients.

Disclosure

Nothing to disclose

References

1.Zhang H., Zhang Z., Gao L., Qiao Z., Yu M., Yu B., Yang T. miR-1-3p suppresses proliferation of hepatocellular carcinoma through targeting SOX9. Onco Targets Ther. 2019. Mar 22; 12: 2149–2157. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Matsuzaki J., Ochiya T. Extracellular microRNAs and oxidative stress in liver injury: a systematic mini review. J Clin Biochem Nutr. 2018. Jul; 63(1): 6–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Ni J.S., Zheng H., Huang Z.P., Hong Y.G., Ou Y.L., Tao Y.P., Wang M.C., Wang Z.G., Yang Y., Zhou W.P. MicroRNA-197-3p acts as a prognostic marker and inhibits cell invasion in hepatocellular carcinoma. Oncol Lett. 2019. Feb; 17(2): 2317–2327. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Xu T., Li L., Hu H.Q., Meng X.M., Huang C., Zhang L., Qin J., Li J. MicroRNAs in alcoholic liver disease: Recent advances and future applications. J Cell Physiol. 2018. Jan; 234(1): 382–394. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.102

OP112 The Alteration of Gut Microbiome in Liver Transplantation: First Results From The Transplantlines Biobank and Cohort Study

Y Li ^1,^2,^✉, S Hu ^1,², C Swarte ^1,³, R Gacesa ^1,², A Vich Vila ^1,², R Douwes ⁴, H Harmsen ⁵, D Jansen ¹, V de Meijer ⁶, J Fu ^2,⁷, A Zhernakova ², C Wijmenga ², E Festen ^1,², S Bakker ³, H Blokzijl ¹, R Weersma ¹

Introduction

Liver transplantation (LT) is a curative treatment for patients with end-stage liver disease, hepatic tumors and genetic metabolic disorders. While the role of the gut microbiome in liver disease is increasingly being understood, little is known in the setting of LT and its outcomes. Animal studies showed that a shift in the gut microbiome after LT predicted acute cellular rejection ⁴. in humans the dynamics of gut microbiota during LT and the relationship with mortality after LT remain unknown. A better understanding of the gut microbiome in the setting of end-stage liver disease and LT enables the development of gut microbiome directed therapies to improve long-term outcomes of LT patients.

Aims & Methods

The aim of this study is to characterize the microbial composition and microbial community functions of LT recipients, investigate the composition of the gut microbiota in the period of pre- and post-LT, and longitudinally evaluate dynamic changes in the gut microbiota. Shotgun metagenomic sequencing was performed on 425 fresh frozen fecal samples from 375 patients of screening period (pre-LT), short-term (3, 6, 12 months), 1-5 years, 5-10 years and long-term post-LT (>10 years) derived from the TransplantLines Biobank and 1183 matched healthy controls derived from the Dutch Microbiome Project. Clinical characteristics of underlying disease, disease severity, surgical parameters, immunosuppressive regime, comorbidities and patient survival were extracted from hospital records. Microbial diversity and community description were compared using Wilcoxon test between the groups by calculating Shannon diversity and Bray-Curtis dissimilarities. Next, multivariable general linear mixed model analyses, which corrected for identified confounders such as age, sex, PPI and antibiotic exposure were performed to compare microbial taxa, pathways, antibiotic resistant genes and virulence factors. Significant differences were defined according to the FDR adjusted p-val-ue for multiple testing. The association between microbiota and patient mortality was calculated using Cox proportional-hazards analysis.

Results

Compared to healthy controls, patients with end-stage liver disease had a significantly lower gut microbial diversity (p< 0.001). in addition, patients with severe liver disease (MELD>18 and Child-pugh C class) showed the lowest microbial diversity. Interestingly, the diversity was significantly decreased in peri-operative LT (within 3 months) and increased long-term post-LT (> 1-year post-LT), compared to pre-LT samples (p< 0.001). Microbiome of Pre-LT patients was characterized by a loss of short chain fatty acid producing bacteria including Subogranulum spp. and Ak-kermansia spp. (P_FDR< 0.20). The use of immunosuppressives was associated to an increase in the abundance of microbial pathways involved in biosynthesis of cyclosporine and mycophenolic acid (P_FDR< 0.20). Strikingly, the high microbial diversity was associated with a lower patient mortality post-LT (adjusted p-value=0.043, hazard ratio 0.34, 95%CI 0.12 to 0.97).

Conclusion

This study is the largest survey on the role of microbiome in the context of liver transplantation to date. We found several significant gut microbial signatures pre- and post-LT that could have important clinical implications. Moreover, microbial diversity shows to be associated with survival post-LT, therefore revealing a new potential biomarker or therapeutic target.

Disclosure

Nothing to disclose

References

1.Chassaing B., Etienne-Mesmin L., Gewirtz A.T. Microbiota-liver axis in hepatic disease. Hepatology. 2014. doi: 10.1002/hep.26494 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Bajaj J.S., Betrapally N.S., Gillevet P.M. Decompensated cirrhosis and microbiome interpretation. Nature. 2015. doi: 10.1038/nature14851 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Macnaughtan J., Jalan R. Clinical and pathophysiological consequences of alterations in the microbiome in cirrhosis. Am J Gastroenterol. 2015. doi: 10.1038/ajg.2015.313 [DOI] [PubMed] [Google Scholar]
4.Ren Z., Jiang J., Lu H. et al. Intestinal microbial variation may predict early acute rejection after liver transplantation in rats. Transplantation. 2014. doi: 10.1097/TP.0000000000000334 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.103

OP113 Response of Liver and Spleen Stiffness To Portal Pressure Lowering Drugs in A Rat Model of Cirrhosis

O Elshaarawy ^1,^✉, S Alquzi ¹, T Peccerella ¹, J Mueller ¹, V Rausch ¹, S Mueller ¹

Introduction

Liver stiffness (LS) is increasingly used to screen for liver fibrosis, in addition, spleen stiffness (SS) is an established parameter to assess portal hypertension which is tightly related to the hemodynamics of blood flow and vascular resistance. Little is known about the response of LS and SS to vasoactive substances.

Aims & Methods

We here studied LS and SS in an TAA-induced cirrhosis rat models after exposure to various vasoactive drugs using a miniaturized Fibroscan platform (liFibroscan).

We induced cirrhosis in 24 wildtype 8 weeks old adult male Wistar rats with 200 mg/Kg dosage of Thioacetamide (TAA) through intraperitoneal injection of 50 mg/ml solution 2 times per week for 6 weeks. The six groups consisted of control (sodium chloride), metoprolol, udenafil, enalapril, terlip-ressin and carvidelol. LS and SS were measured by iFibroscan (Echosens, Paris). The rats underwent general anesthesia with isoflurane inhalation. After anesthesia, abdominal aorta, inferior vena cava and portal vein were cannulated with 24-gauge cannula and connected to Power lab device (AD instruments) to continuously measure the mean arterial pressure (MAP), heart rate (HR) and portal vein pressure (PVP). Drugs were injected systemi-cally and data were collected at time points 0, 15 and 30 mins.

Results

LS and SS was significantly higher in TAA treated rats than in the control group (23.8 vs 3.8 kPa and 19.6 vs 47.8 kPa, P< 0.0001). in addition, they had significantly bigger and heavier spleens (6 vs 4 cm and 2.7 vs 1 gm, P< 0.0001, respectively). in all drugs, LS and SS followed tightly the change of the portal vein pressure (r=0.681and 0.622, P< 0.01, respectively). Also, SS was significantly correlated with spleen size and weight (r=0.723 and 0.663, respectively< 0.01). Noteworthy, a significant decrease of PVP ranging from 22% to 34% (P< 0.05) was seen after 15 to 30 minutes with metoprolol, udenafil, enalapril and carvedilol which was accompanied with a significant decrease in LS and SS ranging from 18.2 to 44% (P< 0.05). (see the Table) Interestingly, with terlipressin, LS and SS only slightly decreased which could be explained by counteracting PVP and

MAP. Thus, while PVP decreased by 20% (P< 0.001), MAP increased by 35% (P< 0.001). Overall, carvedilol showed the best response regarding the decrease of PVP, LS and SS. of note, the heart rate increased after metoprolol and udenafil injection (ca. 10%, P< 0.05), while it decreased in response to terlipressin and carvedilol by ca 30% (P< 0.01).

Conclusion

LS and SS strongly correlates PVP and responded differently to various vasoactive drugs. Combined non-invasive LS and SS measurement could be useful to monitor the patient's response and compliance to portal pressure lowering drugs.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.104

OP114 Association of Gut Microbiota with Coronary Artery Disease in Patients with Nonalcoholic Fatty Liver Disease

O Kurinna ^1,^✉

Introduction

Coronary artery disease (CAD) is a leading cause of mortality among patients with non-alcoholic fatty liver disease (NAFLD) with unfavorable epidemic dynamics. Only in the United States, there are approximately 85 million of NAFLD patients, and it is expected to deteriorate. Data on pathological links between CAD and NAFLD with focus on gut microbiota are scarce.

Aims & Methods

Thus, the aim of the study was to investigate association of main bacterial phylotypes of gut microbiota with CAD in NAFLD patients.

In our study we enrolled 95 subjects with NAFLD, of whom there were 24 patients with CAD and 9 practically healthy volunteers. On all subjects we investigated the gut microbiota composition, its main phylotypes, by qPCR with relevant primers. and all patients were diagnosed and treated in compliance with actual clinical guidelines. For NAFLD diagnostics we used ultrasound steato- and elastometry and excluded all other etiologies for fatty liver. Statistical analysis of results was performed using the statistical software package ‘STATISTICA 13.1’. The distribution of all studied parameters was different from normal, we used non-parametric statistics. The results were considered significant if p < 0.05.

Results

Among patients there was no statistically significant changes in liver function, lipid and carbohydrate metabolism. We found an increase in the concentration CRP was increased in patients with concomitant CAD by 24.17%, but the dynamics did not reach statistical significance (p=0.05). Contrary, the concentration of TNF-α was significantly increased in patients with concomitant CAD by 18.62% (p<0.05).

We found the tendency towards an increase in the concentration of Bac-teroidetes by 37.11% with a decrease in Firmicutes by 7.38% with respective reduction in the F/B ratio by 22.62%. Conversely, in NAFLD patients with concomitant CAD we found a decrease in the Actinobacteria concentration by 41.37% (p<0.05). To assess treatment impact, we developed a mnemonic positional code to study the multifactorial relation of the prescribed treatment with changes in the gut microbiota, that demonstrated a tendency to change the Actinobactera concentration.

Conclusion

Our study showed significant changes of the gut microbiota, Actinobacteria phylotype, in NAFLD patients with concomitant CAD. Actin-bacteria concentration was significantly decreased in CAD patients with tendency to depend on prescribed treatment. It can be explained by a sedentary lifestyle but needs further research.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.105

OP115 Liver Mirna-222-3P Is Activated in Non-Alcoholic Steatohepatitis and Ablation Prevents Disease Pathogenesis

AL Simào ^1,^✉, PM Rodrigues ², MB Afonso ¹, AFA Tomás ¹, Á Santos-Laso ², R Jimenez-Agüero ², E Eizaguirre ², L Bujanda ², MJ Pareja ³, R Macias ⁴, JM Banales ², CMP Rodrigues ¹, RE Castro ¹

Introduction

Non-alcoholic fatty liver disease (NAFLD) encompasses several disease stages, from simple steatosis to non-alcoholic steatohepatitis (NASH), each associating with specific disease hallmarks, such as hepatic lipid accumulation, inflammation, ballooning and fibrosis. Accumulating evidence supports a functional role for microRNAs (miRNAs/miRs) in regulating such processes. in particular, anti-apoptotic miR-222-3p expression has been reported to increase during disease progression in experimental NAFLD.

Aims & Methods

Here, we aim to elucidate whether increased miR-222-3p expression represents a key, primary event in NAFLD progression and identify disease hallmarks that correlate with miR-222-3p, hinting at its multiple roles in disease pathogenesis.

C57BL6 mice were fed five different NAFLD-inducing diets, namely a methionine and choline-deficient (MCD) diet for 2 and 8 weeks; a high-fat choline-deficient diet for 14 weeks; a high fat, 2% cholesterol diet with high fructose for 25 weeks; a high-fat/calorie diet with high fructose/glucose in drinking water (HFC+HF/G) for 16 weeks; and a choline-deficient amino acid-defined diet for 32 weeks. MCD and HFC+HF/G-fed mice were also treated with or without antagomiR-222-3p.

Biochemical and histological features of NASH were determined in serum and liver samples. Liver miR-222-3p expression was evaluated by qPCR in human biopsies obtained from 165 NAFLD patients characterized at the histological and metabolic levels (NAS < 2: n=26; NAS 3 to 4: n=62; NAS ≥ 4: n=77).

Results

Mice fed with any of the five diets developed different degrees of NASH and fibrosis, paralleling liver steatosis, inflammation, insulin resistance and weigh gain. Regardless of the model, liver miR-222-3p expression levels were significantly increased in all diseased mice compared with control diet-fed animals.

Remarkably, miR-222-3p inhibition counteracted MCD-induced inflammation, collagen deposition and oxidative stress, as well as HFC+HF/G-induced steatosis, inflammation and liver injury.

In NAFLD patients, liver miR-222-3p expression was found to progressively increase with steatosis, lobular inflammation and ALT. Likewise, miR-222-3p expression levels were significantly higher in patients with bridging fibrosis and cirrhosis. in addition, univariate analysis indicated that liver miR-222-3p expression positively correlates with histological features (steatosis, lobular inflammation and fibrosis), hepatic triglyceride content and serum hepatic enzymes.

Conclusion

Overall, our results indicate that activation of miR-222-3p appears to be a key player in NAFLD/NASH pathogenesis, correlating with several well-characterized disease hallmarks. Targeting miR-222-3p may

represent an appealing prospective therapeutic approach for preventing NAFLD. (Gilead Sciences International - Research Scholars Program in Liver Diseases; LCF/PR/HP17/52190004, “la Caixa” Foundation; OLD-HEPAMARKER, 0348_CIE_6_E, Spain; PTDC/MED-PAT/31882/2017, FCT, Portugal).

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.106

OP116 High Dose Oral Thiamine Reduces Chronic Fatigue in Patients with Quiescent Inflammatory Bowel Disease: A Randomised Double Blinded Crossover Trial

P Bager ^1,^2,^✉, CL Hvas ¹, C Rud ¹, JF Dahlerup ¹

Introduction

Fatigue in inflammatory bowel disease (IBD) is one of the most burdensome symptoms, and the prevalence of fatigue in quiescent IBD is approximately 40% [1]. The causes of fatigue are multifactorial and have led to a variety of interventions. Despite the impact of fatigue in IBD, only few interventions have demonstrated clinical effect [2]. A pilot study indicated a beneficial effect of high dose thiamine given for 20 days [3]. Thiamine is a water-soluble vitamin with no risk of accumulation in the body in patients with normal kidney function.

Aims & Methods

The aim was to evaluate high dose oral thiamine for chronic fatigue in patients with IBD in remission.

We conducted a randomised, double-blinded, placebo controlled, crossover trial. Patients with quiescent IBD and chronic fatigue, normal renal function, no obvious other reasons for fatigue, and no deficiencies were included at the outpatient clinic at our university hospital. Patients were allocated 1:1 to: Group 1 (high dose oral thiamine for 4 weeks, a 4 weeks wash-out period, followed by 4 weeks oral placebo; or Group 2 (oral placebo for 4 weeks, a 4 weeks wash-out period, followed by 4 weeks high dose oral thiamine). Doses of thiamine ranged between 900-1800 mg /day, based on gender and body weight. Fatigue was measured using the Inflammatory Bowel Disease-Fatigue Questionnaire (IBD-F) [4]. IBD-F, section I (range 0-20) measures the severity of fatigue. We included patients with a score >12, equivalent to the 95% percentile of fatigue in the background population [5]. The primary endpoint was an improvement of > 3 points on the IBD-F, section I. The study was conducted according to GCP and was monitored by the University Hospital GCP Unit.

Results

A total of 40 patients were included and randomised. All participants had >80% adherence to treatment. We observed no statistical differences between the groups at baseline. Analyses of possible carryover revealed no carryover effect.

Both groups had reductions of fatigue after the whole study period. The primary efficacy endpoint was met at week 12, with reduction of mean fatigue of 4.7 points (95% CI 3.4-6.0; p< 0.001).

A delayed treatment model was assessed, showing a 4.7 points (95% CI 3.4-5.9) reduction in fatigue after thiamine treatment, compared to a 1.4 point (95% CI 0.3-2.4) reduction in fatigue after placebo (p< 0.001). Furthermore, 55% of Group 1 and 75% of Group 2 showed an improvement of more than 3 points while on thiamine treatment compared with 25% of Group 1 and 35% of Group 2 while on placebo. No serious adverse events were detected. Adverse events were temporary and included: headache, nausea, pyrosis, sleeplessness, or oral blisters.

Conclusion

High dose oral thiamine given for 4 weeks, showed statistically and clinically significant effects on chronic fatigue in patients with quiescent IBD and no deficiencies. The treatment was acceptable and safe for the patients.

Disclosure

Nothing to disclose

References

1.van Langenberg D.R., Gibson P.R. Systematic review: fatigue in inflammatory bowel disease. Aliment Pharmacol Ther. 2010; 32: 131–43. [DOI] [PubMed] [Google Scholar]
2.Farrell D., Artom M., Czuber-Dochan W. et al. Interventions for fatigue in inflammatory bowel disease. Cochrane Database Syst Rev. 2020; 4: Cd012005. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Costantini A., Pala M.I. Thiamine and fatigue in inflammatory bowel diseases: an open-label pilot study. J Altern Complement Med. 2013; 19: 704–8. [DOI] [PubMed] [Google Scholar]
4.Czuber-Dochan W., Norton C., Bassett P. et al. Development and psychometric testing of inflammatory bowel disease fatigue (IBD-F) patient self-assessment scale. J Crohn Colitis. 2014; 8: 1398–1406. [DOI] [PubMed] [Google Scholar]
5.Bager P., Vestergaard C., Juul T. et al. Population-based normative data for the inflammatory bowel disease fatigue scale - IBD-F. Scand J Gastroenterol. 2018; 53: 1274–9. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.107

OP117 Cobitolimod For Moderate-To-Severe Left-Sided Ulcerative Colitis (Conduct): A Phase Iib, Randomised, Double-Blind, Placebo-Controlled, Dose-Ranging Trial

R Atreya ^1,^✉, L Peyrin-Biroulet ², A Klymenko ³, M Augustyn ⁴, I Bakulin ⁵, D Slankamenac ⁶, P Miheller ⁷, A Gasbarrini ⁸, X Hebuterne ⁹, K Arnesson ¹⁰, T Knittel ¹⁰, J Kowalski ¹⁰, MF Neurath ¹¹, W Sandborn ¹², W Reinisch ¹³

Introduction

Cobitolimod is a topically administered DNA-based oligo-nucleotide that binds the Toll-like receptor 9 (TLR9), triggering an anti-inflammatory mucosal response. We assessed the efficacy and safety of different dose regimens of cobitolimod in patients with moderate-to-severe left-sided ulcerative colitis (UC).

Aims & Methods

The CONDUCT trial was a multicenter, randomised, placebo-controlled dose-ranging phase 2b study in 12 European countries. Patients had left-sided, moderate-to-severe UC defined by a full Mayo score of 6 to 12 with a centrally read endoscopic subscore >2 and no other individual subscore < 1. Patients enrolled had an inadequate response to conventional or biological therapies. 211 patients were randomised and received enemas of cobitolimod 31 mg (n=40), 125 mg (n=43), or 250 mg (n=42) at weeks 0 and 3, cobitolimod 125 mg at weeks 0, 1, 2, and 3 (n=42), or placebo (n=44) in a double blinded fashion stratified for concomitant use of steroids and previous use of a TNF-inhibitor. The primary efficacy endpoint was proportion of patients in clinical remission at week 6, defined

by the 3-component Mayo score i) rectal bleeding subscore (RBS) of 0, ii) stool frequency subscore (SFS) of 0 or 1 (with at least one point decrease from baseline), and iii) centrally read endoscopy score (ES) of 0 or 1. Secondary exploratory endpoints at week 6 included clinical remission by full Mayo score, symptomatic remission (RBS 0, and SFS ≤1 with ≥1 point decrease from baseline), and endoscopic improvement (ES 0 or 1).

Results

Significantly more patients achieved clinical remission at week 6 in the cobitolimod 2x250 mg group than with placebo (21.4% [nine] versus 6.8% [three]; odds ratio 3.8 [80% CI 1.53-9.47]; one-sided p=0.0247 using non-responder imputation for missing data). Other dose groups showed no significant difference in clinical remission versus placebo: 2x31 mg (12.5% [five]; p=0.18); 2x125 mg (4.7% [two]; p=0.66); and 4x125 mg (9.5% [four]; p=0.33). Clinical remission defined by full Mayo score and symptomatic remission were also statistically significant in the cobitolimod 2x250 mg group, whereas endoscopic improvement and proportion of patients with fecal calprotectin < 250 mg/kg among those with levels beyond that threshold at baseline were numerically in favor of the cobitolimod 2x250 mg group compared to placebo. Adverse events occurred in 47.7% (21) of patients in the placebo (4.5% [2] with serious events), 25.0% (10) in the cobitolimod 2x31 mg (5.0% [2] serious), 39.5% (17) in the 2x125 mg (none serious), 35.7% (15) in the 4x125 mg (4.8% [2] serious), and 42.9% (18) in the 2x250 mg (9.5% [4] serious) group with worsening of UC representing 8 out of 10 SAEs.

[Selection of secondary exploratory endpoints at week 6.]

Full Analysis Population - Observed Data	Cobitolimod 2x31 mg	Cobitolimod 2x125 mg	Cobitolimod 4x125 mg	Cobitolimod 2x250 mg	Placebo
Clinical remission, full Mayo score	15.2% (5/33)	2.4% (1/41)	7.7% (3/39)	*20.0^ (7/35)**	7.7% (3/39)
Symptomatic remission	27.0% (10/37)	26.2% (11/42)	25.0% (10/40)	*35.1^ (13/37)**	20.9% (9/43)
Endoscopic improvement	20.6% (7/34)	12.2% (5/41)	25.6% (10/39)	40.5% (15/37)	30.0% (12/40)
Clinical response	51.5% (17/33)	43.9% (18/41)	38.5% (15/39)	57.1% (20/35)	51.3% (20/39)

Open in a new tab

^*

one-sided p-value<0.10, based on Cochran Mantel-Haenszel test, adjusted for stratification factorsa

Conclusion

The CONDUCT study demonstrates that topical administration of 2x250 mg of the TLR9 agonist cobitolimod is effective to induce clinical remission at week 6 and no specific safety signals were detected. The concept of TLR9 activation is a promising novel therapeutic option in UC and is planned to be confirmed in an upcoming phase III program.

Disclosure

The CONDUCT study was funded by InDex Pharmaceuticals AB. RA was coordinating investigator of the CONDUCT study and receives research grant and consultancy fee from InDex Pharmaceuticals AB. LPB was principle investigator in the CONDUCT study and receives consultancy fee from InDex Pharmaceuticals AB. AK, MA, IB, DS, PM, AG, XH was principle investigators in the CONDUCT study. KA is an employee and shareholder of InDex Pharmaceuticals AB. TK and JK are consultants and shareholders of InDex Pharmaceuticals AB. MFN receives research grant from InDex Pharmaceuticals AB. WJS receives consultancy fee from InDex Pharmaceuticals AB. WR was medical advisor in the CONDUCT study and receives consultancy fee from InDex Pharmaceuticals AB.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.108

OP118 Pattern of Biological Use and Treatment Outcomes in Patients with Inflammatory Bowel Diseases - A Danish Nationwide Cohort Study

M Zhao ^1,^✉, M Sall Jensen ², T Knudsen ³, J Kelsen ⁴, M Coskun ⁵, J Kjellberg ², J Burisch ¹

Introduction

Therapeutic management of inflammatory bowel diseases (IBD) is rapidly evolving with an expanding armamentarium of biological drugs. Treatment persistence and switch are markers of effectiveness and drug tolerance, however, real-world data from population-based cohorts on the pattern and outcomes of biological use is limited.

Aims & Methods

This study aimed to investigate trends in treatment pattern of biological drugs and risk of surgery in a nationwide cohort of biological-naïve IBD patients. The patient population consisted of adult patients diagnosed with ulcerative colitis (UC) or Crohn's disease (CD) who initiated treatment with a biological drug between 2011 to 2018. Patients were identified in the Danish National Patient Registry using a ten-year look back period and individual-level data on patient age, gender, diagnoses, biological treatment and surgery was collected. Treatment patterns with first- and second-line biological drugs as well as treatment persistence beyond the induction phase (defined as day 0-100) were assessed using Kaplan-Meier survival curves. Risk factors of major surgery after treatment start were assessed in a COX-regression model.

Results

During 2011 to 2018, a total of 6.036 IBD (52% UC, 48% CD) patients were treated with a biological drug for the first time. Biological exposure in UC patients increased over time from 5.0% per year in 2011 to 10.8% per year in 2018, and similarly from 8.9% to 14.5% per year in CD. A trend towards earlier biological use was seen in CD (p=0.01) where the proportion of patients exposed to a biological drug within two years from diagnosis increased from 47.0% in 2011 to 57.7% in 2018, while no trend was seen in UC (43.3% in 2011, to 42.2% in 2018). in both UC and CD, infliximab (IFX) was the most common first-line drug (93.2% UC, 89.0% CD) followed by adalimumab (ADA) (3.5% UC, 9.3% CD). in UC patients who continued first-line treatment beyond the induction phase, 32% discontinued IFX within one year and 67% within three years. For those who received ADA and golimumab (GOL) as first-line drug, 22% and 17%, respectively, discontinued treatment within one year, while 58% and 50% discontinued treatment within three years. in CD, treatment persistence was similar for IFX and ADA with 40% discontinuing treatment within one year and more than half discontinued treatment within three years. in 30.7% of UC and 31.0% of CD patients, a change to a second biological drug was necessary. in UC, ADA was the most common second-line drug (49.8%) followed by GOL (29.5%) and vedolizumab (VDZ) (17.9%). ADA (71.1%) was also the most common second-line drug in CD, followed by VDZ (12.2%).

A trend towards decline in surgery rate within three years after initiation of biological treatment was seen in both UC (23.7% in 2011; 19.1% in 2015) and CD (18.7% in 2011; 13.5% in 2015), however, this was only significant in UC in the COX-regression model (2011 vs. 2015: HR 0.73, 95%CI 0.53 to 0.10, p=0.05). Biological use within two years of diagnosis (p< 0.01), previous surgery (p=0.03) and age above 40 (p< 0.01) significantly increased the risk of surgery in UC. None of these factors were associated with surgery risk in CD.

Conclusion

In real-life settings, discontinuation rate of biological treatment in IBD patients was high and treatment switch was required in one third of patients. The increased use of biological in UC and CD appears to reduce the surgery rate, but surgery need remained high. This highlights the need to optimize the use of available biological drugs in clinical practice.

Disclosure

Funding disclosures: This study was funded by Takeda Pharma A/S.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.109

OP119 Long-Term Evolution After Anti-Tnf Drug Discontinuation in Patients with Inflammatory Bowel Disease

MJ Casanova ^1,^✉, M Chaparro ¹, O Nantes Castillejo ², JM Benitez ³, M Rojas-Feria ⁴, J Castro Poceiro ⁵, JM Huguet-Malavés ⁶, A Martin-Cardona ⁷, M Aicart ⁸, J Tosca ⁹, M Martín-Rodríguez ¹⁰, C Gonzalez Muñoza ¹¹, M Mañosa ¹², E Leo Carnerero ¹³, LJ Lamuela-Calvo ¹⁴, I Pérez-Martínez ¹⁵, L Bujanda ¹⁶, J Hinojosa ¹⁷, R Pajares ¹⁸, F Argüelles-Arias ¹⁹, JL Pérez-Calle ²⁰, GE Rodríguez ²¹, J Guardiola Capon ²², M Barreiro-de Acosta ²³, F Bermejo ²⁴, J Barrio ²⁵, B Beltran ²⁶, F Gomollón ²⁷, RH Lorente ²⁸, A Gutierrez Casbas ²⁹, M Domínguez-Cajal ³⁰, C Dueñas ³¹, A Ponferrada Diaz ³², M Van Domselaar ³³, P Ramírez-de la Piscina ³⁴, L Ramos ³⁵, P Almela ³⁶, M Navarro Llavat ³⁷, B Botella ³⁸, C Castaño Milla ³⁹, C Rodriguez Gutierrez ², E Iglesias-Flores ³, MJ Serrano-Muñoz ⁴, E Ricart ⁵, A Monzó ⁶, M Esteve Comas ⁷, E Garrido ⁸, M Mínguez ⁹, MJ Cabello-Tapia ¹⁰, E García-Planella ¹¹, E Domènech ¹², A Núñez-Ortiz ¹³, S García-López ¹⁴, JP Gisbert ¹

¹Hospital Universitario de La Princesa, IIS-IP, and CIBERehd, Gastroenterology Unit, Madrid, Spain

²Complejo Hospitalario de Navarra, Instituto de Investigación Sanitaria de Navarra (IdiSNA), Gastroenterology Unit, Pamplona, Spain

³Hospital Universitario Reina Sofía, Instituto Maimónides de Investigación Biomédica de Córdoba (IMIBIC), Gastroenterology Unit, Córdoba, Spain

⁴Hospital Universitario de Valme and CIBEREHD, Gastroenterology Unit, Sevilla, Spain

⁵Hospital Clínic i Provincial and CIBEREHD, Gastroenterology Unit, Barcelona, Spain

⁶Consorcio Hospital General Universitario de Valencia, Gastroenterology Unit, Valencia, Spain

⁷Hospital Universitario Mutua Terrassa, and CIBEREHD, Gastroenterology Unit, Terrassa, Spain

⁸Hospital Universitario Ramón y Cajal, Gastroenterology Unit, Madrid, Spain

⁹Hospital Clínico Valencia, Gastroenterology Unit, Valencia, Spain

¹⁰Hospital Universitario Virgen de las Nieves, Gastroenterology Unit, Granada, Spain

¹¹Hospital Santa Creu i Sant Pau, Gastroenterology Unit, Barcelona, Spain

¹²Hospital Universitario Germans Trias i Pujol, and CIBEREHD, Gastroenterology Unit, Badalona, Spain

¹³Hospital Universitario Virgen del Rocío, Gastroenterology Unit, Sevilla, Spain

¹⁴Hospital Universitario Miguel Servet, Gastroenterology Unit, Zaragoza, Spain

¹⁵Hospital Universitario Central de Asturias, Gastroenterology Unit, Oviedo, Spain

¹⁶Hospital Universitario de Donostia, Instituto Biodonostia, Universidad del País Vasco (UPV/EHU), and CIBEREHD, Gastroenterology Unit, Donostia, Spain

¹⁷Hospital Universitario de Manises, Gastroenterology Unit, Valencia, Spain

¹⁸Hospital Universitario Infanta Sofía, Gastroenterology Unit, Madrid, Spain

¹⁹Hospital Universitario Virgen Macarena, Gastroenterology Unit, Sevilla, Spain

²⁰Hospital Universitario Fundación de Alcorcón, Gastroenterology Unit, Madrid, Spain

²¹Hospital Universitario Nuestra Señora de la Candelaria, Gastroenterology Unit, Santa Cruz de Tenerife, Spain

²²Hospital Universitario de Bellvitge, Gastroenterology Unit, L'Hospitalert de LLobregat, Spain

²³Hospital Clínico Universitario de Santiago de Compostela, Gastroenterology Unit, Santiago de Compostela, Spain

²⁴Hospital Universitario de Fuenlabrada, Instituto de Investigación Sanitaria del Hospital La Paz (IdiPAZ), Gastroenterology Unit, Madrid, Spain

²⁵Hospital Universitario Río Hortega, Gastroenterology Unit, Valladolid, Spain

²⁶Hospital Universitario La Fe, and CIBEREHD, Gastroenterology Unit, Valencia, Spain

²⁷Hospital Clínico Lozano Blesa, Instituto de Investigación Sanitaria Aragón (IIS Aragón), and CIBEREHD, Gastroenterology Unit, Zaragoza, Spain

²⁸Hospital General de Ciudad Real, Gastroenterology Unit, Ciudad Real, Spain

²⁹Hospital General Universitario Alicante and CIBEREHD, Gastroenterology Unit, Alicante, Spain

³⁰Hospital General San Jorge, Gastroenterology Unit, Huesca, Spain

³¹Hospital Universitario de Cáceres, Gastroenterology Unit, Cáceres, Spain

³²Hospital Universitario Infanta Leonor, Gastroenterology Unit, Madrid, Spain

³³Hospital de Torrejón de Ardoz, Gastroenterology Unit, Madrid, Spain

³⁴Hospital HUA Álava, Gastroenterology Unit, Vitoria, Spain

³⁵Hospital Universitario de Canarias, Gastroenterology Unit, La Laguna, Spain

³⁶Hospital General Universitario de Castellón, Gastroenterology Unit, Castellón, Spain

³⁷Hospital de Sant Joan DespíMoisés Broggi, Gastroenterology Unit, Sant Joan Despí, Spain

³⁸Hospital Universitario Infanta Cristina, Gastroenterology Unit, Madrid, Spain

³⁹Hospital Universitario Rey Juan Carlos, Gastroenterology Unit, Madrid, Spain

^✉

Contact E-Mail Address: mjcasanova.g@gmail.com

Introduction

The long-term risk of relapse after anti-TNF discontinuation in inflammatory bowel disease (IBD) patients in remission, has not been well established.

Aims & Methods

The aims of the present study were: 1) to assess the risk of relapse at long-term after anti-TNF discontinuation; 2) to identify the factors associated with relapse; 3) to calculate the response rate to re-treatment with the same anti-TNF, 4) to evaluate the safety of re-treatment with these drugs, and 5) to investigate the rate of complications of patients that relapsed.

Retrospective, observational, multicenter study conducted at 39 Spanish centers. Crohn's disease or ulcerative colitis patients who had been treated with anti-TNFs and in whom these drugs had been withdrawn after achieving clinical remission, were included. Follow-up time after anti-TNF discontinuation was at least 6 months.

Results

1,055 patients were included (53% women, mean age at diagnosis 42 years, 68% Crohn's disease). The reasons for discontinuation the anti-TNF therapy were: 75% elective decision, 18% adverse events, and 7% remission due to top-down strategy. Treatment with immunomodulators (IMMs) was maintained in 68% after anti-TNF discontinuation. of the 562 patients with available endoscopy or magnetic resonance imaging (MRI) at the moment of anti-TNF discontinuation, 460 patients were in deep remission (clinical remission plus normal endoscopy or normal MRI). The median follow-up time was 34 months. The incidence rate of relapse was 12% per patient-year (95%CI=11-14), and the median time to relapse was 17 months. The cumulative incidence of relapse was 50% (95%CI=47-53): 19% at one year, 31% at two years, 38% at three years, 44% at four years, and 48% at five years of follow-up. in the multivariate analysis, older age at IBD diagnosis (HR=0.98; 95%IC=0.97-0.99), and maintenance therapy with IMMs at anti-TNF discontinuation (HR=0.64; 95%CI=0.52-0.8) were associated with a lower probability of relapse. Discontinuation due to elective decision vs. top-down (HR=1.88; 95%CI=1.20-2.94), and discontinuation due to adverse events vs. top-down (HR=2.01; 95%CI=1.24-3.25) were the associated with a higher probability of relapse. of the 60% patients retreated with the same anti-TNF after relapse, 73% (95%CI=68-78) regained

remission, and 29% (95%CI=11-22) relapsed at the end of follow-up. of the 75 patients who did not respond, 48% (95%CI=35-62) achieved remission with other therapies. of the 190 patients who started other therapies after relapse, 62% (95%CI=54-70) achieved remission with the new treatment, and of these, 31% % (95%CI=22-41) relapsed at the end of follow-up. After retreatment, 16% had adverse events, the majority of them mild.

Conclusion

The incidence rate of relapse after anti-TNF discontinuation in IBD patients that were in remission was 12% per patient-year. Almost half of the patients relapsed at five years of follow-up. Older age at diagnosis and maintenance treatment with IMMs were associated with a lower risk of relapse. Anti-TNF discontinuation due to elective decision or adverse events (vs. top-down) was associated with a higher probability of relapse after anti-TNF discontinuation. Retreatment of relapse with the same anti-TNF was effective and safe. Less than half of the patients who did not respond to anti-TNF retreatment achieved remission with other therapies.

Disclosure

María José Casanova1, has received education funding from Pfizer, Takeda, Janssen, MSD, Ferring and Abbvie. María Chaparro1, has served as a speaker, or has received research or education funding from MSD, Abbvie, Hospira, Pfizer, Takeda, Janssen. O. Nantes2, has served as a speaker, or has received education funding from MSD, Abbvie y Pfizer. JM. Benítez3, has served as a speaker, a consultant and advisory member or has received research or education funding from Dr. Falk Pharma, Faes Farma, Ferring, Shire Pharmaceuticals, MSD, Abbvie, Takeda and Janssen. JM. Huguet6, has served as a speaker, a consultant and advisory member for or has received research funding from MSD, Ferring, Abbvie, Janssen, Biogen, Sandoz, Kern Pharma, and Takeda. MM. Martín-Rodríguez10, has received education funding from MSD, Takeda, Janssen, Abbvie, Tillotts Pharma y Ferring. M. Mañosa12, has served as a speaker, has received research or educational funding or advisory fees from Danone, Allergan, Almirall, MSD, AbbVie, Takeda, Janssen, Ferring. E. Leo-Carnerero13, has served as a speaker, a consultant and advisory member for or has received research funding from MSD, Pfizer, Abbvie, Takeda, Janssen, Tillotts Pharma, Shire Pharmaceuticals, Ferring, Dr. Falk Pharma y Otsuka Pharmaceutical. J. Hinojosa17, has served as a speaker, a consultant and advisory member for, and has received research funding from MSD, Abbvie, Ferring, Faes Farma, Shire Pharmaceuticals, Chiesi, Otsuka Pharmaceutical, Pfizer - Hospira, Kern Pharma, UCB Pharma, Vifor Fharma, Sandoz, Biogen, Janssen, Takeda and Dr. Falk Pharma. F. Argüelles-Arias19, has served as a speaker, a consultant and as an advisory member for or has received research funding from Janssen, MSD, Abbvie, Pfizer, Kern Pharma, Biogen, Sandoz, Takeda, Ferring, Faes Farma, Shire Pharmaceuticals, Dr. Falk Pharma, Tillotts Pharma, Gebro Pharma, Amgen and Vifor Pharma. J. Pérez-Calle20, has served as a speaker or received research or educational funding or advisory fees from Ferring, MSD, Takeda, Jannsen, Faes, Dr. Falk, Pfizer. J. Guardiola22, has served as a speaker, a consultant, or an advisory member for or has received research funding from MSD, Abbvie, Kern Pharma, Takeda, Janssen, Pfizer, Ferring, Shire Pharmaceuticals, Ge-bro Pharma, and Tillots. M. Barreiro-de Acosta23, has served as a speaker, a consultant and advisory member for or has received research funding from MSD, Abbvie, Celltrion, Takeda, Janssen, Pfizer, Ferring, Faes Farma, Shire Pharmaceuticals, Dr. Falk Pharma, Chiesi, Gebro Pharma, Roche, Otsuka Pharmaceuticals, and TillottsPharma. JP. Gisbert1: has served as a speaker, a consultant and advisory member for or has received research funding from MSD, Abbvie, Hospira, Pfizer, Kern Pharma, Biogen, Takeda, Janssen, Roche, Celgene, Ferring, Faes Farma, Shire Pharmaceuticals, Dr. Falk Pharma, Tillotts Pharma, Chiesi, Casen Fleet, Gebro Pharma, Otsuka Pharmaceutical, Vifor Pharma. All the rest of authors: none

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.110

OP120 Ustekinumab Is More Effective Than Azathioprine To Prevent Endoscopic Postoperative Recurrence in Crohn'S Disease

A Buisson ^1,^✉, S Nancey ², L Manlay ³, DT Rubin ⁴, X Hebuterne ⁵, B Pariente ⁶, M Fumery ⁷, D Laharie ⁸, X Roblin ⁹, G Bommelaer ³, B Pereira ³, L Peyrin-Biroulet ¹⁰, L Vuitton ¹¹

Introduction

Preventing postoperative recurrence (POR) is a major concern in Crohn's disease (CD). While azathioprine is an option, no data is available on ustekinumab efficacy in this situation.

Aims & Methods

We aimed to assess whether ustekinumab is more effective than azathioprine to prevent endoscopic POR in patients with CD. We retrospectively collected data from all consecutive CD patients treated with ustekinumab after intestinal resection in 9 centers. All these patients received intravenous induction with a median interval between restoration of bowel continuity and onset of ustekinumab of 30.5 days IQR [16.8-49.8 days]. All the subsequent subcutaneous injections were performed using a dose of 90 mg. The control group (azathioprine 2.5 mg/kg/day monotherapy) was composed of patients who participated in a randomized controlled trial conducted in the same centers, comparing azathio-prine alone or in combination with curcumin. Azathioprine was started 14 days after restoration of bowel continuity. We included patients with CD older than 18 years-old who had undergone a surgical resection for ileal, ileocolonic, or colonic CD with ileocolonic anastomosis. All the macroscopic lesions had to be removed during the surgery. The anastomosis had to be reachable by ileocolonoscopy.

The primary endpoint was endoscopic POR evaluated 6 months (M6) after intestinal resection or following stoma closure, defined as Rutgeerts’ index > i2. The secondary endpoints were alternative definitions of endoscopic POR using a Rutgeerts’ index > i2b as the cut-off value (i2b-endoscopic POR), and Rutgeerts’ index > i3 as the cut-off value (severe endoscopic POR). Endoscopic procedures at M6 were scored by independent central readers in the control group. For the patients treated with ustekinumab, all endoscopic evaluations were performed by local readers who were experienced in assessing Rutgeerts’ index in clinical trials. Propensity-matched analyses (Inversed Probability of Treatment weighting = IPTW) were applied to make the two groups comparable.

Results

Overall, 32 patients were included in the ustekinumab group and 31 in the azathioprine group. No patient received azathioprine in the ustekinumab group. The median level of 6-TGN was 365 pmol at M3 and 260 pmol at M6. The propensity-matched analysis was adjusted on the main risk factors (smoking, fistulizing phenotype, prior bowel resection, resection length > 30 cm and > 2 biologics before surgery) but also on thiopurines or ustekinumab exposure prior to surgery making the two arms comparable (½d½< 0.2). in this cohort, 93.7 % (30/32) and 83.9% (26/31) of the patients had at least one of the main risk factors of POR in ustekinumab and azathioprine group, respectively.

After IPTW, the rate of endoscopic postoperative recurrence at 6 months was lower in patients treated with ustekinumab compared to azathioprine (28.0% vs 54.5%, p =0.029). After IPTW, the rates of i2b-endoscopic POR (Rutgeerts’ index ≥ i2b) and severe endoscopic POR (Rutgeerts’ index ≥ i3) were 20.8% vs 42.5% (p =0.066) and 16.9% vs 27.9% (p =0.24), in the ustekinumab and azathioprine groups, respectively. No adverse event was reported in the ustekinumab group while two serious adverse events occurred in the control group.

Conclusion

In this cohort study, we found that ustekinumab was more effective than azathioprine to prevent endoscopic POR in patients with CD. Hence, ustekinumab could be an interesting option in this situation.

Disclosure

Anthony BUISSON: Consulting fees for Abbvie, Amgen, Biogen, Janssen, MSD, Pfizer, Roche, Takeda and Tillots. Lecture fees for Abbvie, Amgen, Biogen, Janssen, Mayoly-Spindler, MSD, Norgine Pfizer, Roche, Takeda and Tillots. Stéphane NANCEY: consulting fees from Merck, Abbvie, Takeda, Ferring, Norgine, Vifor Pharma, Novartis, Janssen Cilag, Hospira, Takeda and HAC-Pharma. Luc MANLAY: none. David T RUBIN: consultant fees for Abbvie, Abgenomics, Allergan, Inc., Amgen, Celgene Corporation, Forward Pharma, Genentech/Roche, Janssen Pharmaceuticals, Merck & Co., Inc., Miraca Life Sciences, Napo Pharmaceuticals, Pfizer, Salix Pharmaceuticals, Inc., Samsung Bioepis, Sandoz Pharmaceuticals, Shire, Takeda and grant support from Abbvie, Genentech/Roche, Janssen Pharmaceuticals, Prometheus Laboratories, Shire, Takeda and UCB Pharma. Xavier HEBUTERNE: Abbvie, Abivax, Alphasigma, Arena, Astellas, Baxter, Bristol Myers Squibb, Cellgen, Gilead, Eli Lilly, Enterome, Ferring, Frese-nius-Kabi, InDex Pharmaceuticals, Janssen, MSD, Nutricia, Pfizer, Roche, Sanofi-Advantis, SAlix, Sangamo, Servier, Takeda, Theravance, TillotAb-bvie, MSD, Pfizer, Janssen Cilag, and Takeda. Benjamin PARIENTE: Consulting or lecture fees: AbbVie, MSD, Ferring, Takeda, Janssen, Biogaran, Pfizer, Biogen, Sandoz, Tillots, Lilly. Mathurin FUMERY: Consulting or lecture fees for Abbvie, Amgen, Biogen, Gilead, Janssen, MSD, Pfizer, Ferring, Tillots, Celgene, Boehringer and Takeda. David LAHARIE: consulting and lecture fees from AbbVie, Ferring, Janssen Cilag, MSD, Pfizer, and Takeda. Xavier ROBLIN: Abbvie, MSD, Pfizer, Janssen Cilag, and Takeda. Gilles BOMMELAER: none Bruno PEREIRA: none Laurent PEYRIN-BIROULET: consulting fees from Merck, Abbvie, Janssen, Genentech, Ferring, Norgine, Tillots, Vifor, Shire, Therakos, Pharmacosmos, Pilege, BMS, UCB-Pharma, Hospira, Celltrion, Takeda, Biogaran, Boerhinger-Ingelheim, Lilly, Pfizer, and HAC-Pharma. Lecture fees from Merck, Abbvie, Takeda, Janssen Cilag, Ferring, Norgine, Tillots, Vifor, Therakos, HAC-Pharma, and Mitsubishi. Lucine VUITTON: lecture fees from Abbvie, Ferring, Mayoli, MSD, Pfizer, Jans-sen and Takeda; consulting fees from Abbvie, Ferring, Gilead, Janssen and Takeda; research grants from MSD, Pfizer and Takeda.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.111

OP121 Mass Spectrometry-Based Protein Profiling of Circulating Exosomes For Predicting The Prognosis of Intraductal Pancreatic Mucinous Neoplasms

N Park ^1,^✉, WH Paik ¹, HM Lee ², C Lee ³, KP Kim ⁴, SH Lee ¹, J-K Ryu ¹, Y-T Kim ¹

Introduction

Intraductal pancreatic mucinous neoplasms (IPMNs) have a spectrum of neoplastic transformation ranging from low-grade dysplasia to invasive carcinoma. The current guidelines rely largely on radiologic findings to evaluate the potential of malignancy, but the levels of evidence are low.

Aims & Methods

This study aimed to find out key exosomal proteins that distinguish benign cyst from high risk of malignant potential in IPMN. To investigate differentially expressed proteins (DEPs) comprising exo-somes from IPMN and IPMN associated carcinoma (IPMC) patients, circulating exosomes from 400 microliters of plasma were isolated using commercial exosome isolation kit (ExoLutE®, Rosetta Exosome, Inc.). Size-exclusion chromatographic analysis was performed to verify the purity of isolated circulating exosomes. To identify subsets of exosomal DEPs, 30 micrograms of protein from each purified exosomes were analysed using tandem-mass-tag quantitative proteomics technology. We here performed two-step statistical analysis. in discovery session, we enrolled six patients (three IPMN and three IPMC) and analysed the exosomal DEPs from the two groups. To validate the result, eight additional patients (four IPMN and four IPMC) were enrolled and the purified exosomes were applied to the quantitative proteomics analysis under blindness.

Results

Approximately 30-70 micrograms in protein quantity of highly pure exosomes were isolated from each patient. Size-exclusion chromato-graphic analysis has shown that the area of the peak corresponding to exosomes in all samples were > 85%, indicating all samples are sufficiently pure for downstream quantitative proteomics analysis. As the result of quantitative proteomics analysis using equal protein quantity of the purified exosomes from each patient, in discovery session, we identified 91 exosomal DEPs (>1.2 in fold change) from the two groups with p-value < 0.05 among 640 quantifiable proteins. Using the DEPs found in discovery session, the differential diagnosis in validation cohort was perfectly possible under blindness. in addition, we further identified 77 exosomal DEPs (>1.2 in fold change, p-value< 0.05) from the independent clinical subset in validation session. By combinatorial analysis of proteins common to DEPs in both session, we further identified a subset of 9 exosomal proteins highly potential for predicting IPMCs from IPMNs.

Conclusion

Based on potential roles of exosomes in pathophysiology, analysing protein component of circulating exosomes could open a new avenue to predict and identify IPMC with malignant potential.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.112

OP122 Modelling Plasticity and Dysplasia of Pancreatic Ductal Organoids Derived From Human Pluripotent Stem Cells

J Merkle ^1,^✉, M Breunig ¹, MK Melzer ¹, T Engleitner ², PC Hermann ¹, L Perkhofer ¹, J Krüger ¹, S Heller ¹, M Müller ¹, T Seufferlein ¹, R Rad ², M Wagner ¹, M Meier ³, M Hohwieler ¹, A Kleger ¹

Introduction

Personalized in vitro models for dysplasia and carcinogenesis in the pancreas have been constrained by insufficient engineering approaches to generate human exocrine pancreatic tissue. Human pluripotent stem cells (hPSCs) may provide the appropriate platform for biomedical studies due to their capability to differentiate into every cell type in the human body. We established a differentiation matrix to generate virtually pure pancreatic duct-like organoids (PDLOs) from hPSCs. These PDLOs showed morphological, transcriptional and functional characteristics of human pancreatic ducts.

Aims & Methods

We generated an early in vitro cancer model by introduction of a Doxycycline-inducible piggyBac system in hPSCs which allows timed expression of oncogenic KRAS^G12D. Wildtype hPSCs as well as a CRISPR/Cas-engineered CDKN2A^KO/KO line both armed with oncogenic KRAS were differentiated according to our protocol to the ductal lineage, followed by oncogene activation in vitro and in vivo.

Results

These PDLOs showed a characteristic growth phenotype of lumen-filling after KRAS^G12D induction, which was further characterized on the molecular level and demonstrated senescence and growth arrest accompanied with increased epithelia-to-mesenchymal transition (EMT). These oncogene-induced responses were even more pronounced in PD-LOs with mutated KRAS and genetic covariance of CDKN2A^KO/KO. Ortho-topic transplantation was additionally incorporated into the experimental workflow. While transplanted PDLOs with oncogenic KRAS alone developed to well-differentiated tumours, PDLOs with KRAS and CDKN2A-loss resulted in dedifferentiated human pancreatic ductal adenocarcinomas (PDAC) in vivo.

Conclusion

Summarized, our novel and unique differentiation platform generates human untransformed ducts and allows modelling of plasticity, dysplasia, and cancer formation in a human and genetically defined background in the pancreas. Thereby, we also show that human ducts are indeed permissive to generate PDAC. Tailored genome editing strategies mimicking the mutational make-up of human PDAC will open novel opportunities to provide a unique and valuable human PDAC model.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.113

OP123 Bariatric Surgery Decreases Prevalence of Pancreatic Cancer in Patients with Prior Diabetes and Obesity: A 20-Year National Analysis

A Syed ^1,^2,^✉, T Seoud ¹, K Farah ¹, G Eid ¹, S Thakkar ³

Introduction

The prevalence of pancreatic cancer (PC) is on the rise and is predicted to be second-leading cause of cancer related deaths in the near future. Obesity and type 2 diabetes (DM) are well known risk factors for PC via chronic inflammation, excess hormones and growth factors released by adipose tissue. Bariatric surgery (BS) has been shown to improve hyperglycemia and even resolve it in many cases more so than medical therapy alone.

Aims & Methods

We aim to evaluate the role of BS in the prevalence of PC in patients with prior DM and obesity. We queried the population-based Explorys database (IBM, New York) from 1999-2019 to collect aggregated de-identified electronic health records. Data was obtained using the SNOMED-CT coding system to search for patients with “malignant tumor of pancreas” and “type 2 diabetes mellitus.” A temporal relationship was established to observe patients with (1) DM and obesity (defined as BMI > 35) with or without subsequent (2) bariatric surgery (either Roux-en-Y gastric bypass or sleeve gastrectomy) and (3) lastly having a diagnosis of PC.

Results

We identified 1,435,350 patients with concomitant DM and obesity, of which 10,620 (0.7%) had record of BS subsequently. The majority of patients did not undergo BS (98.9%). The prevalence of PC was significantly higher in patients with predisposing DM and obesity without record of BS versus those who did have BS (0.32% vs. 0.19%, P = 0.0179). Patients with DM and obesity that underwent BS had higher rates of being female, younger (18-65 yrs. of age), and Caucasian or African-American (P < 0.05).

Conclusion

Bariatric surgery significantly decreased the prevalence of pancreatic cancer with prior DM and obesity versus patients with no record of bariatric surgery. Clinicians should consider bariatric surgery in eligible patients with metabolic disorders such as diabetes and obesity.

Disclosure

Nothing to disclose

[Baseline Demographic Data of Patients with Type 2 Diabetes and Obesity]

	DM-Obesity with BS (N = 10,620)	DM-Obesity without BS (N = 1,418,950)	P
Sex, n (%)
Male	2,870 (27.0)	601,950 (42.4)	<0.0001
Female	7,750 (73.0)	816,810 (57.6)	<0.0001
Age, n (%)
18-65	8,390 (79.0)	770,900 (54.3)	<0.0001
> 65	2,240 (21.1)	647,030 (45.6)	<0.0001
Race, n (%)
Caucasian	8,290 (78.1)	1,083,410 (76.4)	<0.0001
African American	1,960 (18.5)	244,750 (17.2)	0.0004
Asian	40 (0.4)	11,810 (0.8)	<0.0001
Hispanic/Latin American	150 (1.4)	23,340 (1.6)	0.1016

Open in a new tab

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.114

OP124 Echoendoscopic Guided Fine Needle Aspiration Versus Circulating Cell-Free Dna For Genetic Profiling of Borderline and Non-Resectable Pancreatic Ductal Adenocarcinoma

S Bunduc ^1,^✉, B Stoica ^1,², RA Iacob ^1,³, A Sorop ⁴, I Manea ^4,⁵, D Constantinescu ⁴, E Moise ⁴, L Dima ⁴, M Sirbu ⁴, AM Pantazica ⁴, A Ghionescu ⁴, R Chelaru ⁴, G Becheanu ^1,³, M Dumbrava ¹, A Croitoru ^1,³, S Dima ^1,^3,⁴, I Popescu ^1,^3,⁴, C Gheorghe ^1,^3,⁴

Introduction

Incidence of pancreatic cancer is increasing, and the prognosis remains dismal, despite the progress in treatment strategies. Better biomarkers are needed for improved stratification of patients prior to therapy.

Aims & Methods

Our study aimed to evaluate adequacy for Next Generation Sequencing (NGS) of genomic DNA (gDNA) isolated from tumoral tissue sampled by echoendoscopic guided fine needle aspiration (EUS-FNA) and circulating cell free DNA (ccfDNA) isolated from peripheral venous blood in patients with borderline and unresectable PDAC. Concordance of NGS results between the two biological sampling methods was inves-tigated. During 11.2018-02.2020, patients with borderline and unresect-able histologically confirmed PDAC, diagnosed by EUS-FNA were enrolled. Twenty-two Gauge (22G) or 19G needles were used for FNA. Venous blood was collected from the same patients prior to EUS-FNA. NGS was conducted for a group of 20 patients for both gDNA and ccfDNA, using a custom-made hot-spot panel including 1121 somatic mutations in 40 genes. The average sequencing depth was 11500 x, with a 5% mutation frequency filter for data analysis.

Results

Genomic DNA and ccfDNA were isolated from 53 histologically confirmed PDAC patients. Significantly better DIN values were obtained with 19G needles (8.27 vs. 7.15, p = 0.001). All ccfDNA samples had adequate purity by on-chip electrophoresis. Overall 77.27% of EUS-FNA samples and 97.22% of ccfDNA samples were adequate for downstream NGS applications. Overall concordance of NGS results was 91.92%. Our analysis identified pathogenic and potentially pathogenic mutations in FNA and ccfDNA samples in 31.28% vs 25.8% of cases respectively. Most frequent pathogenic mutations in FNA samples were KRAS (65%), ERBB2 (40%), TP53 (35%), ATM (30%) and PALB2 (10%). For ccfDNA samples mutation concordance was good for ERBB2, ATM and PALB2 but poor for KRAS and TP53.

Conclusion

EUS-FNA and ccfDNA yielded adequate quantity and quality of DNA for downstream NGS applications. There was overall good concordance (>90%) between NGS results using the two biological samples but mutation-specific concordance variations were identified.

Disclosure

Nothing to disclose

Acknowledgement

This work was supported by the research Grant PN-III-P1-1.2-PCCDI-2017-0797 (PANCNGS).

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.115

OP125 Copy Number Alterations and Gene Fusions Correlate with Immune Cell Infiltration and Survival in Pancreatic Cancer

IJM Levink ^1,^✉, LAA Brosens ², MR Aberle ³, SS Rensen ³, DL Cahen ¹, SWM Olde Damink ⁴, MJ Bruno ¹, SI Buschow ¹, GM Fuhler ¹, MP Peppelenbosch ¹

Introduction

Immunotherapy has not been effective for pancreatic ductal adenocarcinoma (PDAC). However, factors underlying antitumor immunity in PDAC are poorly understood. We hypothesized that, for PDAC, not the quantity of MHC-binding neoantigens, but neoantigen quality and frequency of both copy number alterations (CNAs) and gene fusions are related to effective immune cell responses and survival. Additionally, we expect that an organoid can be a proper model for neoantigen selection for personalization of immunotherapy.

Aims & Methods

We performed extensive genomic and transcriptomic profiling of paired normal and tumor tissue of PDAC patients with distinct survival (n=13; table 1) and matched PDAC-derived organoids (n=2). Additionally, neoantigen MHC-binding prediction (NetMHC) and immunological profiling (CIBERSORTx¹) were performed. Data was analyzed using Spearman's correlation.

Table 1.

[Cohort description]

	N=13
Survival, range (months)	5-144
Age, range (years)	29-78
Gender, % female	46
Tumor stage, %stage III	54
Neoadjuvant treatment, % gemcitabine	85

Open in a new tab

Results

Mutations or CNAs were most often detected in KRAS, TP53, TTN, ZNF471, ZNF667 and LOC101″2″″″2 (>4 patients). MHCI or MHCII binding of neoantigens occurred most frequently as a result of mutations in KRAS, TP53, TTN, APOB, COL6A (>3 patients). The range of neoantigen-per-mu-tation ratio was 0.53-1.85 for MHCI (HLA rank≤2) and 0.38-43.02 for MCHII (HLA rank≤10). Neoantigen quality (MHCI only), as based on the mean and upper 10% of immunogenicity, similarity-to-self and similarity-to-known score, did not correlate with survival.

Immunological profiling showed that survival time was correlated with tumor-infiltrating plasma cells (r=0.63), monocytes (r=0.63), M0 macrophages (r=-0.62) and activated dendritic cells (r=-0.63), but not with (cytotoxic) T-cell infiltration (r=0.36, p=0.23).

While mutational load and neoantigen burden did not correlate with survival or immune cell infiltration, the number of CNAs negatively correlated with survival time (r=-0.62), cytotoxic T-lymphocytes (r=-0.64), plasma cells (r=-0.55), follicular helping T-cells (r=-0.73), and monocytes (r=-0.59).

Conversely, the number of gene fusions positively correlated with survival time (r=0.62), infiltrating CD4+ resting memory T-cells (r=0.75) and monocytes (r=0.57). As compared to gene mutations, gene fusions less often resulted in MHC-binding neoantigens, with neoantigen-per-fusion ratios of 0.00-0.51 for MHCI and 0-3.08 for MCHII.

When comparing paired tumor tissue and organoids, 49-57% of mutations, 17-34% of CNAs and 11-15% of fusions showed overlap. Interestingly, 50-56% of fusions were only detected in organoids and may have developed during culturing. Frequent neoantigens present in tumor tissue (KRAS, TP53, TTN, APOB, COL6A) were preserved in organoids.

Conclusion

In contrast to other cancer types, not tumor mutational load and neoantigen quantity or quality, but rather the number of CNAs were negatively, and gene fusions positively, correlated with survival and immune cell infiltration. While CD8+ T-cell infiltration did not correlate to survival in our study, plasma cell and macrophage-lineage derived cell infiltration did. More research is required to identify pathways responsible for these correlations. Organoids retain the mutational burden of the parent tumor to some extent, yet less for fusions and CNAs.

Disclosure

Nothing to disclose

References

Newman A.M., Steen C.B., Liu C.L. et al. Determining cell type abundance and expression from bulk tissues with digital cytometry. Nat Biotechnol 37, 773–782 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.116

OP126 The Effect of The Pringle Maneuver During Hepatectomy On Post-Operative Outcomes: An Analysis of The Nsqip Database

F Daniel ^1,^✉, T Nammour ¹, M Hosni ², H Tamim ³, A Mailhac ³, M Khalife ⁴, W Faraj ⁴, F Jamali ⁴

Introduction

The Pringle maneuver is a surgical technique that involves clamping the hepatoduodenal ligament to interrupt the blood flow from the hepatic artery and portal vein. It was postulated that this technique helps to minimize blood loss during hepatectomy procedures. We aim to investigate the correlation between the use of the Pringle maneuver during hepatectomy and the 30-day post-operative outcomes.

Aims & Methods

We studied all patients undergoing hepatectomy from the National Surgery Quality Improvement Program (NSQIP) database. NSQIP includes prospective validated outcomes and anonymized data for patients’ demographics, functional statuses, preoperative risk factors, laboratory data, and 30-day postoperative outcomes for patients undergoing major surgery in more than 500 hospitals. Descriptive statistics were presented as % for categorical variables and mean ± SD for continuous variables. Bivariate analysis was conducted to check for correlation between the short-term 30-day postoperative outcomes and the use of the Pringle maneuver. A p-value of 0.05 was considered statistically significant.

Results

15748 patients undergoing hepatectomy were analyzed according to the surgical maneuver used during the operative procedure. The Pringle maneuver was performed in 3706 (23.5%) of the patients while the rest, which included 12042 patients (76.5%), had either no or other techniques used. The post-operative 30-day mortality risk was higher in the Pringle group as compared to the non-Pringle group (2% vs. 1.3%). Postoperative complications were also higher including bile leakage, wound infection, respiratory and urinary complications, thromboembolism, sepsis, bleeding, and composite morbidity risk. The Pringle maneuver was also associated with higher readmission rates, longer operations times, and total length of stay. Cardiac complications and return to the operation room did not appear to differ between the two groups.

[Comparing 30-day outcomes among patients who underwent hepatectomy according to the surgical maneuver used during the operative procedure]

Hepatectomy Maneuver	Other (12042, 76.5%)	Pringle (3706, 23.5%)	p-value
Invasive operation	132 (9.6%)	391 (10.6%)	0.0003
Bile-leakage	823 (6.9%)	353 (9.6%)	<0.0001
Mortality	156 (1.3%)	73 (2.0%)	0.0027
Wound	956 (7.9%)	346 (9.3%)	0.0069
Cardiac	166 (1.4%)	64 (1.7%)	0.1221
Respiratory	587 (4.9%)	220 (5.9%)	0.0104
Urinary	171 (1.4%)	72 (1.9%)	0.024
Thromboembolism	301 (2.5%)	128 (3.45%)	0.0018
Sepsis	654 (5.43%)	245 (6.61%)	0.0068
Bleed	1997 (16.6%)	765 (20.6%)	<0.0001
Return to OR	339 (2.8%)	125 (3.4%)	0.0791
Composite Morbidity	1769 (14.7%)	683 (18.4%)	<0.0001
Readmission	1008 (8.4%)	380 (10.3%)	0.0004
Operation time	233.0 (121.5)	3706 (268.7)	0.015
Length of stay	6.5 (6.4)	7.6 (7.4)	0.005

Open in a new tab

Conclusion

The Pringle maneuver was found to be associated with higher short-term surgical mortality and morbidity. Future randomized trials are needed for better assessing the usefulness of this surgical method.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.117

OP127 Intrathoracic Versus Cervical Anastomosis After Minimally Invasive Oesophagectomy For Oesophageal Cancer: A Randomised Controlled Trial

M Verstegen ^1,^2,^✉, F van Workum ^1,², B Klarenbeek ¹, S Bouwense ^1,³, S Gisbertz ⁴, G Hannink ¹, JW Haveman ⁵, J Heisterkamp ⁶, E Kouwenhoven ⁷, J van Lanschot ⁸, G Nieuwenhuijzen ⁹, D van der Peet ¹⁰, F Polat ¹¹, S Ubels ¹, M Rovers ¹, group Rosman C, ICAN collaborative ¹

Introduction

Robust evidence is lacking whether transthoracic minimally invasive oesophagectomy (MIO) with intrathoracic anastomosis or cervical anastomosis should be preferred for patients with oesophageal or

gastro-oesophageal junction (GEJ) cancer. We aimed to study the effectiveness of transthoracic MIO with intrathoracic anastomosis compared to transthoracic MIO with cervical anastomosis.

Aims & Methods

In this multicenter randomized controlled trial, patients with oesophageal (below the level of the carina) or GEJ cancer (up to 2 centimetres into the GEJ) planned for curative resection were recruited. Patients were randomly assigned (1:1) to transthoracic MIO with intra-thoracic or cervical anastomosis. The primary endpoint was anastomotic leakage requiring endoscopic, radiologic or surgical intervention. Secondary outcome parameters were overall anastomotic leakage rate, other postoperative complications, length of stay, mortality and quality of life.

Results

A total of 262 patients were randomly assigned to transthoracic MIO with intrathoracic anastomosis (n=130) or cervical anastomosis (n=132). Seventeen patients were excluded from the trial due to incorrect inclusion (n=2), physical unfitness for surgery (n=3), patients’ choice (n=3), interval metastases (n=5) or peroperative metastases (n=4). Anastomotic leakage necessitating reintervention occurred in 15 of 122 (12.3%) patients with intrathoracic anastomosis and in 39 of 123 (31.7%) patients with cervical anastomosis after transthoracic MIO (risk difference -19.4%, 95% CI -29.5% - -9.3%). Overall anastomotic leakage rate was 12.3% in the intrathoracic anastomosis group and 34.1% in the cervical anastomosis group (risk difference -21.9%, 95% CI -32.1% - -11.6%). ICU length of stay, mortality rates and overall quality of life were comparable between groups, but intrathoracic anastomosis was associated with better quality of life in three subdomains.

Conclusion

This study provides evidence for a lower rate of anastomotic leakage requiring reintervention after transthoracic MIO with intrathorac-ic anastomosis compared to transthoracic MIO with cervical anastomosis in patients with mid to distal oesophageal or GEJ cancer.

Disclosure

Funding: ZonMw.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.118

OP128 Clinical Outcomes After Total Pancreatectomy: A Prospective Multicenter Pan-European Snapshot Study

A Latenstein ^1,^✉, L Scholten ², H Ahmad Al-Saffar ³, B Björnsson ⁴, G Butturini ⁵, G Capretti ^6,⁷, N Chatzizacharias ⁸, C Dervenis ⁹, I Frigerio ¹⁰, T Gallagher ¹¹, S Gasteiger ¹², A Halimi ³, K Labori ¹³, G Montagnini ¹⁴, L Muñoz-Bellvis ¹⁵, G Nappo ¹⁶, A Nikov ¹⁷, E Pando Rau ¹⁸, M de Pastena ¹⁹, J de la Peña-Moral ²⁰, D Radenkovic ²¹, K Roberts ⁸, R Salvia ²², F Sanchez-Bueno ²⁰, C Scandavini ³, M Serradilla ²³, S Stättner ¹², A Tomazic ²⁴, M Varga ²⁵, H Zavrtanik ²⁶, A Zerbi ¹⁶, M Erkan ²⁷, J Kleeff ²⁸, M Lesurtel ²⁹, MGH Besselink ³⁰, J Ramia-Angel ³¹; European-African Hepato-Pancreato-Biliary Association (E-AHPBA)

Introduction

Current studies on total pancreatectomy (TP) mostly originate from single high-volume centers and span long time periods and therefore may not reflect daily practice. International prospective multicenter short-term studies on clinical outcomes are therefore needed.

Aims & Methods

The aim of this prospective pan-European multicenter snapshot study is to assess short-term clinical outcomes among patients undergoing elective TP. Patients who underwent elective (primary or completion) TP in 43 centers in 16 European countries between June 2018 and June 2019 were included. Major postoperative complications (Cla-vien-Dindo >3), in-hospital mortality, and 90-day mortality were assessed. Subgroup analysis included two cut-off values for the annual volume of pancreatoduodenectomies (< 40 vs. ≥40 and < 60 vs. ≥60). Predictors for major complications and in-hospital mortality were assessed in multivariable logistic regression.

Results

In total, 277 patients underwent total pancreatectomy, mostly for malignant disease (73%). Major postoperative complications occurred in 70 patients (25%). Median hospital stay was 12 days (IQR 9-18) and 40 patients were readmitted within 90 days (15%). In-hospital mortality was 5% and 90-day mortality 8%. in the subgroup analysis, in-hospital mortality was lower in patients operated in centers with >60 pancreatoduodenec-tomies compared to lower-volume centers (4% vs. 10%, p=0.046). in multivariable analysis, annual volume < 60 pancreatoduodenectomies (OR 3.78, 95% CI 1.18-12.16, p=.026), age (OR 1.07, 95% CI 1.01-1.14, p=.046), and estimated blood loss >2L (OR 11.9, 95%CI 2.64-53.61, p=.001) were associated with in-hospital mortality. ASA >3 (OR 2.87, 95%CI 1.56-5.26, p=.001) and estimated blood loss >2L (OR 3.52, 95%CI 1.25-9.90, p=.017) were associated with major complications.

Conclusion

This pan-European prospective snapshot study found a 90-day mortality of 8% after total pancreatectomy. Risk factors for major complications and mortality were identified which may be used to improve outcomes by better patient selection or selective patient referral.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.119

OP129 Selective Histopathological Examination Following Cholecystectomy Is Oncologically Safe: Results of The Multicentre Prospective Fancy Study

VP Bastiaenen ^1,^✉, JLP van Vliet ², EAJ de Savornin Lohman ³, BJGA Corten ⁴, J de Jonge ⁵, AC Kraima ¹, HA Swank ¹, GJD van Acker ², AAW van Geloven ⁵, KH in ‘t Hof ⁶, L Koens ⁷, PR de Reuver ³, CC van Rossem ⁸, GD Slooter ⁴, PJ Tanis ¹, V Terpstra ⁹, MGW Dijkgraaf ¹⁰, WA Bemelman ¹; Dutch Snapshot Research Group

Introduction

Over the past decades, there has been an ongoing debate about whether routine histopathological examination following cholecys-tectomy for benign gallbladder disease is still necessary. in order to save costs and diminish pathologists’ workload, a more selective approach has been suggested, but firm evidence regarding its oncological safety is lacking.

Aims & Methods

The primary aim of this multicentre prospective study was to determine whether selective histopathological examination of cholecystectomy specimens following initial macroscopic assessment by the surgeon is oncologically safe. During a study period of nine months, all gallbladders removed for cholecystitis or gallstone disease were opened and systematically examined by surgeons in 43 Dutch hospitals. Following visual inspection and digital palpation, the surgeon reported whether macroscopic abnormalities suspicious for gallbladder cancer (GBC) were present, and if he or she believed histopathological examination by the pathologist was indicated. Regardless of the surgeon's findings, all gallbladder specimens were sent for histopathological examination. The primary outcome was oncological safety of a selective policy, assessed by calculating the number of patients in whom a histopathological diagnosis of GBC with clinical consequences potentially benefitting the patient would have been missed. All diagnostic or therapeutic procedures performed as a consequence of the diagnosis were considered potentially beneficial. A selective policy was considered safe if, within the group of specimens that would not have been sent for histopathological examination, the upper limit of the two-sided 95% confidence interval of the proportion of missed GBC with clinical consequences potentially benefitting the patient was below 3 per 1000 specimens.

Results

In total, 10,041 patients were included, of whom twenty-eight (0.28%) were diagnosed with GBC. Following predefined systematic macroscopic examination, surgeons considered that 7846 (78.1%) gallbladder specimens could be refrained from histopathological examination. Within this group, eight (0.10%) patients were diagnosed with GBC. The diagnosis led to additional diagnostic or therapeutic procedures in seven patients, implying that in a hypothetical situation of a selective policy, the diagnosis of GBC with clinical consequences potentially benefitting the patient would have been missed in 0.89 per 1000 patients (upper limit 95% CI 1.40 per 1000 patients). None of these patients would have experienced clear prognostic benefit from routine histopathological examination: two patients underwent additional surgery, without radical removal of residual disease, and the remaining five patients only underwent diagnostic procedures without any further treatment.

Conclusion

Selective histopathological examination of gallbladder specimens following initial macroscopic assessment by the surgeon is onco-logically safe and might even prevent patients from harm due to unnecessary surgery. Due to an expected reduction of almost 80% in the number of submitted gallbladder specimens, implementation of a selective policy will likely result in significant cost savings and reduced workload for pa-thologists.

Disclosure

This investigator initiated study was supported by the Netherlands Organisation for Health Research and Development (ZonMw) under grant number 843002822, of which the Dutch Ministry of Health, Welfare and Sports and the Dutch Organisation for Scientific Research are the main commissioning organisations. ZonMw has not played a role in designing the study and data collection, nor has it influenced the analysis and writing process of the abstract or manuscript.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.120

OP130 Long-Term Outcomes and Survival After Endoscopic Or Surgical Resection For T1 Colorectal Cancer: A Multicenter Real-World Observational Study

H Tanaka ^1,^✉, S Kuribayashi ¹, M Sekiguchi ², A Iwamoto ³, Y Hachisu ⁴, Y Fukai ⁵, T Nakayama ⁶, K Furuya ⁷, T Masuda ⁸, K Takahashi ⁹, K Marubashi ¹⁰, T Uraoka ¹; Gunma GI Research Group

Introduction

According to the European Society of Gastrointestinal Endoscopy (ESGE) and the Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines, the curative criteria of endoscopic resection (ER) for T1 colorectal cancer (CRC) are well/moderately differentiated adenocarcinoma or papillary carcinoma, no lympho-vascular invasion, submucosal invasion depth < 1000μm, and budding grade 1. Although an additional surgery is recommended for non-curative ER, the incidence of lymph node metastasis (LNM) is less than 12%. Although some patients do not undergo surgery due to their old age and serious comorbidities and perspective in clinical practice, their clinical outcomes are not fully evaluated.

Aims & Methods

The aim of this study was to clarify clinical outcomes in patients with T1 CRC. A total of 476 consecutive patients were enrolled from April 2009 to August 2019 at 10 institutions. Exclusion criteria included evidence of familial adenomatous polyposis, hereditary nonpolyposis colorectal cancer, or inflammatory bowel disease; presence of active, malignant diseases in any other organs; presence of synchronous or metachronous advanced CRC. Risk factors of LNM and recurrence, and survival were analyzed. Data was shown as mean ± standard deviation.

Results

The mean age of patients was 68.6 ± 10.4 years. The location of the lesion was 333 (68.5%) and 153 (31.5%) in the colon and rectum, respectively. A number of patients with ER without additional SR (ER alone) were 156 (curative resection 78, non-curative resection 78), those with SR (SR alone) were 247, and those with ER and additional SR (ER + additional SR) were 83.

(i) Risk factor of LNM: LNM was found in 10 (12.0%) patients with ER + additional SR and in 34 (13.8%) patients with SR alone. in multivariate analysis, positive lymphatic invasion was the only significant independent risk factor for LNM (OR 6.5, 95% CI [2.87-14.9], p< 0.05).

(ii) Recurrence: During the mean observation period of 1281 ± 922 days, recurrence was found in 15 (3.1%) patients including 6 of lung metastasis, 6 of liver metastasis, 4 of local recurrence, 1 of lymph node recurrence, 1 of peritoneal dissemination and 1 of pelvic recurrence (Including multiple recurrences). No recurrence was found in the ER curative resection group. Although 48% (78/161) patients with ER non-curative resection did not receive additional SR, there was no significant difference in recurrence rate between ER alone with non-curative resection and ER + additional SR groups (6.4% (5/78) vs. 2.4% (2/83), respectively; p=0.27). in the SR alone group, recurrence was found in 3.2% (8/247). The significant independent risk factor for recurrence was budding grade 2/3 (OR 3.8, 95% CI [0.64-22.6], p< 0.05).

(iii) Survival: There were 16 deaths during the observation period, but CRC-related death was found in only 2 cases (0.42%, 2/476); one was ER alone with non-curative resection and the other was SR alone. Deep submucosal invasion, positive lymphatic invasion and LNM were found in the latter case, and subsequently the patient died of recurrence of liver metastasis.

Conclusion

This study suggests that the possibility of expanding the indication for ER of T1 CRC, because positive lymphatic invasion was the only risk factor for LNM and better long-term outcomes were shown despite including significant number of patients with ER alone with non-curative resection.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.121

OP131 Anal Slow Wave Activity in Health and Faecal Incontinence

A Rasijeff ^1,^✉, L Wiklendt ², P Dinning ², C Knowles ¹, SM Scott ¹

Introduction

High-resolution anorectal manometry (HR-ARM) is a diagnostic tool used to assess anal sphincter dysfunction in faecal incontinence (FI). Anal resting pressure, a broad measure of internal anal sphincter (IAS) function, discriminates poorly between health and FI. HR-ARM can also assess rhythmic pressure oscillations, referred to as slow waves, which reflect myogenic activity of the IAS, and may be important for tone generation and the maintenance of high anal resting pressure. However, no HR-ARM study has systematically compared anal slow wave activity between health and FI.

Aims & Methods

Using HR-ARM, our aim was to compare the amplitude of the oscillations of slow waves at a range of frequencies between: i) healthy nulliparous and ii) parous female volunteers (HV); and iii) women with FI. in addition, we determined if the slow waves displayed a direction of propagation at each of the frequencies. Consecutive patients without excluding criteria attending the GI Physiology Unit at the Royal London Hospital (UK) between January 2018 and December 2018 for investigation of FI were considered for inclusion if they had a St Marks’ incontinence score >10. All subjects underwent an HR-ARM study according to the IAPWG protocol¹; recordings with >120 seconds of combined ‘stabilisation’ and ‘resting’ periods were included in the study.

Anal slow wave frequencies were measured between 0.5 and 32 cpm. For each recording channel, a continuous wavelet transform² was computed and the root-mean-square over time and channels calculated to obtain the average amplitude per frequency. The cross-wavelet transform was calculated between adjacent channels, and the time and channel averages of the phase-difference computed to reveal the average direction of propagation per frequency. Results were compared between groups using Bayesian linear regression with Gaussian process responses³.

Results

Nulliparous HV (median: 32, range: 18-67, n=41) were significantly younger than parous HV (45, 31-64, p < 0.01, n=32) and FI (45, 30-72, p < 0.01, n=50). By conventional HR-ARM measures, 20% of patients had hypo-tonia, 38% had hypocontractility, 6% had hypotonia with hypocontractili-ty, and 36% had normal findings. The dominant slow wave frequency in all groups was ∼16 cpm, with a second peak frequency at ∼1.5 cpm. Patients with FI had significantly lower amplitude across all frequencies compared to both parous and nulliparous HV. Parous HV also had a significantly lower amplitude across all frequencies compared to healthy nulliparous females. in all three groups there was a biphasic direction of propagation of slow waves; retrograde at 16 cpm; antegrade at ∼4 cpm. This did not differ between the groups.

[Table. Amplitude of slow wave frequencies between groups]

Anal slow wave frequency (cpm)
Group	0.5	1	1.4	2.8	4	8	16	23	32
Nulliparous HV	1.43 (2.5)	1.89 (2.65)	2.18 (2.36)	2.02 (1.84)	1.87 (1.75)	1.39 (2.04)	3.43 (2.44)	1.49 (2.03)	0.33 (1.81)
Parous HV	1.05 (2.39)	1.34 (2.53)	1.55 (2.33)	1.46 (1.81)	1.35 (1.68)	1.02 (1.86)	2.54 (2.24)	1.11 (1.89)	0.25 (1.7)
FI	0.84 (2.43)	0.98 (2.44)	1.12 (2.39)	1.09 (1.95)	1.05 (1.79)	0.84 (1.93)	2.13 (2.5)	0.94 (2.15)	0.21 (1.77)
Values represent mean amplitude (SD)

Open in a new tab

Conclusion

Wavelet analysis represents a novel, computational method for analysing anal slow waves captured by HR-ARM. The dominant frequency demonstrated herein agrees with findings from conventional ma-nometry⁴ and animal studies⁵. This analysis indicates that slow wave amplitude is reduced in female FI compared to both healthy nulliparous and parous women. Childbirth also appears to reduce slow wave amplitude in comparison to nulliparous women. Retrograde propagation observed at ∼16 cpm may represent an important physiological mechanism to ‘clean’ the anal canal and help maintain continence.

Disclosure

MS has received honoraria for teaching from Laborie. CHK has received financial remuneration from Medtronic Inc. as speaker fees and for expert advisory committees

References

1.Carrington E.V. et al. (2020) The international anorectal physiology working group (IAPWG) recommendations: Standardized testing protocol and the London classification for disorders of anorec-tal function. Neurogastroenterology & Motility 32:e13679. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Wiklendt L. et al. (2020) Autonomic Neuroscience: A Novel Method for Electrophysi-ological Analysis of EMG Signals using MesaClip. Manuscript submitted for publication.
3.Wiklendt L. et al. (2019) Bayesian Functional Mixed-Effects Model with Gaussian Process Responses for Wavelet Spectra of Spatiotemporal Colonic Manometry Signals. arXiv preprint arXiv:1912.02389. Unpublished. [DOI] [PMC free article] [PubMed]
4.Sun W.M., and Rao S. Manometric assessment of anorectal function. Gastroenterology Clinics of North America, 2001; 30(1): 15–32. [DOI] [PubMed] [Google Scholar]
5.Hall K.A. et al. Spatial organization and coordination of slow waves in the mouse anorectum. J Physiol 2014; 592: 3813–3829 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.122

OP132 Topographical Distribution of Microscopic Colitis, Number of Biopsies Necessary For Diagnosis, and Importance of Orientation of Paraffin Embedded Biopsies

A-M Kanstrup Feihn ¹, PJH Engel ², F Lanzarotto ³, D Goudkade ⁴, S Landolfi ⁵, LK Munck ^6,^7,^✉, V Villanacci ⁸

Introduction

The diagnosis microscopic colitis (MC) in patients with chronic watery diarrhea relies on specific histopathological findings in colon biopsies. The number of biopsies and targeted colon segments needed to diagnose MC remains unknown. Consensus on the MC subgroup diagnosis of patients enrolled in a European prospective follow-up study has been established.

Aims & Methods

To determine the number and site of biopsies necessary for the diagnosis and the effect of perpendicular orientation when embedding the biopsies. This multicenter European study included 42 patients with a consensus diagnose of collagenous colitis (CC), 51 with lymphocytic colitis (LC) and three with incomplete lymphocytic colitis. The number of individually diagnostic biopsies from each patient were determined by a retrospective reevaluation of all biopsies from each of these patients.

Results

Overall, the diagnostic rate of 744 individual biopsies from 96 patients with MC was 69.5% for the specific MC subgroup, increasing to 79.4% for MC and 93.4% when including both MC and incomplete MC (MCi). The risk of missing a diagnosis of the specific subgroup of MC when analyzing four biopsies was 0.87% decreasing to 0.18% for MC and 0.0019% if accepting MCi as diagnostic. More biopsies from the right colon were diagnostic for the specific MC subgroup (76.3% vs. 64.0%, p=0.0014). The number of patients with biopsies not diagnostic for the consensus diagnosis in either the right or the left colon was four and six, respectively. Use of cellulose filters for embedding biopsies with a perpendicular orientation resulted in a higher diagnostic rate for the specific MC subgroup (73.1% vs. 65.0%, p=0.0201) mainly related to the diagnosis of CC in the right colon.

Conclusion

Histological changes diagnostic for MC were present in most of the examined biopsies with right colon and orientated biopsies more often being diagnostic. Our findings indicate that four biopsies from the colon, rectum excluded, are sufficient to diagnose MC.

Disclosure

The study was partly funded by a UEG link award

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.123

OP133 Fasting Serum C4 Combined with Primary Fecal Bile Acids From A Single Stool Sample Is Able To Diagnose Patients with Bile Acid Diarrhea

M Camilleri ^1,^✉, P Vijayvargiya ¹, A Taylor ¹, I Busciglio ¹, EV Loftus ¹, L Donato ²

Introduction

Bile acid (BA) diarrhea (BAD) affects 25-33% of patients with diarrhea-predominant irritable bowel syndrome (IBS-D), but requires 48 hour stool collection or ⁷⁵SeHCAT retention test for diagnosis of BAD.

Aims & Methods

To compare the diagnostic yield of combined fasting serum C4 and single stool sample BA to the current diagnostic gold standard for BAD: 48h fecal BA. We studied 25 healthy volunteers (HV,18F), 59 patients with IBS-D (45F), or 4 (all M) with >10cm terminal ileum (TI) resection. Parameters of diarrhea were defined as in a previous population study: stool frequency >3 bowel movements/day and stool consistency >5.5 on Bristol Stool Form Scale. We measured serum C4 and total and individual fecal BAs from a random, single stool sample, and a sample from 48h stool collection. BAD was defined by 3 criteria based on 48h collection: total fecal BA >2,337μmol, total fecal BA >1,000μmol/48h with primary BA >4%, or primary BA >10%. Data are mean±SD and comparisons are based on Kruskal Wallis test. Logistic regression generated ROC curves to determine utility of serum C4 and single stool sample BAs to diagnose BAD.

Results

There were no differences in groups’ average age. BAD was diagnosed in 42% of IBS-D and 100% of TI resection patients. Only 20% (12/59) of patients who presented with diarrhea had elevated serum C4 >52.5ng/ mL, which is less than half of the patients that were diagnosed with IBS-D with BAD based on 48h stool. The results of percentages and concentrations of the primary BAs, DCA, CDCA, and CA, across groups were similar in the random stool sample as observed in the 48h stool sample. Single stool primary BA (%) differentiated BAD compared to IBS-D without BAD and HV (p=0.002). ROC curve analysis demonstrated that C4 and single stool primary BA (%) can diagnose BAD with AUC 0.86, sensitivity 66%, specificity 95%, positive predictive value 86%, and negative predictive value 85%.

Table. Demographics, fasting serum C4 and fecal BA measured from random single and 48hr stool samples]

	Healthy	IBS-D without BAD	IBS-D with BAD	Terminal ileum resection	Kruskal Wallis p
N	25	34	25	4
Age (years)	39.5±10.7	41.8±10.7	46±12.9	50.75±22.6	0.2
Sex (F/M)	18/7	26/8	19/6	0/4	0.01
BMI	24.8±3.1^#	30.6±8.0#	33.5±7.0^	28±6.8	0.0001
Stool frequency (BM/day)	1.3±0.7^*^{^}^#	2.2±1.1^*^{^}^#	3.3±2.0^{^}	5.4±1.7^*	<0.001
Stool consistency (BSFS)	4.1±0.5^*^{^}	4.4±1.1^*^{^}	5.5±0.8^{^}	6.2±0.2^*	<0.001
7 αC4	13.6±9.0^{^}	19.8±14.2^{^}^#	50.5±31.5^{^}^#		<0.001
SINGLE STOOL SAMPLE
Primary BA (%)	2.5±4.2^*	2.2±4.1^*^{^}	19.5±27.3^*^{^}	73.4±18.2^*	0.002
Primary BA (μmol/L)	2.8±8.4^*^{^}	2.9±8.4^*^{^}	45.1±85.2^*^{^}	345.2±188.5^*	<0.001
Total fecal BA (μmol)	212±249^{^}	260±290.4	534.2±561.8^{^}	973.8±1201	0.04
Stool weight (g)	106.8±99.8	88.1±67.7	127±100.9	97±82.8	0.26
48 HOUR STOOL SAMPLE
Primary BA (%)	2.94±5.4^*^{^}	2.1±2.6^*^{^}	27.2±21.3^*^{^}	82.4±15.3^*	<0.0001
Primary BA (μmol/L)	1.6±2.9^*^{^}	2.2±3.0^*^{^}	56.9±69.8^*^{^}	258.5±86.4^*	<0.0001
Total fecal BA (μmol)	585.8±465^*^{^}	718±598.2^*^{^}	1891.6±848^{^}	7204±8500^*	<0.0001
Stool weight (g/48h)	318.9±165^{^}	287.3±123^{^}	527.3±257^{^}	993±950	0.0001

Open in a new tab

^*

significant difference (p< 0.05) to terminal ileum resection patients

^{^}

significant difference (p< 0.05) to IBS-D with BAD

^#

significant difference (p< 0.05) to IBS-D without BAD; BSFS=Bristol Stool Form Scale

Conclusion

Individual fecal BAs measured from a random, single stool sample have similar discriminatory capacity to differentiate between patients with IBS-D with BAD, patients with IBS-D without BAD, and HVs. Combined fasting serum C4 and single stool sample BA is promising to diagnose BAD and should undergo further validation.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.124

OP134 Allergy-Related Diseases During Childhood and Risk of Irritable Bowel Syndrome At 16 Years: A Swedish Birth Cohort Study

J Sjölund ^1,^✉, I Kull ^2,³, A Bergström ^4,⁵, H Törnblom ¹, M Simrén ¹, O Olen ^3,⁶

Introduction

Allergy and immune dysregulation has been suggested to play a role in the pathophysiology of irritable bowel syndrome (IBS), but previous studies on allergy-related diseases and IBS are contradictory.

Aims & Methods

We aimed to examine the association between allergy-related diseases during childhood and IBS at 16 years of age. in this prospective Swedish population-based birth cohort study (BAMSE, N=4089, 49.5% female) children and parents answered questionnaires regarding asthma, allergic rhinitis, eczema, and food hypersensitivity at 1, 2, 4, 8, 12, and 16 years. At 16 years children answered questions based on the Rome III Questionnaire on Pediatric Gastrointestinal Symptoms which allowed children to be categorized into irritable bowel syndrome (IBS), functional abdominal pain, and functional dyspepsia. At 12 and 16 years parents were asked if their child had received a doctor's diagnosis of inflammatory bowel disease or celiac disease. We included children with complete follow up at 16 years. Children who fulfilled the criteria for functional abdominal pain or functional dyspepsia, and children with a parent-reported doctors diagnosis of inflammatory bowel disorder or celiac disease were excluded. Crude and sex-adjusted associations were examined using binominal generalized linear model with log link and described as relative risk (RR/adjusted RR [aRR]) with 95% confidence intervals (CI).

Results

From the original cohort we included 2770 adolescents (49.1% female) with complete follow up at 16 years (excluded: non-responders, n=974; missing data, n=130; inflammatory bowel disease or celiac disease, n=36; functional abdominal pain or functional dyspepsia, n=179). The prevalence of IBS at 16 years was 6.4% (176/2770, 63.6% female). Compared to adolescents without IBS, adolescents with IBS more often had asthma at 12 (11.2% vs 6.7%) and 16 (11.1% vs 6.0%) years, eczema at 16 years (12.7% vs 8.2%) and food hypersensitivity at 12 (40.7% vs 29.2%) and 16 (41.3% vs 25.1%) years. in multivariable regression models, asthma (aRR 1.8; 95% CI 1.2-2.9) and FH (aRR 1.6; 95% CI 1.1-2.2) at 12 years were associated with an increased risk of IBS at 16 years both in crude models and when adjusting for sex. Asthma at 1-2 years (aRR 1.3; 95% CI 0.8-2.2) and eczema at 1-2 years (aRR 1.3; 95% CI 1.0-1.8), 4 years (aRR 1.3; 95% CI 1.0-1.9), and 8 years (aRR 1.4; 95% CI 0.9-1.9) were also associated with an increased risk of IBS at 16 years, but this was not statistically significant (P >0.05 for all analyses). Asthma (RR 1.8; 95% CI 1.2-2.9) eczema (RR 1.5;

95% CI 1.0-2.3) and food hypersensitivity (RR 2.0; 95% CI 1.5-2.7) were all significantly associated with concurrent IBS at 16 years in crude models, but after adjusting for sex, only the associations between asthma (aRR 1.8; 95% CI 1.2-2.8) and FH (aRR 1.9; 95% CI 1.4-2.6) remained significant.

Conclusion

Asthma and food hypersensitivity at 12 years were associated with an increased risk of IBS at 16 years and asthma, eczema, and food hypersensitivity were all associated with an increased risk of concurrent IBS at 16 years. These associations suggest a shared pathophysiology between these disorders and IBS.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.125

OP135 Prevalence and Impact of Overlapping Rome Iv Disorders of Gut-Brain Interactions: Results of A Global Epidemiology Study

AD Sperber ^1,^✉, T Freud ¹, I Aziz ², OS Palsson ³, D Drossman ⁴, DL Dumitrascu ⁵, X Fang ⁶, S Fukudo ⁷, UC Ghoshal ⁸, J Kellow ⁹, R Khatun ¹⁰, E Okeke ¹¹, EM Quigley ¹², MJ Schmulson ¹³, M Simrén ¹⁴, J Tack ¹⁵, WE Whitehead ³, P Whorwell ¹⁶, S Bangdiwala ¹⁰

Introduction

The phenomenon of overlap among Disorders of Gut-Brain Interactions (DGBIs), for example IBS and functional dyspepsia, is long recognized. The co-presence of several DGBIs has been reported as prevalent (1) and has a deleterious effect on disease severity, function, and quality of life. in a recent study conducted in the US, UK, and Canada (2) one third of the participants reported involvement of more than 1 anatomical region (esophageal, gastroduodenal, bowel, and ano-rectal). Overlapping disorders were characterized by increased somatization, poorer QOL, and greater healthcare utilization.

Using the database of the Rome Foundation global epidemiology study (3), we conducted an analysis of 54,127 survey participants to ascertain the degree of overlap among DGBIs and the implications of an increased number of overlapping regions.

Aims & Methods

An Internet survey was completed by 54,127 adults (26,578 females and 27,549 males) in 26 countries, with representative geographic representation, on 6 continents. Quota-based sampling ensured comparable sex ratio and age group distributions among countries. The data were analyzed by anatomical region and associations were assessed between the number of involved regions for each participant (1-4 regions) and the associated variables detailed above. The primary aim was to assess the prevalence and associations of overlapping DGBIs by sex and age. The secondary aims were to assess the association of overlapping DGBIs with (a) quality of life (physical and mental), (b) anxiety, depression, and somatization, and (c) healthcare utilization in the entire study population and with severity of disease in the sub-population of patients with IBS.

Results

40.3% of the 54,127 study participants met the diagnostic criteria for at least one DGBI. The percentages for one, two, three, and four regions were 27.5%, 9.0%, 2.8%, and 0.9%, respectively. All degrees of overlapping regions were more prevalent among women and decreased with age. The table shows the mean scores for severity of disease in the sub-set of individuals meeting IBS criteria, and quality of life, somatization, anxiety, depression, and healthcare utilization (number of medications and number of abdominal surgeries) in the entire study population. As predicted, IBS severity (mean IBS-SSS score) increased incrementally with the number of involved regions, as did the mean scores for somatization, anxiety and depression, number of medications used, and number of abdominal surgeries (P< 0.001 for all variables). The mean scores for quality of life (physical and mental) decreased with an increasing number of overlapping regions (P< 0.001 for both).

Conclusion

A substantial percentage of the adult population surveyed in 26 countries suffered from at least one DGBI, and many of them had DGBIs in more than 1 gastrointestinal region. An increased number of involved regions (overlapping DGBIs) was associated with higher rates of psychological co-morbidity, reduced quality of life, and increased utilization of healthcare services. Among individual with IBS, a higher number of overlapping regions was associated with greater disease severity. Overlapping DGBIs are prevalent and clinically important, so clinicians should identify them and know the implications for the management of these patients.

[Table. IBS severity,quality of life,somatization,anxiety and depression,medications,and surgery scores,by overlap regions (P<0.0001 for all).]

Overlapping regions	IBS-SSS	PROMIS-10 physical QOL	PROMIS-10 mental QOL	PHQ-12 somatization	PHQ-4 anxiety/depression	Different types of medications	Abdominal surgeries
No DGBI	Irrelevant	15.15	14.24	4.35	2.02	0.80	0.28
1	207.59	13.89	12.89	6.28	3.37	1.31	0.31
2	244.27	12.79	11.97	8.05	4.51	2.08	0.37
3	269.37	12.04	11.39	9.45	5.35	2.82	0.44
4	291.64	11.30	11.09	11.35	6.05	3.89	0.53

Open in a new tab

Disclosure

Nothing to disclose

References

1.Locke G.R. 3rd, Zinsmeister A.R., Fett S.L., Melton L.J. 3rd, Tal-ley N.J. Overlap of gastrointestinal symptom complexes in a US community. Neurogastroenterology and motility: the official journal of the European Gastrointestinal Motility Society. 2005; 17: 29–34. [DOI] [PubMed] [Google Scholar]
2.Aziz I., Palsson O.S., Törnblom H., Sperber A.D., Whitehead W.E., Simrén M. The Prevalence and Impact of Overlapping Rome IV-Diagnosed Functional Gastrointestinal Disorders on Somatization, Quality of Life, and Healthcare Utilization: A Cross-Sectional General Population Study in Three Countries. American Journal of Gastroenterology. 2018; 113: 86–96. [DOI] [PubMed] [Google Scholar]
3.Sperber A.D., Bangdiwala S.I., Drossman D.A., Ghoshal U.C., Imren M., Tack J. et al. Worldwide Prevalence and Burden of Functional Gastrointestinal Disorders, Results of Rome Foundation Global Study. Gastroenterology. 2020. doi: 10.1053/j.gastro.2020.04.014 [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.126

OP136 Treatment Outcome of Tofacitinib Dose Reduction To 5 Mg Bid Vs Remaining On 10 Mg Bid in Patients with Uc Who Were in Stable Remission On 10 Mg Bid: 6-Month Data From The Double-Blind, Randomised Riveting Study

S Vermeire ^1,^✉, C Su ², N Lawendy ², T Kobayashi ³, WJ Sandborn ⁴, DT Rubin ⁵, I Modesto ⁶, S Gardiner ⁶, N Kulisek ⁷, H Zhang ⁷, W Wang ⁷, J Panés ⁸

Introduction

Tofacitinib is an oral, small molecule JAK inhibitor for the treatment of ulcerative colitis (UC). Efficacy and safety of tofacitinib were evaluated in Phase 3 trials, including maintenance treatment with tofacitinib 5 and 10 mg twice daily (BID) [1,2]. We present primary completion analysis of the RIVETING study (NCT03281304), designed to evaluate the efficacy and safety of dose reduction to 5 mg BID vs remaining on 10 mg BID in patients (pts) with UC in stable remission on tofacitinib 10 mg BID maintenance therapy.

Aims & Methods

RIVETING is an ongoing, double-blind, randomised, parallel-group study. Eligible pts had received tofacitinib 10 mg BID for ≥ 2 consecutive years in an open-label, long-term extension study (NCT01470612), and been in stable remission on this dose for ≥ 6 months prior to enrolment. The aim of RIVETING was to estimate treatment difference between tofacitinib 5 and 10 mg BID doses. This primary analysis was conducted when all pts completed their Month 6 study visit. Efficacy was assessed at Month 6; safety was assessed throughout.

Results

In total, 140 pts were randomised (1:1) to tofacitinib 5 or 10 mg BID. 77.1% and 90.0% of pts were in remission based on modified Mayo score at Month 6 (primary endpoint) in the 5 and 10 mg BID groups, respectively (adjusted difference 12.9%; 95% confidence interval 0.5, 25.0); when analysed by subgroup (baseline endoscopic subscore; prior tumour necrosis factor inhibitor [TNFi] treatment failure), differences ranged from 9.5% to 21.1%. Consistent dose-group differences were seen for the secondary endpoints of remission and clinical response, according to total Mayo score, and mucosal healing. Rates of adverse events (AEs) and serious AEs were similar across dose groups. AEs of special interest included (Table): serious infections (5 mg BID, n=2); herpes zoster (5 mg BID, n=1; 10 mg BID, n=3; all non-serious); pulmonary embolism (10 mg BID, n=1). There were no cases of deep vein thrombosis or death.

Conclusion

These 6-month data from RIVETING showed that most pts in stable remission on tofacitinib 10 mg BID maintenance therapy maintained remission after dose reduction to 5 mg BID. Differences between dose groups were 11.4% to 12.9% for various efficacy endpoints, favouring

[OP136 Table. Summary of efficacy and safety in the RIVETING study]

Efficacy at Month 6 (FAS, NR^b)	Tofacitinib 5 mg BID (N=70) n/N (%)	Tofacitinib 10 mg BID (N=70) n/N (%)	Difference % (95% CI)
Primary efficacy endpoint
Modified Mayo score remission^c	54/70 (77.1)	63/70 (90.0)	12.9 (0.5, 25.0)^d
Stratified by subgroup:
Baseline endoscopic subscore = 0	42/51 (82.4)	47/51 (92.2)	9.8 (-3.0, 22.6)
Baseline endoscopic subscore = 1	12/19 (63.2)	16/19 (84.2)	21.1 (-6.1, 48.2)
Prior TNFi treatment failure: No	34/43 (79.1)	31/35 (88.6)	9.5 (-6.6, 25.6)
Prior TNFi treatment failure: Yes	20/27 (74.1)	32/35 (91.4)	17.4 (-1.6, 36.3)
Select secondary efficacy endpoints
Remission^e	53/70 (75.7)	61/70 (87.1)	11.4 (-1.5, 24.1)
Mucosal healing^f	56/70 (80.0)	64/70 (91.4)	11.4 (-0.3, 23.1)
Clinical response^g	59/70 (84.3)	67/70 (95.7)	11.4 (1.3, 22.0)
Safety outcomes (SAS)	Tofacitinib 5 mg BID^h (N=70)	Tofacitinib 10 mg BID (N=70)	Total (N=140)
Median treatment duration, days (range)	538 (29-722)	529 (174-804)	537 (29-804)
Deaths, n (%)	0 (0.0)	0 (0.0)	0 (0.0)
TEAEs, n (%)
Pts with any AEs	46 (65.7)	49 (70.0)	95 (67.9)
Pts with SAEs	4 (5.7)	4 (5.7)	8 (5.7)
Pts with discontinuations due to AEsⁱ	7 (10.0)	2 (2.9)	9 (6.4)
Most common TEAEs by SOC, n (%)
Infections and infestations	20 (28.6)	22 (31.4)	42 (30.0)
Gastrointestinal disorders	16 (22.9)	23 (32.9)	39 (27.9)
TEAEs of special interest, n (%)
Serious infections	2 (2.9)^j	0 (0.0)	2 (1.4)
All herpes zoster^k	1 (1.4)	3 (4.3)	4 (2.9)
Opportunistic infections^l,m	0 (0.0)	1 (1.4)ⁿ	1 (0.7)
MACE^l	1 (1.4)^o	0 (0.0)	1 (0.7)
Malignancies (excl. NMSC)^l	1 (1.4)^p	0 (0.0)	1 (0.7)
NMSC^l	1 (1.4)	0 (0.0)	1 (0.7)
Gastrointestinal perforations^l,q	0 (0.0)	0 (0.0)	0 (0.0)
Deep vein thrombosis	0 (0.0)	0 (0.0)	0 (0.0)
Pulmonary embolism	0 (0.0)	1 (1.4)	1 (0.7)

Open in a new tab

^a

Pts were required to have been in stable remission for ≥6 months prior to enrolment. Stable remission was defined as meeting the following criteria: a partial Mayo score of ≤2 with no individual subscore >1, and an RB subscore of 0, at each study visit where data were available during the 6-month period in the open-label, long-term extension study, prior to and including baseline of the current study; at least 1 assessment of remission based on total Mayo score (a confirmed, locally read endoscopic subscore of 0 or 1) was required in the 6 months prior to randomisation, and all assessments based on Mayo score during this period were required to show remission; no corticosteroid use to treat ulcerative colitis for at least 4 weeks prior to baseline;

^b

Pts with missing scores were treated as non-responders. Pts in the tofacitinib 5 mg BID group with dose escalation due to flare were treated as non-responders for visits after dose escalation

^c

Defined as an endoscopic subscore of ≤1 with a stool frequency subscore of ≤1, and an RB subscore of 0

^d

Difference is weighted difference based on the Cochran-Mantel-Haenszel weight method and 95% CI using the Newcombe method, stratified by baseline endoscopic subscore (0,1)

^e

Defined as a total Mayo score of ≤2 with no individual subscore >1, and an RB subscore of 0

^f

Defined as a Mayo endoscopic subscore of ≤1, as per the OCTAVE Open protocol (NCT01470612) approved prior to publication of the US Food and Drug Administration draft guidance on the definition of mucosal healing

^g

Defined as a decrease from induction study baseline total Mayo score of ≥3 points and ≥30%, plus a decrease in RB subscore of ≥1 point or an absolute RB subscore of ≤1

^h

Events reported after dose escalation were not included in dose group analyses of tofacitinib 5 mg BID

ⁱ

Pts who have an AE record that indicates that the AE caused the pt to be discontinued from the study

^j

Serious infections (number of events): cellulitis (1), urinary tract infection (1)

^k

Non-serious and serious herpes zoster; no cases were classed as serious

^l

Adjudicated events

^m

Excludes tuberculosis and herpes zoster with 2 adjacent dermatomes

ⁿ

Herpes zoster (non-adjacent or >2 adjacent dermatomes)

^o

MACE (number of events): cerebrovascular accident (1)

^p

Malignancy (number of events): vulvar cancer (1)

^q

Gastrointestinal perforation excludes preferred terms of pilonidal cyst, perirectal abscess, rectal abscess, anal abscess, perineal abscess and any preferred terms containing the term fistula AE, adverse event; BID, twice daily; CI, confidence interval; FAS, full analysis set; MACE, major adverse cardiovascular events; N, number of pts treated in the treatment group; n, number of unique pts with a particular event; NMSC, non-melanoma skin cancer; NR, non-responders; pts, patients; RB, rectal bleeding; SAE, serious AE; SAS, safety analysis set; SOC, system organ class; TEAE, treatment-emergent AE; TNFi, tumour necrosis factor inhibitor

the 10 mg BID group. For pts who reduced dose to 5 mg BID, pts with baseline endoscopic subscore 0 and those without prior TNFi failure were more likely to maintain remission vs pts with subscore 1 and pts with prior TNFi failure, respectively. No new safety risks were identified in this limited dataset.

Disclosure

S Vermeire has received research support and consultancy and speakers’ bureau fees from AbbVie, Janssen, MSD, Pfizer Inc and Takeda; consultancy and speakers’ bureau fees from Ferring and Hospira; consultancy fees from Abivax, Arena Pharmaceuticals, Celgene, Eli Lilly, Galapagos, Genentech/Roche, Gilead, GSK, Mundipharma, Progenity, Second Genome and Shire; and speakers’ bureau fees from Tillotts. T Ko-bayashi has received lecture fees from AbbVie, Alfresa Pharma Corporation, Janssen, Mitsubishi Tanabe Pharma, Mochida Pharmaceutical Co., Ltd., Pfizer Japan Inc and Takeda; consigned/research expenses from EA Pharma and Nippon Kayaku; and has a course affiliation with AbbVie, EA Pharma, JIMRO Co., Ltd., Otsuka Holdings Co., Ltd. and Zeria Pharmaceutical Co., Ltd. WJ Sandborn has received grants and consultancy fees from AbbVie, Amgen, Eli Lilly, Genentech, Gilead, Janssen, Pfizer Inc, Prometheus Laboratories (now Prometheus Biosciences) and Takeda; grants from Atlantic Healthcare Limited and Celgene/Receptos; consultancy fees from Allergan, Arena Pharmaceuticals, Avexegen Therapeutics, BeiGene, Boehringer Ingelheim, Celgene, Celltrion, Conatus, Cosmo, Escalier Biosciences, Ferring, Forbion, Gossamer Bio, Incyte, Kyowa Kirin, Landos Bio-pharma, Oppilan, Otsuka, Reistone, Ritter Pharmaceuticals, Robarts Clinical Trials, Seres Therapeutics, Shire, Sienna Biopharmaceuticals, Sigmoid Biotechnologies, Sterna Biologicals, Sublimity Therapeutics, Theravance Biopharma, TiGenix, Tillotts, UCB, Ventyx Biosciences, Vimalan Biosciences and Vivelix Pharmaceuticals; and is a shareholder of BeiGene, Escalier Biosciences, Gossamer Bio, Oppilan, Prometheus Biosciences, Ritter Pharmaceuticals, Ventyx Biosciences and Vimalan Biosciences. DT Rubin has received grants and personal fees from AbbVie, Genentech/Roche, Jans-sen, Prometheus Laboratories and Takeda; personal fees from AbGenom-ics, Allergan, Biomica, Boehringer Ingelheim, Bristol-Myers Squibb, Cel-gene, Check-Cap, Dizal Pharma, Eli Lilly, GalenPharma/Atlantica, Gilead, Glenmark, GSK, Mahana Therapeutics, Narrow River Management, Pfizer Inc, Reistone, Seres Therapeutics, Shire, TARGET PharmaSolutions and Techlab; and other fees from ACG, Cornerstones Health and GoDuRn. C Su, N Lawendy, I Modesto, S Gardiner, N Kulisek, H Zhang and W Wang are employees and shareholders of Pfizer Inc. J Panés has received grants and personal fees from AbbVie and MSD, and personal fees from Arena Pharmaceuticals, Boehringer Ingelheim, Celgene, Genentech/Roche, Gilead, GoodGut, GSK, Immunic, Janssen, Nestlé, Oppilan, Pfizer Inc, Progenity, Robarts Clinical Trials, Takeda, Theravance Biopharma and TiGenix.

References

1.Sandborn et al. N Engl J Med 2017; 376: 1723–36. [DOI] [PubMed] [Google Scholar]
2.Lichtenstein et al. J Crohns Colitis 2020; 14(Suppl 1): DOP61. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.127

OP137 Single-Cell Transcriptomics in Peripheral Blood Mononuclear Cells Reveals Cell Population Differences That Define Vedolizumab Response Among Cd Patients

AYF Li Yim ^1,^✉, I Hageman ², V Joustra ³, MMAM Mannens ¹, GR D'Haens ³, WJ de Jonge ³, P Henneman ¹

Introduction

Vedolizumab (VDZ) is one of the more recent additions to the repertoire of biologicals used to ameliorate inflammation in Crohn's disease (CD). Despite its efficacy, non-response to VDZ remains rife. While epigenetic and transcriptomic blood-based biomarkers have been reported, interpreting the biology thereof has been hampered by the fact that such experiments were performed on bulk material and could therefore be confounded by differences in cell populations.

Aims & Methods

In this study, we investigated whether VDZ response was associated with differences in the cellular composition by using single-cell transcriptomics. VDZ response assessment was performed during routine care at a median of 42 weeks into treatment and was defined as endoscopic- (≥50% drop in SES-CD) in combination with biochemical (≥50% reduction in CRP and fecal calprotectin) and/or (steroid-free) clinical response (≥3 point drop in HBI). Peripheral blood mononuclear cells (PBMCs) were isolated at a median of 54 weeks into treatment from 4 responders (R) and 4 non-responders (NR). Single cells were isolated and prepared in a multiplexed fashion using cell hashing antibodies on the Chromium 10X. Sequenced on the HiSeq4000 to a total depth of 936M reads. Reads were aligned using Cellranger and analyzed in R using Seurat.

Results

Unsupervised graph-based clustering analysis revealed 4 distinct clusters, which we identified as the T-cells (CD3D and IL7R), B-cells (MS4A1 and CD19), mononuclear phagocytes (MNP; CST3, HLA-DRA, CD14 and FC-GR3A) and thrombocytes (PPBP). Non-responding patients presented on average 12% more T-cells and 4.3% fewer MNPs. Further analysis indicated that the T-cell signal was largely driven by a 14% larger population of the T-helper cells (CD4 and NFKB1) among non-responders, whereas the MNP signal was driven by a concordant decrease in the proportions of both dendritic cells (FCER1A, CLEC4C and IL3RA) and monocytes (CD14 and FCGR3A) by 1% and 3%, respectively.

Conclusion

Our data suggest that patients not responding to VDZ display decreased concentrations of innate immune cells. While our observations provide an avenue into understanding the various manifestations of VDZ response, an important consideration is that we cannot ascertain whether the differences are attributable to VDZ treatment or the overall reduction in inflammation. Further research is necessary to identify whether the aforementioned differences in cell population are associated with VDZ treatment or are due to persistent inflammation.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.128

OP138 Diet-Induced Remission in Pediatric Crohn'S Disease Is Associated with A Marked Decrease of Kynurenine, A Tryptophan Metabolite Associated with Inflammation

M Ghiboub ^1,^2,^✉, CM Verburgt ^1,², S Penny ³, A Levine ⁴, R Sigall Boneh ⁴, E Wine ⁵, A Cohen ⁶, KA Dunn ⁷, D Pinto ³, MA Benninga ², WJ de Jonge ^1,⁸, J Van Limbergen ^1,^2,⁹

Introduction

Both Crohn's Disease Exclusion Diet (CDED) and exclusive enteral nutrition (EEN) reduce dietary exposure to foods that might negatively impact the microbiome, the intestinal barrier and innate immunity (1). These diets are able to induce remission in pediatric Crohn's disease (CD) patients, but CDED is better tolerated and able to sustain remission (1). Although the specific mechanisms of nutritional therapy are still unknown, microbiome community changes are strongly associated with the therapeutic outcome (1,2). Metagenome studies have associated CD with microbial amino acids metabolism (2,3). Tryptophan (Trp) is an essential amino acid, that is obtained from protein-rich diets (4). Trp is necessary for protein synthesis and the formation of serotonin and melatonin. More than 90% of dietary Trp is metabolized through the kynurenine pathway (4). Trp metabolites such as kynurenine have been suggested to play a central role in modulating the intestinal immune response, notably via the aryl hydrocarbon receptor (AhR) (4,5).

Aims & Methods

We investigated changes in 16 Trp metabolites in stool samples from a 12-week prospective trial comparing two different nutritional therapies (CDED + Partial enteral nutrition (EEN) vs EEN) for induction of remission in mild to moderate pediatric CD. After 6 weeks, the EEN group could return to free diet with 25% of calories from PEN. Endpoints of the study were intention to treat (ITT) remission at week 6 (defined by Pediatric Crohn's Disease Activity Index (PCDAI) < 10) and corticosteroid-free ITT sustained remission at week (W)12. Trp metabolites at W0, W3, W6 and W12 of 102 samples were measured by liquid chromatography coupled to mass spectrometry and analyzed according to clinical outcome groups of sustained remission, non-sustained remission and no-remission.

Results

There were no significant differences in PCDAI score or tryp-tophan metabolites between CDED and EEN samples at baseline. CDED induced remission in 15 out of 17 patients (88%) at week 6 (n=16, 9, 17, 9 at W0, W3, W6, W12, respectively), which was maintained in 8 out of 9 patients (88%) at week 12; in the EEN group 9 out of 14 patients (64%) achieved remission at week 6 (n=14, 11, 14, 12 at W0, W3, W6, W12, respectively) and 7 out of 12 patients (55%) maintained remission until week 12. Pediatric CD patients showed higher fecal levels of kynurenine, a tryp-tophan metabolite associated with inflammation. Kynurenine decreased significantly (p< 0.05) by week 3 in the EEN group. in CDED, the drop in kynurenine became more significant with duration of therapy (p< 0.05 at week 3, p< 0.01 at week 6 and week 12). This decrease in kynurenine was strongly associated with achieving remission. There were no significant changes in other Trp metabolites.

Conclusion

CDED- and EEN-induced remission is associated with a strong reduction in kynurenine. CDED-induced sustained remission proposes kynurenine reduction as one of the factors mediating diet-induced remission. Further studies are warranted to explore the effect of kynurenine on Ahr-signaling in dietary treatment of CD.

Disclosure

Nothing to disclose

References

1.Levine A., Wine E., Assa A. et al. Crohn's Disease Exclusion Diet Plus Partial Enteral Nutrition Induces Sustained Remission in a Randomized Controlled Trial. Gastroenterology. 2019; 157(2): 440–450.e8. doi: 10.1053/j.gastro.2019.04.021 [DOI] [PubMed] [Google Scholar]
2.Jones C.M.A., Connors J., Dunn K.A. et al. Bacterial Taxa and Functions Are Predictive of Sustained Remission Following Exclusive Enteral Nutrition in Pediatric Crohn's Disease. Inflamm Bowel Dis. 2020; izaa001. doi: 10.1093/ibd/izaa001 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Douglas G.M., Hansen R., Jones C.M.A. et al. Multi-omics differentially classify disease state and treatment outcome in pediatric Crohn's disease. Microbiome. 2018; 6(1): 13. Published 2018 Jan 15. doi: 10.1186/s40168-018-0398-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Agus A., Planchais J., Sokol H. Gut Microbiota Regulation of Tryptophan Metabolism in Health and Disease. Cell Host Microbe. 2018; 23(6): 716–724. doi: 10.1016/j.chom.2018.05.003 [DOI] [PubMed] [Google Scholar]
5.Kaiser H., Parker E., Hamrick M.W. Kynurenine signaling through the aryl hydrocarbon receptor: Implications for aging and healthspan. Exp Gerontol. 2020; 130: 110797. doi: 10.1016/j.exger.2019.110797 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.129

OP139 Increasing Use of Biologics Did Not Reduce Hospitalization and Surgery Rates in Paediatric Ibd: A Population-Based Study From The Epi-Iirn Database

D Turner ¹, N Asayag ², A Mendelovici ², G Focht ^2,^✉, E Matz ³, A Cahan ⁴, N Ledderman ⁵, Z Haklai ⁶, I Dotan ⁷, Y Chowers ⁸

Introduction

Biologics are being increasingly used and are becoming the main factor in the increased costs of paediatric IBD (PIBD) treatment. However, their ability to change the natural history of the disease has never been proven in children.

Aims & Methods

We aimed to explore in this population-based study the temporal association between utilization of biologics and the rates of hospitalization and surgeries of PIBD in Israel.

We also report contemporary rates of steroid dependency and mortality among IBD children.

The analysis was performed on data from the epidemiology cohort of the Israeli IBD Research Nucleus (epi-IIRN). The epi-IIRN cohort is a validated population-based cohort including clinical and accurate drug prescription data of all IBD patients in Israel.

This study utilized data from 3/4 national health maintenance organizations (HMOs) covering 48% of the population of Israel. Children diagnosed with IBD prior to 18 years of age were included. Each child was individually matched by age, jurisdiction, HMO and gender to 3 non-IBD random controls. The two cohorts (i.e. cases and controls) were linked to the National Israeli Registries that record real-life reports of all hospitalizations and surgeries in Israel. We excluded outcomes occurring prior to the index date (i.e. IBD diagnosis in the cases and the identical age in the matched controls).

Results

Of 16,004 included IBD patients, 2,555 were diagnosed during childhood and included in the current analysis: 1,735 with Crohn's disease (CD) and 820 with ulcerative colitis (UC). There were 8,358 patient-years followed by 18 years of age in the IBD group and 7,343 non-IBD matched controls.

Hospitalizations in the CD and UC groups were recorded by 18 years of age in 1,131 (65%) and 302 (37%), respectively, as compared with 771 (10%) in the non-IBD matched controls (p< 0.001). Similarly, IBD-related surgeries were more common in CD (n=313 (18%)) and UC (n=49 (6%)) than controls (n=76 (1%); p< 0.001.

Biologics were ever used by 18 years of age in 693 (40%) children with CD and 147 (19%) with UC, 57% of the total did so during the first year after diagnosis. Despite a sharp increase in biologics utilization, hospitalization and surgery rates did not decline from 2005, when biologics entered the standard treatment algorithms, to 2018.

Steroid dependency defined as >3 months treatment was ever noted in 246 children with CD (19%) and 150 with UC (19%). There were 5 deaths in the IBD group (0.2%) by 18 years of age (only one of whom possibly related to IBD or its treatment with gastrointestinal hemorrhage), compared with 4/7276 (0.05%) in the controls (p=0.043).

Conclusion

In this large population-based study, hospitalizations and surgery rates did not decline, despite extensive biologics use.

Disclosure

This study was funded by the Helmsley charity fund. Last 3 years DT received consultation fee, research grant, royalties, or honorarium from Janssen, Pfizer, Hospital for Sick Children, Ferring, Abbvie, Take-da, Atlantic Health, Shire, Celgene, Lilly, Neopharm, Roche, ThermoFisher

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.130

OP140 Efficacy and Safety of Adalimumab in Pediatric Patients with Moderate To Severe Ulcerative Colitis: Results of A Randomized-Controlled Phase 3 Study

NM Croft ^1,^2,^✉, WA Faubion ³, S Kugathasan ⁴, J Kierkuš ⁵, F Ruemmele ⁶, NM Mostafa ⁷, M Venetucci ⁷, T Finney-Hayward ⁷, M Bereswill ⁸, A Lazar ⁸, D Turner ⁹

Introduction

Currently, biologic treatment options are limited for children with ulcerative colitis (UC).

Aims & Methods

This study assessed efficacy and safety of adalimumab in children with moderate to severe UC; 93 children (4-17 years) were enrolled (intention-to-treat [ITT]/safety population). During 8-wk induction period, 77 patients (pts) were randomized to receive double-blind adalimumab 2.4 mg/kg (max 160 mg) at wk 0, and wk 1 (high dose) or matching placebo (PBO) at wk 1 (standard dose), 1.2 mg/kg (max 80 mg) at wk 2, and 0.6 mg/kg (max 40 mg) at wks 4 and 6, and analyzed for primary wk 8 efficacy endpoint (ITT-E); 16 pts received open-label high-dose adalimumab after protocol change. of all wk 8 responders per Partial Mayo Score (PMS; defined as decrease of ≥2 points and ≥30% from baseline in PMS) in the ITT population, 62 pts were randomized to adalimumab high (0.6 mg/ kg every wk) or standard dose (0.6 mg every other wk) and were analyzed for primary wk 52 endpoints (mITT-E); 12 pts randomized to PBO were excluded from wk 52 primary analyses per protocol change to remove PBO arm. Nonresponders at wk 8 were discontinued. Co-primary endpoints were proportion of pts achieving PMS remission (defined as PMS ≤2 and no individual subscore >1) at wk 8 and Full Mayo Score (FMS) remission (defined as FMS <2 and no individual subscore >1) at wk 52 among wk 8 responders. Ranked secondary endpoints at wk 52 were percentage of pts who achieved FMS clinical response among wk 8 responders, mucosal healing (Mayo endoscopy subscore ≤1 by central reading) among wk 8 responders, FMS remission among wk 8 PMS remitters, and corticosteroid-free FMS remission among wk 8 responders. Adalimumab groups were tested against historical adult PBO rates per protocol amendment.

Results

Among the ITT population, mean age was 14.1 yrs, 55% were female, mean duration of UC was 2.3 yrs, 47% had systemic corticosteroid use, 59% had immunosuppressant use, and 84% were tumor necrosis factor inhibitor naïve at baseline. After 8-wk induction, 53% of pts in the ITT-E population achieved PMS remission (60% with high-dose and 43% with standard-dose adalimumab; Table). At wk 8, 74/93 (80%) ITT pts were responders. During the maintenance period, among the 62 mITT-E pts receiving high- or standard-dose adalimumab, a significantly higher percentage of pts achieved all wk 52 co-primary and secondary endpoints (except corticosteroid-free FMS remission) both in the pooled and high-dose adalimumab groups versus historical PBO. in the safety population, 56% of pts reported ≥1 adverse event (AE) and 11% reported a serious AE during the induction period. Throughout any adalimumab exposure in the study, 78% and 23% of pts experienced an AE and serious AE, respectively. No deaths, malignancies, active TB, or demyelinating disease were observed in the study.

[Table. Co-Primary and Ranked Secondary Endpoints.]

Endpoints, mean (n)	ADA Pooled	ADA high dose	ADA standard dose	External Placebo
PMS Remission, wk 8, ^**	53.2^* (n=77)	59.6^* (n=47)	43.3 (n=30)	19.8
FMS Remission, wk 52, ^**	37.1^* (n=62)	45.2^* (n=31)	29.0 (n=31)	18.4
FMS Response, wk 52, %	64.5^* (n=62)	67.7^* (n=31)	61.3^* (n=31)	26.1
Mucosal healing, wk 52, %	45.2^* (n=62)	51.6^* (n=31)	38.7 (n=31)	22.0
FMS remission in remitters, wk 52, ^***	44.2^* (n=43)	45.5^* (n=22)	42.9 (n=21)	14.8
Corticosteroid-free FMS remission, wk 52, %	31.0 (n=29)	31.3 (n=16)	30.8 (n=13)	24.1

Open in a new tab

ADA, adalimumab; FMS, Full Mayo Score; PBO, placebo; PMS, Partial Mayo Score. Analysis populations: ITT-E for week 8 endpoint and mITT-E for week 52 endpoints; all wk 52 analyses were among wk 8 responders unless noted otherwise.

^*

Statistically significant (alpha 5%) per pre-specified sequentially rejective multiple test procedure using 1-sample 2-sided chi-square tests of respective ADA group versus external placebo.

^**

Co-primary endpoints.

^***

FMS remission among wk 8 remitters.

Conclusion

Clinically meaningful rates of remission/response and mucosal healing were achieved by children with moderate to severe UC receiving adalimumab. No new safety signals were observed, suggesting that adalimumab is an efficacious and safe treatment option for children with moderate to severe UC.

Disclosure

N.M. Croft: Dr. Croft's institution received speaker fees, advisory board fees, and research funding from Abbvie, Eli-Lilly, Takeda, Shire, Pfizer, 4D Pharma. W.A. Faubion Jr: has received consulting fees from Boehringer Ingelheim, Takeda Pharmaceuticals International, AbbVie, Janssen and Robarts. S. Kugathasan: has commercial relationship/interests with Jessen and Takeda. J. Kierkus: has received consultation fees, research grants, or honoraria from Janssen, Abbvie, Takeda, Egis, Nestle, and Nutricia, F.M. Reummele: has received grants and research support from Nestlé Nutrition Institute, AbbVie, MSD, and Janssen; served as a member of advisory boards for Centocor (DEVELOP), AbbVie (CAPE and LEA), MSD France (SAC), Nestlé Nutrition Institute, Nestlé Health Science, Danone, Mead Johnson, Nutricia, Takeda, Celgene, Biogen, Shire, Pfizer, and Therakos; and has received payment and honorarium for lectures from AbbVie, Danone, Nutricia, and Nestlé. N.M. Mostafa, M. Venetucci, T. Finney-Hayward, M. Bereswill, and A. Lazar are full-time AbbVie employees and may own AbbVie stock and/or options. Dan Turner: has received consultation fees, research grant, royalties, or honorarium from Janssen, Pfizer, Hospital for Sick Children, Ferring, AbbVie, Takeda, Atlantic Health, Shire, Celgene, Lilly, Roche, ThermoFisher, and BMS. ACKNOWLEDGMENTS The authors would like to thank Nisha Kwatra for contributing to the study. AbbVie funded the study, contributed to its design, and was involved in the collection, analysis, and interpretation of the data, and in the writing, review, and approval of the abstract. Medical writing support was provided by Maria Hovenden, PhD, of ICON plc (North Wales, PA), which was funded by AbbVie.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.131

OP141 The Incremental Benefit of Dye-Based Chromoendoscopy in Addition To High-Definition White Light and Virtual Chromoendoscopy For Lesion Assessment and Prediction of Submucosal Invasion

M Sidhu ^1,^2,^✉, A van Hattem ¹, S Vosko ³, N Shahidi ¹, DJ Tate ⁴, MJ Bourke ^1,²

Introduction

Analysis of the surface pit pattern (SPP), of large (>/=20mm) laterally spreading colonic lesions (LSL) can help predict the risk of sub-mucosal invasion (SMI) and identify lesions not suitable for piecemeal endoscopic mucosal resection (EMR). Expert opinion mandates the use of dye-based chromoendoscopy for a reliable assessment of SPP. However, in the era of advanced optics the utility of virtual chromoendoscopy (VC), combined with high definition white light (HDWL) for lesion assessment remains unknown.

Aims & Methods

We sought to assess the incremental benefit of dye-based chromoendoscopy in addition to HDWL plus VC for the assessment of SPP and prediction of SMI in colonic LSL referred for EMR. A prospective observational study of consecutive lesions referred for EMR at a single tertiary referral centre was performed (NCT03506321). Prior to resection all lesions were assessed for the following surface features; SPP as per the Kudo classifcation and an area of demarcation [where a regular neoplastic pit pattern (Kudo III/IV) became disordered (Kudo V)], initially performed using HDWL + VC [Narrow band imaging (NBI),Olympus,Tokyo)] by two trained independent blinded observers. The results were recorded by an independent observer. Thereafter, indigo-carmine (0.2%) was sprayed directly onto the lesion surface and a repeat assessment performed by the same blinded observers. Overall inter-observer agreement was calculated, and significance reported using kappa co-efficient.

Results

Over 22 months to September 2019, 205 consecutive LSL in 205 patients (50.7% - male) were enrolled. Median lesion size was 38mm (IQR: 30-50mm), 46.8% - right colon. The overall rate of SMI was 9.2% (19/205). Presence of a demarcated area on the lesion surface for all lesions had a NPV of [95.1%,95%CI (90.5-97.6)] for predicting SMI at histology. There was no incremental benefit from the addition of dye-based chromoendoscopy [94.6%,95%CI (90.0-97.3)]. 23/205 (11.2%) LSL contained a demarcated area with HDWL plus VC and 20/205 with addition of dye-based chromoendoscopy.

In addition, there was a high rate of interobserver agreement as to the presence of a demarcated area between the two blinded observers, independent of whether dye-based chromoendoscopy was used (Table 1); k-0.98 (SE - 0.03) with HDWL plus VC and k-0.96 (SE - 0.03) with HDWL plus VC and dye-based chromoendoscopy.

Table 1.

[Overall agreement & Diagnostic Accuracy; VC - Virtual Chromoendoscopy,CI - conifdence interval,HDWL - High-definition white light]

Assessment of Demarcated Area	HDWL + VC (95% CI)	HDWL + VC + Dye-Based Chromoendoscopy (95% CI)
Overall Agreement, % (95% CI)	99.5 (97.3-99.9)	99.0 (96.5-99.7)
Kappa (Standard Error)	0.98 (0.03)	0.95 (0.03)
Assessment of submucosal invasion (SMI), n=205
HDWL + VC		HDWL + VC + Dye-Based Chromoendoscopy
Sens (95% CI) / Spec (95% CI)	52.6% (29.5-74.8) / 93.0% (88.1-96.1)	Sens (95% CI) / Spec (95% CI)	47.4% (25.2-70.5) / 94.1% (89.4-96.9)
NPV (95% CI) / PPV (95% CI)	95.1% (90.5-97.6) / 43.5 % (23.8-65.1)	NPV (95% CI) / PPV (95% CI)	94.6% (90.0-97.3) / 45.0 % (23.8-67.9)

Open in a new tab

Conclusion

The absence of a demarcated area (where a regular neoplas-tic pit- pattern becomes disordered) within LSL is strongly predictive for the absence of submucosal invasion histologically. It can be determined using high definition white light and virtual chromendoscopy without the need for dye-based chromoendoscopy and has a high rate of interobserver agreement amongst experts.

Disclosure

Bourke, MJ - Research support from Olympus, Boston, Cook

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.132

OP142 Thermal Ablation of The Mucosal Defect Margin After Endoscopic Mucosal Resection: A Prospective, International, Multi-Center Trial

M Sidhu ^1,^2,^✉, S Gupta ¹, S Vosko ³, A van Hattem ¹, N Shahidi ¹, LF Hourigan ⁴, S Raftopoulos ⁵, AC Moss ⁶, S Zanati ⁷, SJ Williams ⁸, EYT Lee ⁸, NG Burgess ⁸, S Heitman ⁹, DJ Tate ¹⁰, MJ Bourke ¹

Introduction

Thermal ablation of the defect margin after endoscopic mucosal resection

(TAM-EMR) in the treatment of large (>/=20mm) laterally spreading lesions (LSL) has been shown to be efficacious in a clinical trial setting, with a fourfold reduction, in residual or recurrent adenoma (RRA) at 6 months first surveillance colonoscopy (SC1) [1]. The clinical effectiveness of this treatment is unknown.

Aims & Methods

We sought to evaluate the effectiveness of TAM-EMR and the rate of RRA in an international, multi-center prospective trial (NCT02957058). We conducted a study of consecutive LSL, across six tertiary centers, referred for EMR. The primary endpoint was the rate of RRA at SC1. TAM-EMR was performed using soft coagulation (ERBE - Tübingen, Germany:80W, Effect 4) via the snare-tip (STSC) to create a minimum 2-3mm rim of completely ablated tissue around the entire circumference of the resection defect. All endoscopists underwent an educational intervention prior, with two videos circulated at each centre and image review performed of treated LSL, critiqued to assess for uniform completeness of TAM-EMR.

Detailed demographic, procedural and outcome data were recorded. Recurrence was defined endoscopically as the presence of a neoplastic pit pattern within an EMR scar and confirmed histologically. Exclusion criteria included LSL involving the ileo-caecal valve/appendi-ceal orifice and circumferential LSL.

Results

Over 40 months to September 2019, 866 LSL (54.7% - right colon, median size - 35mm) were enrolled and underwent TAM-EMR. TAM-EMR with uniform completeness was achieved in 795 LSL. 71 LSL (median size - 40mm) had incomplete treatment with TAM-EMR (poor access - 26, pa-tientrelated - 15, deep mural injury (>/=3) - 10, risk of significant stenosis - 8, other - 12).

For patient and procedure details, see Table 1. 424/494 (85.4%) of eligible LSL underwent SC1 at median 5.8 months (IQR:4.8-6.9). 9/424 (2.1%) cases had evidence of RRA (endoscopic or histologic). RRA was commonly unifocal (7/9), diminutive (6/9) and occurred at the edge of the scar (6/9). All recurrences were treated endoscopically. For LSL with incomplete TAM-EMR, the rate of RRA was 28% (14/50), in those eligible for SC1.

Table 1.

[Lesion Details and Outcomes,DMI (III-V) - Sydney DMI Classfication [2], CSPEB - clinically signficant post endoscopic bleeding]

	TAM-EMR n = 866
Mean Age, yrs (SD) / Male, n (%)	67.5 (11.2) / 464 (53.6)
Median Lesion Size - mm (IQR)	35 (25-40)
*Right colon, n (%)	468 (54)
Invasive Cancer, n (%)	39 (4.9)
Major DMI, n (%)	4 (0.5)
IPB, n (%) / CSPEB, n (%)	191 (24) / 21 (2.6)
Underwent SC1, n / Eligible for SC1, n	424/494
Recurrence SC1, n (%)	9/424 (2.1)
Recurrence size (</= 5mm), n (%) / Site (edge of the scar), n (%)	6/9 (66.7) / 6/9 (6

Open in a new tab

Conclusion

In clinical practice routine thermal ablation of the defect margin after EMR is highly effective in reducing recurrence. This simple and inexpensive technique should be universally employed. Incomplete treatment, in difficult lesions, is associated with a higher rate of recurrence. Therefore, endoscopists should ensure thermal ablation is performed with uniform completeness in all LSL undergoing EMR.

Disclosure

Bourke, MJ - Research support from Olympus, Boston, Cook

References

1.Klein et al. Gastro, 2018 [Google Scholar]
2.Burgess et al. Gut, 2017 [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.133

OP143 Treatment Outcomes of Deep Mural Injury After Endoscopic Mucosal Resection in An International Cohort of 3462 Large Non-Pedunculated Colorectal Polyps

I Bar-Yishay ^1,^✉, N Shahidi ^1,^2,³, S Vosko ¹, S Gupta ^1,², A van Hattem ¹, S Schoeman ¹, M Sidhu ^1,², DJ Tate ⁴, LF Hourigan ⁵, R Singh ⁶, A Moss ⁷, S Raftopoulos ⁸, G Brown ⁹, S Zanati ¹⁰, SJ Heitman ¹¹, NG Burgess ¹², K Byth ¹², MJ Bourke ¹², SJ Heitman ¹¹, NG Burgess ¹, K Byth ¹², MJ Bourke ¹²

Introduction

Although perforation is the most feared adverse event associated with endoscopic mucosal resection (EMR), limited data exists concerning its management. Therefore, we sought to evaluate the short-and long-term outcomes of intra-procedural deep mural injury (DMI) in an international multi-center observational cohort of large (≥ 20mm) non-pedunculated colorectal polyps (LNPCPs).

Aims & Methods

Consecutive patients who underwent EMR for a LNPCP ≥ 20mm were evaluated. Significant DMI (S-DMI), was defined as Sydney DMI Classification type III (muscularis propria injury, target sign) or type IV/V (perforation without or with contamination, respectively). Short-term outcomes included successful defect closure, technical success (removal of all polypoid tissue) and referral to surgery. Long-term outcomes included the absence of recurrence at first surveillance colonoscopy (SC1).

Results

Between July-2008 to October-2019, 3462 LNPCPs underwent EMR. Median lesion size was 35mm (interquartile range (IQR) 25 to 45mm). Significant DMI was identified in 93 cases (2.7%), with successful defect closure in 91 (97.8%) using a median of 4 through-the-scope clips (TTSCs; IQR 3 to 6 TTSCs). 3 (3.2%) patients underwent S-DMI-related urgent surgery. EMR was successful in 86 (92.5%) patients, with 44 (47.3%) admitted to hospital (median duration 1 day; IQR 1 to 2 days). Comparing LNPCPs with and without S-DMI, no difference in technical success (86 (92.5%) vs. 3187 (94.6%); p = 0.4), or SC1 recurrence (13 (22.0%) vs. 359 (13.6%); p = 0.1) were identified.

Conclusion

Significant DMI is readily managed endoscopically and does not affect the technical success of EMR or long-term outcomes.

Disclosure

Michael Bourke: Research Support: Olympus Medical, Cook Medical, Boston Scientific. The remaining authors have no conflicts of interest to disclose.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.134

OP144 Cap-Assisted Endoscopic Mucosal Resection Versus Endoscopic Submucosal Dissection For Small Rectal Neuroendocrine Tumors: A Randomized Controlled Trial

Y Li ¹, S Liu ¹, Y Chen ¹, K Sun ^1,^✉

Introduction

Cap-assisted endoscopic mucosal resection (EMR-C) and endoscopic submucosal dissection (ESD) have both been reported to be effective treatment methods for small rectal neuroendocrine tumor (NET) in limited studies. However, which one is better and should be recommended to remove small rectal NET has not been determined.

Aims & Methods

We aimed to compare the efficacy and safety of EMR-C and ESD for the treatment of small rectal NET.

We prospectively identified patients diagnosed with primary localized rectal NET less than 10mm and randomly assigned to EMR-C or ESD group. The primary end-point is pathological complete resection rate, and the clinicopathological characteristics, en bloc resection rate, operating time, complications, hospitalization cost and recurrence were also analyzed.

Results

Sixty-two patients fulfilled the criteria and were enrolled in the study with 31 patients received EMR-C and 31 patients received ESD treatment. The baseline characteristics of the patients and the tumor size were well balanced between the two groups. En bloc resection was achieved in all patients of both groups, and there was no significant difference concerning about pathological complete resection rate (96.77% vs. 96.77%, p=0.382) and complication rate (3.22% vs. 3.22%, p=0.258) between EMR-C group and ESD group, respectively. However, shorter operating time (7.20±4.72min vs. 25.73±19.76min, p=0.005) and less hospitalization cost (529±473$ vs. 958±518 $, p=0.001) were observed in EMR-C group. No recurrent cases were observed in the two groups after an average follow-up time of 8.20 months.

Conclusion

EMR-C was preferable for the removal of small rectal NET (≤10 mm) with shorter operating time and less hospitalization cost compared with ESD (NCT03982264).

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.135

OP145 Thermal Ablation of The Defect Margin After Duodenal Endoscopic Mucosal Resection Significantly Reduces The Rate of Adenoma Recurrence

M Sidhu ^1,^2,^✉, J Fritzsche ³, A Klein ⁴, A van Hattem ⁵, S Vosko ⁶, N Shahidi ¹, DJ Tate ⁷, MJ Bourke ^1,²

Introduction

Duodenal adenomas are pre-malignant lesions which are conventionally treated with endoscopic mucosal resection (EMR). Post-EMR recurrence is a major limiting factor as it occurs in 10-20% of cases at first surveillance (SE1) and can be difficult to safely treat in the thin walled retroperitoneal duodenum where the risk is substantial.

Aims & Methods

We examined the utility of thermal ablation of the defect margin after duodenal EMR (dTAM-EMR) in the prevention adenoma recurrence at SE1. All duodenal laterally spreading lesions (LSL), >/=10mm in size, referred for EMR, at a single tertiary referral center, were eligible (NCT2006603). Data was prospectively collected. Standard inject and resect EMR technique was used.

The primary end-point was rate of histologic recurrence at SE1. A previous well characterized prospective cohort of duodenal LSL treated by conventional EMR (cEMR) was the comparator. After successful EMR, defined as complete excision of all macroscopically visible adenoma, thermal ablation was performed using soft coagulation via the snare-tip (ERBE - Tübingen, Germany: 80W, Effect 4) to create a minimum 2-3mm rim of completely ablated tissue around the entire circumference of the resection defect. Exclusion criteria included; involvement of papilla and 100% circumferential lesions. SE1 was performed at 6 months after the index procedure. All scars were routinely biopsied. The presence of a neoplastic pit pattern within EMR scar was regarded as recurrence. Any suspected recurrence was biopsied and then treated endoscopically.

Results

Over 46 months to August 2019, 50 LSL in 50 patients, underwent dTAM-EMR. 128 LSL underwent cEMR in the comparator arm. Median lesion size was 30mm (interquartile range (IQR): 20-40mm) in both groups with a majority of the lesions found in D2 (77.5%) (Table1).

Table 1.

[Lesion characterstics and procedure outcomes,CSPEB - clinically significant post endoscopic bleeding]

	cEMR n=128	dTAM-EMR n=50	Total	p
Age, years (SD)	68.0 (12.4)	65.4 (10.9)	66.1 (11.4)
Sex, n (%) M/F	28 (56.0)/ 22 (44.0)	71 (55.5)/ 57 (44.5)	99 (55.6)/ 79 (44.4)	0.95
Lesion Size, mm (IQR)	30 (20-40)	30 (20-40)	30 (20-40)	0.94
Paris, n (%) - IIa	35 (70.0)	86 (67.2)	121 (67.0)	0.42
Location, n (%) - D1/D2 junction	38 (76.0)	100 (78.1)	138 (77.5)	0.05
Complete endoscopic resection, n (%)	50 (100%)	122 (95.3%)	172 (96.6%)	0.12
IPB, n (%)	20 (39.2)	65 (49.6)	85 (46.7)	0.21
CSPEB, n (%)	9 (18.0)	22 (17.2)	31 (17.4)	0.90
Admission, n (%)	21 (42.0)	60 (46.9)	81 (45.5)	0.56

Open in a new tab

Of those eligible in the TAM-EMR group, 92.3% LSL underwent SE1, at median time of 6 months (IQR: 5-6). of those eligible in c-EMR group, 82.8% LSL underwent SE1, at a median time of 5 months (IQR: 3-7). EMR scars were identified in all cases and biopsied to assess for histologic recurrence.

The rate of recurrence was significantly lower in the dTAM-EMR group as compared to c-EMR; 2.8%(1/36) vs. 20.4% (22/108), p=0.01) (Table 1). There was no statistical difference in the rate of other intra or post-procedural adverse events between the two groups.

Conclusion

Thermal ablation of the defect margin after duodenal EMR results in a significantly reduced rate of adenoma recurrence at first surveillance endoscopy. This technique is safe in the duodenum and has the potential to significantly improve the effectiveness of duodenal EMR, avoiding the adverse outcomes associated with treatment of recurrence and the burden of endoscopic surveillance with conventional treatment.

Disclosure

Bourke, MJ - Research support from Olympus, Boston, Cook

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.136

OP146 Association Between Ex Vivo Human Ulcerative Colitis Explant Cytokine Secretion Profiles and Disease Behaviour

R Corcoran ^1,^2,^✉, P McDonagh ³, F O'Connell ⁴, J O'Sullivan ⁴, D Kevans ¹

Introduction

The clinical course of ulcerative colitis (UC) is variable with a proportion of patients experiencing sustained remission while others experience progressive disease with risk of hospitalisation and surgery. Identifying patients at high risk of disease progression would allow personalised therapeutic strategies for UC patients with the potential for improved outcome. There is therefore an unmet clinical need for biomarkers of UC disease behaviour and clinical response. We hypothesized that UC tissue microenvironment cytokine profiles would provide valuable information on UC disease behavior.

Aims & Methods

We aimed to evaluate the association between ex vivo human UC explant cytokine secretion profiles and disease behaviour. Twenty four patients with UC undergoing endoscopy for UC were prospectively recruited. Endoscopic biopsies were collected, washed with wash buffer (PBS, 10% FCS, Pen/Strep, Gentamicin, Fungizone), and then cultured for 24 hours in Media (M199, 10% FBS, 1% Pen-Strep, 1% Fungizone, 0.1% Gentamicin, 1ug/ml Insulin) with 0.1% DMSO to generate tissue conditioned media (TCM). TCM was analysed via 54 V-plex ELISA to assess cytokine secretions. Total protein content of biopsies was quantified to allow normalisation of secretions. Patients demographics, baseline characteristics and disease behaviour were characterised. Subsequent to index endo-scopic assessment, disease progression was defined as the requirement for corticosteroid therapy, UC-related hospitalisation, UC-related surgery or the introduction of a new immunomodulatory agent. As a secondary endpoint cytokine profiles were compared between anti-TNF failure patients compared with patients not exposed to or successfully treated with anti-TNF therapy. Comparison of secreted cytokine concentrations based on progression status and anti-TNF failure status was performed. P values < 0.05 were considered significant in analyses.

Results

Twenty four patients were recruited (age [mean,range] 51 years, 21-72; 50% female; disease duration [mean,range] 7.8 years, 1-17). Eight patients (33%) were receiving biologic therapy (n=8). Disease progression during follow up occurred in 10 (42%) patients with a mean time to disease progression of 9.2 months. IL2 secretion was significantly lower in individuals with disease progression compared to those without progression, median IL2 concentration 11.0 pg/ml per ug of protein [IQR 7.5 - 17.7] versus 29.4 [IQR 15.7 - 45.2] respectively, p=0.01. IL8 secretion was significantly lower in individuals with disease progression compared to those without progression, median IL8 concentration 2360.3 pg/ml per ug of protein [IQR 1898.5-2482.9] versus 3967.6 [IQR 2322.7-6613.2] respectively, p=0.05. Anti-TNF failure status was significantly associated with an increase in the following secreted cytokines: Exotaxin-3, FLT-1, GMSCF, IL10, IL12/23p40, IL12p70, IL17A/F, IL8, MDC, VEGF, PIGF, SAA and TARC (all p values < 0.05).

Conclusion

Reduction in secreted IL2 and IL8 were associated with disease progression in ex vivo human UC explants. IL2 is a pleotropic cytokine known to have immunoregulatory functions. IL8 has functions in immune cell chemotaxis and angiogenesis promotion. Increased secretion of IL-23 pathway- associated cytokines was observed in anti-TNF failure patients consistent with previous reports. While further evaluation and validation is required, the ex vivo human UC explant model has potential as a precision medicine tool in inflammatory bowel disease.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.137

OP147 Metabolic Profiling of Colonic Mucosal Wound Healing in Ulcerative Colitis May Lead To Novel Therapeutic Options

JT Bjerrum ^1,^✉, Y Wang ², J Zhang ², OH Nielsen ¹, JB Seidelin ¹

Introduction

Research on wound healing in the healthy colon and in ulcerative colitis (UC) is critical in order to identify new therapeutic targets that potentially may improve tissue regeneration and ultimately quality of life in affected patients; however, in vivo research is scarce.

Aims & Methods

Consequently, the present study was initiated to reveal mechanisms of colonic mucosal wound healing at the molecular level using a state-of-the-art metabolomic approach in humans with or without UC. Thirty individuals (21 patients with well-established UC and in endoscopi-cal and histological remission as well as nine healthy controls) participated. An innovative in vivo wound healing assay was applied. Here small experimental wounds were created in the colonic mucosa using a biopsy forceps at endoscopy, and at day two and seven post-wounding the subsequent wound healing process was surveilled by further serial endoscopies and sampling of biopsies across the experimental wounds. Kinetics were evaluated via an in-house developed mucosal wound healing score based on well-known kinetics from skin wounds,¹ and all biopsy specimens were analyzed by liquid chromatography coupled with mass spectrometry producing a metabolic profile focusing on phospholipids and eicosanoids. Multivariate statistical analyses were performed with the SIMCA-P+ tool (v12.0, Umetrics, Umeâ, Sweden).

Results

The wound scores were significantly higher in patients with UC as compared to controls both at day two (p=0.001) and seven (p< 0.0001), demonstrating a delay in the kinetics of wound healing in UC. This was reflected in the metabolic profiles of the wound biopsies, where sustained and significant decreased levels of several phospholipids were identified in UC at both day two and seven, but not in controls, cf. Table 1.

Table 1.

[Altered phospholipids during wound healing in ulcerative colitis and controls]

Lipid class	Total no. annotated lipids	Con day 0 Wounding	Con day 2	Con day 7	UC day 0 Wounding	UC day 2	UC day 7
PC	21	7265.3±507.7	5258.7±281.2αα	6324.8±639.6	7069.5±385.3	5013.1±299.6ααα	5575.9±255αα
LPC	5	97.4±11.3	66.7±6.3	92.7±7.8	90.2±8.7	58.0±4.8αα	75.7±5.3
PE	5	57.0±7.4	61.7±3.1	67.4±3.8	67.2±7	54.7±5.5	53.6±3.9β
LPE	1	9.0±6.3	0	5.4±3.6	5.2±3.9	0	0
PA	8	514.9±120.8	307.0±30	366.1±65.6	542.3±58.2	258.6±23.7ααα	296±32.7ααα
LPA	3	6.4±0.9	4.1±0.4α	5.0±0.8	6.3±0.5	3.4 ±0.4ααα	3.5 ±0.3ααα
PG	6	56.9±5.8	51.3±3.5	57.5±6.3	52.3±3.2	38.2±2.8αα	39.7±2.2αα/β
PS	8	355.6±20	357.9±25	401.3±39.6	328.1±12	285.2±15.9α	271.4 ±17.1α/β
PI	33	299.5±20.6	256.9±11.3	265.0±32	271.7±12.3	226.7±10.5αα	226.8±11.7α
SM	8	1429.6±126.1	1307.4±107.7	1457.8±184.6	1439.3±91	1226.1±51.2α	1282.6±61.2

Open in a new tab

Similarly, a range of AA derivatives, e.g., prostaglandins and thrombox-anes, were found (day 2 + 7) at significantly decreased levels in UC and controls, however, no significant differences were found between the two groups.

Conclusion

By use of an in vivo wound healing assay we are first to demonstrate a delay in the kinetics of wound healing in patients with UC, and concomitantly to characterize fluctuations of the lipid profile. Previous studies in animal models have verified efficacy of treating chemically induced colitis with phospholipids.^2-4 Similarly, the differences of phospholipids identified between UC and controls provide us with potential novel therapeutic avenues in humans, which need to be validated and investigated in future clinical trials.

Control (Con), Lysophosphatidic acid (LPA), lysophosphatidylcholine (LPC), phosphatidic acid (PA), phosphatidylcholine (PC), phosphatidyl-ethanolamine (PE), phosphatidylglycerol (PG), phosphatidylinositol (PI), phosphatidylserine (PS), sphingomyelin (SM), ulcerative colitis (UC) a: comparing day two or seven with day zero for UC or controls β: comparing controls with UC at day 7 aaa: p< 0.001; aa: 0.001< p< 0.01; a: 0.01< p< 0.05 β: 0.01< p< 0.05.

Disclosure

Nothing to disclose

References

1.Bielefeld K.A. et al. Cell Mol Life Sci 2013; 70: 2059. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Arun S.N. et al. J Lipid Res 2013; 54: 581. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Lin S et al. Am J Pathol 2018; 188: 353. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Wang X et al. Food Funct 2019; 10: 4608. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.138

OP148 Characterising The Small Intestine Microbiota Using Individuals with An Ileostomy Or Ileoanal Pouch in The Context of Ibd

RAA Ruigrok ^1,^2,^✉, V Collij ^1,², MAY Klaassen ^1,², P Sureda ^1,², BH Jansen ¹, M Voskuil ^1,², J Fu ^2,³, C Wijmenga ², A Zhernakova ², RK Weersma ¹, A Vich Vila ^1,²

Introduction

The gastrointestinal tract harbours distinct microbial communities essential for health and implicated in several disorders, such as inflammatory bowel disease (IBD). Most knowledge however is limited to the colonic microbiota, due to the use of faecal samples, leaving more proximal parts of the tract largely unknown. This is particularly true for the small intestine (SI) which, whilst depending on its microbiota, is responsible for 90% of total energy absorption from the diet and key to maintaining host-microbe immune homeostasis. Studying the SI microbiota has great potential to increase our knowledge of disease mechanisms, important for developing personalised diagnostics and therapy.

Aims & Methods

Using subjects with an ileostomy or ileoanal pouch, we aimed to explore the SI microbiota and its potential clinical implications, in an IBD context. Shotgun metagenomes derived from effluent samples of 57 IBD patients with an ileostomy or pouch were compared with faecal metagenomes of the general population (n=1178), IBD patients without bowel resections (n=309) and IBD patients with at least 1 small or large bowel resection (n=169). Phenotypic data was obtained via clinical records and questionnaires. Microbial diversity and community composition were compared between groups by calculating Shannon diversity, Bray-Curtis dissimilarities and taxa and microbial gene abundances.

Next, multivariable boosted general linear model analyses were performed, correcting for calculated confounders such as age, sex and IBD-specific factors. P-values were adjusted using false discovery rate (FDR) to correct for multiple testing, where relevant; FDR< 0.05 was considered significant.

Results

Microbial diversity within the SI was significantly lower than in faecal samples from IBD patients and general population (P-value< 0.05). Moreover, analysing the microbial compositions showed that the relative abundances of 89 & 49, out of 134, species differed significantly between the small intestine and the general population and IBD resected bowel, respectively (FDR< 0.05). The relative abundance of Veillonella atypia, for example, was significantly higher in the SI compared to all three groups with the degree reducing in order from general population, to IBD non-resected bowel and to IBD resected bowel (FDR< 0.01). E.coli and S.salivarius were also significantly increased in the SI compared to the general population (FDR< 0.05 & < 0.01,respectively) however, did not differ significantly from the IBD resected bowel. F.prausnitzii, a usual indicator of gut health was decreased in SI compared to the general population (FDR< 0.01). 243 out of 341 pathways were also differentially abundant between the SI and general population (FDR< 0.05). Around 20% of pathways enriched in the SI are involved in oxidative phosphorylation and simple carbohydrate metabolism.

Conclusion

Clear differences between microbial diversity, composition and pathway abundance in SI and faecal samples were seen, highlighting distinct functional roles of the SI and colonic microbiota that may be clinically relevant. Interestingly,IBD resected bowel metagenomes showed most resemblance to the SI and many of the species found to predominate the SI have previously been associated to IBD. This poses a hypothesis implicating loss of diversity in the colon with concurrent enrichment of non-native SI bacteria in the pathophysiology of IBD. These results highlight the importance of studying the SI microbiota to add other perspectives to disease pathogenesis, enhancing what is currently known and potentially improving disease management.

Disclosure

RKW unrestricted grants: Takeda, Johnson and Johnson, Tramedico, Ferring. Consultant: Takeda The other authors have nothing to disclose.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.139

OP149 A Key Role For Muc13 in Intestinal Barrier Dysfunction During Dextran Sodium Sulphate-Induced Colitis in Mice

T Breugelmans ^1,^2,^✉, B Oosterlinck ^1,², J De Man ^1,², B De Winter ^1,², A Smet ^1,²

Introduction

MUC13 is the main transmembrane mucin expressed in the colon and overexpression of this mucin has been reported in the inflamed mucosae of patients with inflammatory bowel diseases and colorectal cancer patients^1-3. We recently showed that aberrant MUC13 expression coincides with increased intestinal permeability upon experimental colitis in mice². in this study, we further evaluated the role of MUC13 in barrier function upon intestinal inflammation using a MUC13 knockout mouse model.

Aims & Methods

Seven- to nine-week old male wildtype (WT) and MUC13^-/- C57BL6/J mice were treated with 1 or 2 cycles of 2% dextran sodium sulphate (DSS) in their drinking water for 7 days followed by a recovery phase of 7 days with normal drinking water to induce acute (cycle 1) and chronic colitis (cycle 2). Control animals received only drinking water (n = 6-10 animals/group). Colitis severity was evaluated by the assessment of disease activity (based on weight loss, changes in stool consistency and rectal bleeding) and a weekly colonoscopy. At the end of each DSS treatment, animals were euthanized and colonic tissue collected. Intestinal inflammation was evaluated by the colon weight/length ratio, microscopic evaluation of the colon, myeloperoxidase (MPO) activity and cytokine mRNA expression (Tnfa, Illb and Il22) in the colon. Intestinal barrier integrity was investigated by the in vivo intestinal permeability to 4 kDa FITC-dextran and the colonic mRNA expression of mucins (Mucl,2,4& 13) and claudins (Cldn1, 2,3,4 & 7).

Results

No macroscopic abnormalities were observed in the colon of control MUC13^-/-mice. Administration of DSS induced colitis in both WT and MUC13^-/- mice as shown by macroscopic and microscopic evaluation of intestinal inflammation. Knockout of MUC13 significantly increased disease activity (mean difference MUC13+ DSS vs WT DSS: 1.12 ± 0.37; p< 0.01) and MPO activity (mean difference MUC13+ DSS vs WT DSS: 2.01 ± 0.76 U MPO/mg; p< 0.01) after 1 cycle of DSS compared to WT mice, but this phenomenon was not seen after 2 cycles. Furthermore, expression analysis of intestinal cytokines did not reveal significant differences between WT and MUC13^-/- mice. During the course of colitis, DSS-treated WT mice showed an increased intestinal paracellular flux of 4 kDa FITC-dextran in the serum compared to WT control mice (Table 1). Interestingly, FITC-dextran permeability was already increased in control MUC13^-/- mice but did not alter significantly upon DSS-treatment (Table 1). This was associated with increased expression of Cldn2 (FC 2.77, p < 0.001) in control MUC13+ mice compared to WT mice. Finally, MUC13 knockout did not affect the expression of other intestinal mucins, suggesting that these observations were solely the result of an altered MUC13 expression.

Table 1.

[4 kDa FITC-dextran concentration in the serum of control and DSS-treated mice]

	Control	1 cycle DSS 2%	2 cycles DSS 2%
Wildtype	860.4 ± 214.6 μg/mL	6128.29 ± 2316.3 ng/mL	2546.3 ± 953.0 ng/mL
MUC13 knockout	4929.0 ± 1914.5 ng/mL	11461.6 ± 5125.8 ng/mL	1853.8 ± 687.3 ng/mL

Open in a new tab

Conclusion

The aggravated acute DSS-induced colitis seen in MUC13^-/-mice seem to be partially attributed to the increased intestinal permeability already present in healthy MUC13^-/- mice, further suggesting a key role for MUC13 in regulating intestinal barrier function.

Disclosure

Nothing to disclose

References

1.Sheng Y.H., Hasnain S.Z., Florin T.H.J., McGuckin M.A. Mucins in inflammatory bowel diseases and colorectal cancer. J Gastroenterol Hepatol 2012; 27: 28–38. [DOI] [PubMed] [Google Scholar]
2.Breugelmans T., Van Spaendonk H., De Man J.G., De Schepper H.U., Jauregui-Amezaga A., Macken E., Lindén S.K., Pintelon I., Timmermans J-P, De Winter B.Y., Smet A. In depth study of transmembrane mucins in association with intestinal barrier dysfunction during the course of T cell transfer and DSS-induced colitis. J Crohns Colitis 2020. [DOI] [PubMed] [Google Scholar]
3.van Put-ten J.P.M., Strijbis K. Transmembrane Mucins: Signaling Receptors at the Intersection of Inflammation and Cancer. J Innate Immun 2017; 9: 281–99. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.140

OP150 Genetic Architecture of Familial Inflammatory Bowel Disease

H-S Lee ^1,^2,^✉, L Hannes ¹, V Ballet ³, M Ferrante ³, S Vermeire ³, I Cleynen ¹

Introduction

Family history of inflammatory bowel disease (IBD) is the strongest risk factor for IBD. There currently, however, is incomplete understanding of the contribution of genetic risk to familial aggregation of IBD.

Aims & Methods

We included 54 multiple-affected families (≥ 3 first-degree relatives affected) of European ancestry, including 189 affected IBD patients (156 Crohn's disease; 33 ulcerative colitis), and 133 unaffected relatives. All individuals were genotyped using Immunochip. To identify genetic variants associated with IBD, we performed both transmission disequilibrium test (TDT) and family-based association (DFAM) analyses using PLINK v1.90b5.2. We also analysed to what extent the risk for IBD in these families is influenced by the combined effect of common genetic variants. For this, we calculated weighted polygenic risk scores (PRS) and their explained variance (Nagelkerke pseudo-R²) across different p-value thresholds (pT) using PRSice-2.0 and based on summary stats from de Lange et al [1]. Sporadic cases (n=1768) and non-IBD controls (n=868) were used for comparison.

Results

Based on TDT analysis, we found five loci with suggestive evidence (p<5.0x10-04) for association in familial cases. Three of these were also significant in the DFAM analysis (p<5.0x10-03): IL12RB1 (rs426132, pTDT=1.7x10-04 and pDFAM=3.8x10-03), SLC9A8 (rs536092, p=2.1x10-04 and pDFAM=3.8x10-03), and SLC2A13_LRRK2 (rs2253736, pTDT=2.4x10-04 and pDFAM=2.0x10-03). DFAM analysis further identified two loci reaching near genome-wide significance: LCE3D_LCE3E (rs16834214, p=5.0 x10-08) and CCR3 (rs78531959, p=9.9 x10-07). Using pT=0.05 for PRS calculation, we found that affected relatives had a higher PRS than unaffected relatives (p=1.0x10-02), sporadic cases (p=4.6x10-02), and non-IBD controls (p<2.2x10-16). However, the proportion of the explained variance was smaller in familial IBD (r2=2.5%) than in sporadic IBD (r2=15.9%). When checking which PRS (i.e. at which pT) had the highest variance explained, this best pT was different for familial and sporadic IBD. in familial IBD, the best-fit PRS was at pT=6.9x10-03 (r2=5.7%, p=3.1x10-04), whereas in sporadic IBD, the best-fit PRS was at pT=0.08 (r2=16.7%, p=8.5x10-63). in sporadic IBD, variants at all p-value levels (until ∼pT=0.1) contributed to disease risk; while for familial IBD, the difference between those affected or not was mostly with variants of higher significance (i.e. pT=0.01) (Table). On the other hand, the difference between familial and sporadic cases, and familial and sporadic controls, mostly was in lower significant SNPs (Table).

[Explained variance at P-value threshold as indicated.]

	familial IBD vs unaffected relatives	sporadic IBD vs control	familial IBD vs sporadic IBD	unaffected relatives vs control
r2 at pT=5x10-8	3.38%	9.61%	0.03%	1.78%
r2 at pT=1x10-4	5.02%	12.43%	0.12%	2.83%
r2 at pT=0.01	5.05%	13.27%	0.04%	2.41%
r2 at pT=0.05	2.50%	15.90%	0.43%	8.00%
r2 at pT=0.1	2.46%	16.55%	1.38%	13.16%
r2 at pT=1	1.98%	14.83%	0.79%	9.39%

Open in a new tab

Conclusion

Our finding of the novel SLC9A8 locus not previously associated with IBD supports the importance of studying also familial IBD genetics. SLC9A8 has been associated with β-diversity of intestinal microbiota, hereby further implicating shaping the human intestinal microbiota in familial IBD [2]. Our PRS analyses showed that higher IBD polygenic risk increases the risk for familial IBD as it does for sporadic IBD; albeit that this elevated risk mostly comes from SNPs with lower significance. The more significant SNPs on the other hand cause the difference between affected and unaffected within families.

Disclosure

Nothing to disclose

References

1.de Lange K.M. et al. Nat Genet 2017. [Google Scholar]
2.Ruhlemann M.C. et al. Gut Microbes 2018. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.141

OP151 The Impact of Covid-19 On Gi Practice in Europe and Beyond - A Survey

H Heinrich ^1,^2,^✉, C Carretero ³, L Ricciardiello ⁴, G Ianiro ⁵, DL Dumitrascu ⁶, P Acedo ⁷, C Vladut ⁸, A Dignass ⁹

Introduction

The COVID-19 pandemic continues to have relevant impact on medical practice worldwide. Particularly, gastroenterological procedures have been reduced also because of specific risks (e.g. risk of aero-solization or faecal-oral transmission).

Aims & Methods

Our 25-item survey study aimed to assess the effect of the COVID-19 pandemic on various aspects of GI practice in a large cohort of healthcare professionals involved in the management of digestive diseases.

An anonymous, internet-based survey was sent via email to recipients within the UEG community and was promoted by national and specialist GI societies as well as via social media and on the UEG Website.

Results

3325 responses were received from 114 countries. 2063 valid responses were available for analysis. 620 female (30.1%), 1441 male (69.8%), and 2 (0.1%) non-binary GIs with an age range 18-75 responded to the survey. N= 346 were based in private practice (PP), n=313 in both PP/ General Hospital (GH)/University Hospital (UH), n=1362 were based in GH and UH, and n=42 were based in other health institutions. Most units and PP followed local guidance on endoscopy during the COVID-19 pandemic (59%), with 23% and 10% following ASGE and ESGE guidelines, respectively.

A majority of hospitals and PP used only surgical masks for upper endos-copy (43.8%) and lower endoscopy (48.5%), with only 33% and 30%, respectively, using FFP2 masks during highly aerosol generating procedures. 37% of hospital GIs were redeployed to care for COVID-19 patients. 48% of redeployed doctors were GI trainees. 80% of GIs felt sufficiently instructed on PPE use, but 65% of GIs felt that there was not enough PPE available in their country of residence. 72% felt that not enough COVID-19 testing was performed. 9.7 % of GIs reported having tested positive for COVID-19. Upper GI bleeding (96.5%), acute lower GI bleeding, oesophageal obstruction (85%), and acute cholangitis (77.6%) were considered urgent in hospitals and PP. Nutritional support (48.3%), infected walled of necrosis (48.3%), and vacuum therapy for surgical leaks (51%) were inconsistently seen as urgent enough to warrant endoscopy. Symptomatic patients without warning signs (64.7%), surveillance of low-risk lesions (82.1%), CRC screening (70%), bariatric endoscopy (65.8%), and faecal transplants (63.5%) were considered non-urgent.

Furthermore, COVID-19 had a significant impact on research, as 47% of centres reported total suspension of all human clinical trials, and 36% were only allowed to conduct research under certain circumstances. Family (86%), friends (57%), and the local GI team (68%) were reported as major sources of support during the pandemic. Most GIs felt calm, confident, and hopeful during the crisis and reported low to moderate levels of physical (median 3.3 ± 2.8) and psychological exhaustion (5.0 ± 3.0) on a scale from 1 to 10. The overall impact of COVID-19 on practice in PP/GH/UH on a scale from 1-10 was high (8.0 ± 2.1).

Conclusion

The impact of COVID-19 on all aspects of practice is high, with PPE availability, application, and COVID-19 testing being key issues. This is highlighted by the high number of GIs being infected. The impact of the pandemic on patient care, research, and training is substantial; however, the GI community seems to face this challenge with a calm and hopeful attitude.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.142

OP152 Intestinal Expression of The Sars-Cov-2 Receptor Ace2 and Tmprss2 in Biopsies and Derived Organoids From Patients with Inflammatory Bowel Disease

S Verstockt ^1,^✉, B Verstockt ^1,², K Arnauts ¹, J Sabino ^1,², M Ferrante ^1,², S Vermeire ^1,²

Introduction

While the respiratory tract is the primary portal entry of SARS-CoV-2, the host receptor for this coronavirus, Angiotensin-Convert-ing Enzyme 2 (ACE2), is also expressed on differentiated enterocytes. Furthermore, digestive symptoms have been reported in COVID-19 patients. The cellular entry of coronaviruses is further facilitated by the Transmembrane serine protease 2 (TMPRSS2). As intestinal inflammation may affect ACE2 and TMPRSS2 expression and hence alter susceptibility to COVID-19, we studied ACE2 and TMPRSS2 expression in intestinal small and large bowel biopsies and derived organoids from patients with IBD.

Aims & Methods

Ileal and colonic mucosal biopsies were collected from 166 Crohn's disease (CD) patients and 155 ulcerative colitis (UC) patients, with 300 biopsies from inflamed segments and 72 from uninflamed segments. Matched biopsies from non-IBD controls were included as comparison (n=59). Colonic biopsies (inflamed/uninflamed) from eight UC patients and eight non-IBD controls were used for crypt isolation and cultured as organoids for at least four weeks. Organoids were subjected to a predefined inflammatory mix (100 ng/ml TNF-α, 20 ng/ml IL-1β, 1 μg/ml Flagellin) for 24 hours. RNA was extracted and single-end RNA sequencing (Illumina) performed. Two sample t-tests and Wilcoxon tests were applied as appropriate (R 3.6.1), and statistical significance was defined as a p value < 0.05.

Results

Within non-IBD control biopsies, ACE2 levels were strongly increased in ileum compared to colon (fold change (FC)=32.0, p=6.3E-13), whereas TMPRSS2 was lower in ileum compared to colon (FC=-2.9, p=6.3E-13). in inflamed CD ileum, we found significantly decreased ACE2 as compared to uninflamed IBD ileum (FC=-2.5, p=4.1E-11) and control il-eum (FC=-2.8, p=4.6E-07). At colonic level, ACE2 expression was higher in inflamed CD and UC colon compared to control colon (FC=1.4, p=8.3E-03; FC=1.4, p=1.9E-02), while expression levels in uninflamed IBD colon did not differ from those in control colon (p=2.1E-01). TMPRSS2 expression behaved opposite and was upregulated in both inflamed CD ileum and uninflamed IBD ileum versus control ileum (FC=1.4, p=6.3E-13; FC=1.2, p=1.8E-03). No dysregulations were observed for colonic TMPRSS2 across the different study groups. in UC-derived organoids, prior to inflammatory stimulation, ACE2 and TMPRSS2 expression levels were independent of disease (UC/non-IBD control) and inflammatory origin. in contrast, exposure to the inflammatory mix induced transcriptional activation of ACE2 and TMPRSS2 in organoids of UC origin (both inflamed and non-inflamed), resulting in significantly higher levels as compared to organoids of non-IBD origin after stimulation (ACE: FC≥3.4, p≤1.7E-05; TMPRSS2: FC≥1.2, p≤3.6E-02).

Conclusion

Expression of ACE2 and TMPRSS2, the key players in cellular entry of SARS-CoV2, is dysregulated in the intestinal mucosa of IBD patients. Our results in UC-derived organoids and colon biopsies suggest that IBD patients with active disease might be more vulnerable to COV-ID-19 because of the upregulation of ACE2 and TMPRSS2 by inflammatory stimuli, as compared to healthy controls. in contrast, a decrease in ACE2 in inflamed ileum could reflect a loss of absorptive enterocytes. Additional single-cell transcriptomic and functional studies are needed to further unravel SARS-CoV2-related mechanisms in the intestine of IBD patients.

Disclosure

B. Verstockt received financial support for research from Pfizer; lecture fees from Abbvie, Ferring Pharmaceuticals, Janssen, R-biopharm and Takeda; consultancy fees from Janssen and Sandoz;. J. Sabino reports speaker's fees from Abbvie and Nestle Health Sciences; M. Ferrante received financial support for research from Janssen, Pfizer, Takeda; consultancy fees from Abbvie, Boehringer-Ingelheim, Celltrion, Ferring, Janssen, Lilly, Mitsubishi Tanabe, MSD, Pfizer, Takeda; and speaker's fees from Abbvie, Amgen, Biogen, Boehringer-Ingelheim, Chiesi, Falk, Ferring, Janssen, Lamepro, Mitsubishi Tanabe, MSD, Pfizer, Takeda, Tramedico, Tillotts, Zeria; S. Vermeire received financial support for research from MSD, Abbvie, Takeda, Janssen, Pfizer; honoraria or consultation fees from AbbVie, MSD, Takeda, Ferring, Genentech/Roche, Shire, Pfizer Inc, Galapagos, Mundipharma, Hospira, Celgene, Second Genome, Progenity, GSK, Lilly, Arena, Gilead and Janssen; participated in company sponsored speaker's bureau from AbbVie, MSD, Takeda, Ferring, Hospira, Pfizer, Janssen, and Tillots.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.143

OP153 Impact of Covid-19 On Patients with Inflammatory Bowel Disease: Data From An International Registry

R Ungaro ^1,^✉, E Brenner ², RB Gearry ³, GG Kaplan ⁴, M Kissous-Hunt ¹, JD Lewis ⁵, SC Ng ⁶, J-F Rahier ⁷, W Reinisch ⁸, FM Ruemmele ⁹, F Steiwurz ¹⁰, FE Underwood ¹¹, M Agrawal ¹, X Zhang ¹², J-F Colombel ¹, MD Kappelman ²

Introduction

The impact of Coronavirus disease 2019 (COVID-19) on patients with inflammatory bowel disease (IBD) is not well characterized.

Aims & Methods

We aimed to characterize the clinical course of COVID-19 among IBD patients and evaluate the association between demographics, clinical characteristics, and immunosuppressant treatments on COVID-19 outcomes. Surveillance Epidemiology of Coronavirus Under Research Exclusion for Inflammatory Bowel Disease (SECURE-IBD) is a large, international registry created to monitor outcomes of IBD patients with confirmed COVID-19. We calculated age-standardized mortality ratios (SMRs) and utilized multivariable logistic regression to identify factors associated with severe COVID-19, defined as intensive care unit admission, ventilator use, and/or death.

Results

959 cases from 40 countries were reported (Median age 43 years, 52% men). Eighty-six patients (9%) had severe COVID-19, 320 (33%) were hospitalized, and 37 patients died (3.9% case fatality rate). SMRs for IBD patients were 2.0 (95% confidence interval [CI] 1.4-2.7), 1.7 (95% CI 1.1-2.2), and 1.9 (95% CI 1.3-2.5) relative to data from China, Italy, and the US, respectively. The proportion of patients with severe COVID-19 and death stratified by medication class are presented in Table 1.

Table 1.

[Proportion of patients with death and severe COVID-19 stratified by medication class]

Medication Class (n)	Death, n (%)	Severe COVID-19, n (%)
Sulfasalazine/5-aminosalicylate (261)	21 (8)	44 (17)
Systemic corticosteroids (73)	8 (11)	18 (25)
Thiopurine monotherapy (95)	2 (2)	9 (9)
TNF antagonist monotherapy (283)	3 (1)	8 (3)
TNF antagonist + thiopurine/methotrexate (93)	2 (2)	11 (12)
Anti-integrin biologic (88)	2 (2)	7 (8)
Anti-IL12/23 biologic (96)	0 (0)	2 (2)

Open in a new tab

Risk factors for severe COVID-19 among IBD patients included increasing age (adjusted odds ratio [aOR] 1.04, 95% CI 1.02-1.06), 1 comorbidity in addition to IBD (aOR 2.60, 95% CI 1.34-5.01), ≥2 comorbidities (aOR 4.8, 95% CI 2.4-9.6), systemic corticosteroids (aOR 5.1, 95% CI 2.5-10.9), and sulfasalazine or 5-aminosalicylate use (aOR 2.0, 95% CI 1.1-3.6). TNF antagonist treatment was not associated with severe COVID-19 (aOR 0.9, 95% CI 0.5-1.8).

Conclusion

Increasing age, comorbidities, and corticosteroids are associated with severe COVID-19 among IBD patients. Notably, TNF antagonists do not appear to be associated with severe COVID-19.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.144

OP154 Impact of The Covid-19 Pandemic On Endoscopy Training: A Multi-National Survey

S John ^1,^✉, H Neumann ²

Introduction

COVID -19 has caused significant disruption of our personal and professional lives across the world. Within our specialty practice, it has caused major changes in how we deliver in- patient and outpatient services. Endoscopy practice has been particularly affected with societal guidelines recommending variable restrictions on type and volume of procedures. This has a profound impact on training in endoscopy. We conducted a multi-national survey to assess the global impact of the pandemic on training in endoscopy.

Aims & Methods

The aim of our multi-national survey was to assess the impact of the pandemic on endoscopy training and measures required post pandemic to address any issues. The survey was conducted using an online survey platform in the Asia Pacific and Europe. Trainees in Endoscopy were invited to participate through major training centres or training leads in national societies. A survey with 16 questions was designed with a focus on age, gender, year of training, experience in endoscopy prior to the pandemic and details of current training in endoscopy. The respondents were also asked about the impact of the pandemic on their training, duration of impact and their opinions on how to address this post pandemic.

Results

127 trainees responded to the survey from 13 countries in the Asia Pacific and Europe. 76.3% were aged 25-34 years and 22% were between 35-44 years. 27.5% who responded were women. with regards to current level of training, 26.7% were in their first year, 19.7% in the 2^nd year, 15.8% in the 3rd year and the rest in advanced fellowships. 56% suggested that their focus of training was in upper and lower GI endoscopy and 10.24% on ERCP and EUS training .22% stated that their training was currently fo-cussed on upper GI endoscopy. A large majority (85.8%) responded that training had been impacted by the pandemic.

Furthermore, 80.7% of the respondents had training impacted for a significant duration of 1-3 months and only 4% longer than 3 months. Nearly 54% trainees are performing < 20 procedures per month with 41% performing < 10 procedures. Restricted case numbers (64.5%) and hospital policy (30%) were cited as the main reasons. 28% stated that they did not receive adequate training in PPE use in endoscopy. A third of respondents suggested that they are very concerned by the impact on training with another 40% moderately concerned. More than half the trainees felt that they would not achieve the required competencies in training this year and two thirds propose that training centres and national societies should develop additional training measures post pandemic. of interest, nearly 40% suggested that this could be done with an extension of their training period vs 41 % suggesting other measures such as workshops.

Conclusion

The global impact of the pandemic on health care delivery and on the health and well being of essential workers is profound. Gastroenterology is no exception to this, and the practice of endoscopy has been severely affected. It is important to acknowledge the overall impact on our trainees. Our survey was limited to a single aspect i.e.. Endoscopy training. This large multi-national survey emphasises the significance of the impact on training and the clear need for planning resources on improving this post pandemic.

There is an urgent need for all trainees in endoscopy to have adequate PPE training. Additional resources for upskilling, extension of training where required, support with educational tools online as well as increased workshops require planning and funding by training centres and national societies.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.145

OP155 Effects of Sars-Cov2 Emergency Measures On Endoscopic Clinical Practice and High-Risk Lesions Detection: A Multicentre Cross-Sectional Study

M Furnari ^1,^✉, LH Eusebi ², EV Savarino ³, C Petruzzellis ⁴, G Esposito ⁵, MF Maida ⁶, L Ricciardiello ⁷, S Pecere ⁸, A Buda ⁹, M De Bona ⁹, C Spada ⁴, E Di Giulio ¹⁰, G Costamagna ¹¹, I Boskoski ¹², EG Giannini ¹; Young-ENDO-ITA

Introduction

COVID-19 pandemic enforced the interruption of endo-scopic screening programs and routine examinations raising the issue of potential delays in the diagnosis of pre-neoplastic and neoplastic lesions.

Aims & Methods

We aimed to assess the decrease of gastrointestinal and pancreato-biliary lesions detected related to the reduction of the endo-scopic procedures during COVID-19 pandemic. Secondary aims were to evaluate the impairment of training and research activities, and adherence to safety protocols.

A multicentre, retrospective, cross-sectional study to compare the pandemic timeframe from March 9 to April 6, to the equivalent timeframe of 2019, was conducted in eight tertiary centres in terms of: type and number of elective and emergency endoscopies; detection rate of gastrointestinal and pancreato-biliary lesions; performance of colorectal cancer (CRC) screening programs. feasibility of research and training activities; safety measures adopted.

Results

Endoscopic procedures decreased by 73% (7,199 vs 2,086): screening colonoscopies by 91.7%, elective procedures by 70.2% and urgent procedures by 48.2%. Hepato-pancreatic-biliary endoscopy had the lowest reduction (48.2%) compared to upper and lower gastrointestinal endoscopy (71.3%, 78.1%;P< 0.001). The overall number of detected high-grade dysplasia/cancers decreased by 68.3% (486 vs 154;P< 0.001): 25.8% pancreato-biliary cancers; 75% and 73% upper and lower gastrointestinal lesions; 93.3% in screening colonoscopy. Research and training ceased by 62.5%. All centres adhered to safety recommendations.

Conclusion

COVID19 outbreak caused a drastic reduction of emergency and elective endoscopies with an alarming reduction of detected cancer and high-risk lesions. The interruption of CRC screenings reduced advanced lesions detection and increased screening intervals with potential risks of CRC progression. Research and training activities were similarly jeopardized. Our findings highlight the need of strategies aimed at reorganizing endoscopic activity and teaching programs.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.146

OP156 Esoflip Pyloric Dilation in Gastroparesis Improves Gastric Emptying, Pyloric Distensibility and Symptoms

F Murray ^1,^✉, V Schindler ¹, J Hente ¹, L Fischbach ¹, L Schnurre ¹, RA Deibel ¹, F Hildenbrand ¹, D Pohl ¹

Introduction

Gastroparesis is a debilitating condition with limited diagnostic and therapeutic options. Pyloric dysfunction is increasingly being recognized as a driving factor in gastroparesis. Recently, data have emerged demonstrating a link between decreased pyloric distensibility and gastric emptying, as measured with the functional lumen imaging probe (FLIP). The pylorus is amenable to endoscopic therapy, although dilation with a usual 20mm through the scope (TTS) balloon has not shown to be very effective. Alternative treatment methods are sparse and generally limited by lack of objective and/or subjective improvement.

Aims & Methods

This is the first study presenting clinical outcome and combination of the diagnostic use of EndoFLIP with therapeutic use of EsoFLIP in patients with gastroparesis. Following the routine protocol in our clinic, patients evaluated for gastroparesis and gastric emptying studies (GES) of t1/2 ≥180 minutes (min) during ¹³C-octanoic acid breath test and/ or gastric solid remnants during gastroscopy after a minimum 12 hour fasting period, were scheduled for EsoFLIP controlled pyloric dilation. Dilation was conducted tailored to each patient using a self-developed algorithm. Pre- and post-procedure GES, questionnaires (PAGI-SYM [including GCSI] and -QOL) and FLIP metrics (including pre- and post-procedural pyloric distensibility) were documented.

Results

Forty-six patients were analyzed (median age: 39ys; range: 18-88; women 33 [72%]; diabetic 10 [22%], idiopathic 33 [72%], postoperative 3 [6%]). Dilation was conducted to a maximum pyloric diameter of 26.5mm (median: 25.2mm). Post-procedural GES decreased significantly (p=0.001) from a median of 211min to 179min. in accordance, pyloric distensibility increased significantly (p< 0.001) from a median of 9 to 13mm²/mmHg after dilation. Quality of life markers PAGI-SYM, PAGI-QOL and GCSI values improved significantly. After a median follow-up of 3.9 months, 57% of all treated patients with returned questionnaires reported global symptom improvement. Twenty-three (50%) of patients reported post-procedural epigastric pain for a median of one day with a median intensity of 3 (scale from 0 to 10). No other side effects, especially no perforations, hospitalizations or hemodynamically relevant bleeding were documented.

Conclusion

Pyloric EsoFLIP-dilation for gastroparesis appears feasible, well-tolerated and comparatively effective using both objective and subjective outcome markers. with its potential of individualized treatment regimes, good handling and larger dilation diameter to conventional TTS balloons, EsoFLIP constitutes an exciting alternative to existing interventional modalities. Long-term follow-up to assess efficacy and safety as well as controlled studies for comparison to other pylorus-directed treatment strategies are warranted.

Disclosure

Daniel Pohl is a consultant for Medtronic.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.147

OP157 Proton Pump Inhibitors Increase Cortisol Reactivity in Functional Dyspepsia Patients, Independent of Duodenal Barrier and Immune Function

L Wauters ^1,^2,^✉, L Cools ², M Ceulemans ², A Accarie ², J Toth ², J Tack ^1,², T Vanuytsel ^1,²

Introduction

Functional dyspepsia (FD) is a disorder of gut-brain interaction, with gut abnormalities driving central changes and vice versa. We previously reported restored duodenal barrier and immune function in FD patients on proton pump inhibitors (PPI) (1). However, central changes including cortisol reactivity and their response to treatment targeting the GI-tract have not been studied.

Aims & Methods

We aimed to study cortisol reactivity and the link between GI and central changes before and after PPI-therapy in FD patients (Rome IV criteria) compared to healthy controls (HC). The cortisol awakening response (CAR) was determined on the day of the endoscopy with collection of duodenal biopsies. Salivary cortisol was measured with ELISA for 5 samples taken every 15 min upon awakening and area under the curve with respect to baseline (‘increase’ or AUCi) and 0 (‘ground’ or AUCg) were calculated.

Paracellular passage of a fluorescein-labeled dextran (4 kDa) was recorded in Ussing chambers and eosinophils were counted on H&E-stained sections per high-power field (HPF; 0.24 mm2). All subjects filled out the Patient Assessment of GI Disorders Symptom Severity Index (PAGI-SYM), the Diet and Food Habits dimension from quality of life (PAGI-QOL) and Perceived Stress Scale (PSS) questionnaires.

Procedures were repeated after pantoprazole 40 mg OD for 4 weeks (on-PPI). Multilevel (mixed) models were constructed for each dependent variable with treatment (off- or on-PPI) as within- and group (FD or HC) as between-subject independent variables. For CAR, a factor time after awakening was added. Next, dextran-passage, eosinophils or PAGI were separately added in the model with cortisol to test potential mediation.

Results

In total, 29 FD patients (24 female) and 30 HC (21 female) with similar age (31 ± 2 yrs) were included. Duodenal hyperpermeability and eosinophilia in FD patients were significantly reduced on-PPI with higher PSS in FD vs. HV and no effect of treatment (table). PAGI-SYM significantly decreased and PAGI-QOL increased in FD on- vs. off PPI (both F=21.4, p< 0.0001). The significant increase of cortisol over time (F=19.97, p< .0001) significantly differed between groups (F=2.89, p=0.02). Higher baseline cortisol levels in FD vs. HC upon awakening (t=0 min) were also reflected by a higher AUCg in FD vs. HC off-PPI (table). in contrast, a significant decrease in cortisol at 0 min in FD on- vs. off-PPI was reflected by the increase in AUCi, which was different from HC (β=-0.52 ± 0.17, p=0.003). No mediating effect was found for duodenal dextran-passage or eosinophils on the change in salivary cortisol (0 min) in FD on- vs. off-PPI (all p>0.05). in contrast, the treatment effect of PPI on AUCi of cortisol was statistically mediated via changes in PAGI-QOL (F= 5.04, p=0.04) but not PAGI-SYM (p>0.05).

[Table: between- and within-group comparisons of variables of interest]

Variable (mean ± standard error)	FD off-PPI (n= 29)	HC off-PPI (n= 30)	FD vs. HC off-PPI (p-value)	FD on-PPI (n= 29)	FD on- vs. off-PPI (p-value)
Dextran-passage (pmol)	23.46 ± 2.32	16.38 ± 1.58	0.04	16.49 ± 1.93	0.007
Duodenal eosinophils (/ HPF)	12.88 ± 1.32	3.65 ± 0.55	<0.0001	4.71 ± 0.64	<0.0001
Perceived Stress Scale	14.29 ± 2.39	7.64 ± 1.11	0.03	12.67 ± 1.59	0.62
Cortisol 0 min (ng/mL)	7.55 ± 1.38	3.88 ± 0.50	0.03	2.88 ± 0.28	0.007
*AUCg (ngmin/mL)**	511.99 ± 74.02	320.14 ± 36.60	0.04	378.22 ± 36.17	0.16
*AUCi (ngmin/mL)**	58.73 ± 40.61	88.55 ± 19.98	0.16	205.37 ± 40.11	0.001

Open in a new tab

Conclusion

FD patients have higher baseline cortisol levels upon awakening vs. controls but increased incremental cortisol levels after PPI with similar perceived stress. Improved cortisol reactivity may be a marker of therapeutic response to PPI, as this was mediated by improved diet and food habits. The effect of PPI on cortisol levels did not involve duodenal permeability and eosinophilia, suggesting the role of other factors.

Disclosure

Nothing to disclose

References

1.Wauters et al. UEGJ 2019; 7. [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.148

OP158 Inter-Relationships of Duodenal Eosinophilia, Anxiety and Functional Dyspepsia

J Ronkainen ^1,^2,^✉, P Aro ³, M Jones ⁴, MM Walker ⁵, L Agréus ⁶, A Andreasson ⁷, NJ Talley ⁸

Introduction

Functional dyspepsia (FD) is a distressing disorder with a prevalence of around 15% in the population.¹,² A previous population-based endoscopic prospective study identified anxiety was strongly associated with new-onset FD.³ Increased duodenal eosinophils have been observed in FD with evidence of systemic immune activation (circulating homing small intestinal T-cells).² ⁴ ⁵. Epidemiological data suggest gut symptoms may precede the development of anxiety in functional gut disorders but the mechanisms are unknown.⁶

Aims & Methods

We hypothesized that duodenal eosinophilia is a cause of anxiety. Participants (n=3000) were randomly selected from the national Swedish population register and surveyed by a validated abdominal symptom questionnaire (ASQ) and hospital anxiety and depression scale (HADS); 1000 completed an esophagogastroduodenoscopy. All eligible from the latter cohort (n=887, response rate 79%) were invited to a follow-up in 2010 with the ASQ and HADS.

Functional dyspepsia was defined based on the Rome III.³ Anxiety and depression were defined by a score of 11 or higher on HADS. in a nested case-control study (FD n=89) vs. healthy controls (n=124), mean age 62 years (SD=12, 34% male), duodenal histology was evaluated at baseline (pre-specified cut off 23 eosinophils in duodenal bulb (D1) and 24 eosinophils in second part (D2)) .⁷

Data were analyzed by Fisher's exact test and unconditional logistic regression.

Results

Duodenal eosinophilia was observed in 78 subjects in D1 and in 84 subjects in D2 (in 46 subjects, both in D1 and in D2, P< 0.001 for both) at baseline. Anxiety at baseline was found in 9 subjects (4%) and at follow-up in 12 subjects (6%).

Anxiety at baseline was independently associated with duodenal eosinophilia in D2 (8/78 vs. 1/124, P=0.013, OR=15.0, 95% CI 1.76-127.1, adjusting for age, gender and FD at baseline) but not in D1 (p=0.2). Anxiety at follow-up was independently associated with duodenal eosinophilia in D1 at baseline (9/75 vs. 3/124, p=0.025, OR=5.90, 95% CI 1.25-27.90), but not D2 (p=0.5). Anxiety at follow-up was also independently associated with FD at baseline (10/83 vs. 2/116, P=0.004, OR=7.42, 95% CI 1.38-39.95), adjusting for age, gender and baseline FD status. New onset anxiety was more common in individuals meeting criteria for eosinophilia in D1 (7.5%) than those who did not meet criteria (1.8%). The association was strong (OR=4.48, 95% CI 0.84-23.75) but failed to reach statistical significance (p=0.08).

Conclusion

In a prospective population-based 10-year follow-up study, anxiety at baseline was independently associated with a 15-fold increased risk of eosinophilia in D2 while anxiety at follow-up was associated with a 5-fold increased risk with duodenal eosinophilia in D1. This is the first study to our knowledge to demonstrate a possible causal link between duodenal inflammation and psychological distress in a functional gut disorder.

We conclude duodenal eosinophilia may be one mechanism by which anxiety arises in FD.

Disclosure

Nothing to disclose

References

1.Talley N.J., Ford A.C. Functional Dyspepsia. N Engl J Med 2015; 373: 1853–63. [DOI] [PubMed] [Google Scholar]
2.Talley N.J., Walker M.M., Aro P. et al. Non-ulcer dyspepsia and duodenal eosinophilia: an adult endoscopic population-based case-control study. Clin Gastroenterol Hepatol 2007; 5: 1175–83. [DOI] [PubMed] [Google Scholar]
3.Aro P., Talley N.J., Johansson S.E. et al. Anxiety Is Linked to New-Onset Dyspepsia in the Swedish Population: A 10-Year Follow-up Study. Gastroenterology 2015; 148: 928–37. [DOI] [PubMed] [Google Scholar]
4.Powell N., Walker M.M., Talley N.J. The mucosal immune system: master regulator of bidirectional gut-brain communications. Nat Rev Gastroenterol Hepatol 2017; 14: 143–159. [DOI] [PubMed] [Google Scholar]
5.Liebregts T., Adam B., Bredack C. et al. Small bowel homing T cells are associated with symptoms and delayed gastric emptying in functional dyspepsia. Am J Gas-troenterol 2011; 106: 1089–98. [DOI] [PubMed] [Google Scholar]
6.Koloski N.A., Jones M., Talley N.J. Evidence that independent gut-to-brain and brain-to-gut pathways operate in the irritable bowel syndrome and functional dyspepsia: a 1-year population-based prospective study. Aliment Pharmacol Ther 2016; 44: 592–600. [DOI] [PubMed] [Google Scholar]
7.Ronkainen J., Aro P., Walker M.M. et al. Duodenal eosinophilia is associated with functional dyspepsia and new onset gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2019; 50: 24–32. [DOI] [PubMed] [Google Scholar]
8.Gray M.A., Chao C.Y., Stau-dacher H.M. et al. Anti-TNFalpha therapy in IBD alters brain activity reflecting visceral sensory function and cognitive-affective biases. PLoS One 2018; 13: e0193542. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.149

OP159 Safety and Efficacy of Apraglutide in Patients with Short Bowel Syndrome Intestinal Failure: A Double-Blind, Crossover, Randomized, Placebo-Controlled, Phase 2 Trial

J Eliasson ^1,^✉, MK Hvistendahl ¹, C Meyer ², F Bolognani ², PB Jeppesen ¹

Introduction

Treatment with glucagon-like peptide-2 (GLP-2) analogs promotes intestinal adaptation and can reduce parenteral support requirements in patients with short bowel syndrome intestinal failure (SBS-IF). Apraglutide is a novel long-acting GLP-2 analog designed for once weekly dosing.

Aims & Methods

This Phase 2 trial investigated the safety and efficacy of 5 and 10 mg apraglutide in patients with SBS-IF. Eight adult patients with SBS-IF were treated with apraglutide according to Table 1. Safety was the primary endpoint. Secondary endpoints included change from baseline in urine volume compared to placebo that was collected for 48-hours at home before and after each treatment period. Oral fluid intake and parenteral support were kept constant during all urine collections. An analysis of covariance was used to assess the effects of apraglutide on change from baseline compared to placebo.

Table 1.

[Trial design]

Part A: Treatment period 1 (4 weeks) Randomized	Washout 1 (6-10 weeks after last dose	Part A: Treatment period 2 (4 weeks) Randomized	Washout 2 (6-10 weeks after last dose)	Part B: Treatment period 3 (4 weeks) Open-label
Placebo or 5 mg apraglutide subcutaneously once weekly		Placebo or 5 mg apraglutide subcutaneously once weekly		10 mg apraglutide subcutaneously once weekly

Open in a new tab

Results

Common treatment-related adverse events were consistent with the physiological effect of GLP-2 and included polyuria (n=7), stoma complications (n=6), reduced stoma output (n=6), decreased thirst (n=4) and edema (n=2). Serious adverse events were evenly distributed between treatment periods and were not considered related to apraglutide. Compared to placebo, once weekly 5 and 10 mg apraglutide significantly increased urine volume by an adjusted mean of 711 mL/day (95% CI 132 to 1289; p< 0.05) and 795 mL/day (95% CI 195 to 1394; p< 0.05) respectively. No significant differences between the lower and higher dose of apraglu-tide were found.

Conclusion

This trial was the first once weekly, placebo-controlled phase 2 trial of a long-acting GLP-2 analog in patients with SBS-IF. Apraglutide was safe and well tolerated at both 5 and 10 mg and demonstrated significant and clinically relevant increases in urine volume. A phase 3 trial will be initiated to further confirm the safety and efficacy of once weekly treatment of apraglutide.

Disclosure

J. Eliasson: None Declared, M. Hvistendahl: None Declared, C. Meyer Other: Employee of VectivBio AG., F. Bolognani Other: Employee of VectivBio AG., P. Jeppesen Consultant for: VectivBio AG., Speakers Bureau of: VectivBio AG.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.150

OP160 Duodenal Mucosal Resurfacing Combined with Glp-1 Receptor Agonism May Eliminate Insulin Treatment in Type 2 Diabetes While Improving Glycaemic Control and Overall Metabolic Health

S Meiring ^1,^✉, ACG van Baar ², P Smeele ³, T Vriend ⁴, F Holleman ⁵, M Barlag ⁶, N Mostafavi ⁷, JGP Tijssen ⁸, M Soeters ⁹, M Nieuwdorp ¹⁰, JJ Bergman ¹¹

Introduction

Duodenal mucosal resurfacing (DMR) is an endoscopic intervention in which the duodenal mucosa is ablated by hydrothermal energy. DMR has been shown to improve glycaemic control in type 2 diabetes mellitus (T2DM), probably through altered enter endocrine signalling from the duodenum causing insulin sensitization. We studied the feasibility of eliminating exogenous insulin therapy in T2DM by combining DMR with gluca-gon like peptide-1 receptor agonism (GLP-1RA) and lifestyle counselling.

Aims & Methods

Single arm, single centre study in 16 insulin treated T2DM patients (long-acting insulin only, HbA1c ≤8.0%, c-peptide ≥0.5 ng/ ml). After a single endoscopic DMR, insulin therapy was discontinued. Patients followed a 2-week post-procedural diet where after GLP-1 RA (liraglutide) was introduced. Life style counselling was provided throughout the study. The primary endpoint was the percentage of patients free of insulin with an HbA1c ≤ 7.5% at 6 months and 12 months (responders). Secondary endpoints were change in haemoglobin A1c (HbA1c), homeo-static model assessment of insulin resistance (HOMA-IR), fasting plasma glucose (FPG), mixed-meal postprandial glucose levels, metabolic parameters (body mass index [BMI]), total body fat percentage, liver proton density fat fraction [PDFF]) at 6 and, if applicable, at 12 months. A sub analysis was performed to assess change in insulin dose in non-responders after DMR compared to baseline.

Results

ll 16 patients underwent successful DMR without procedure-related serious adverse events during follow-up. At 6 months, 75% of patients (12/16) (95% CI: 0.48 - 0.93) was off insulin therapy with an HbA1c ≤ 7.5%. This was associated with significant improvement of all glycaemic and metabolic parameters in the responders (Table 1). Median HbA1c improved from 7.4 to 6.7% (A -0.6% (95% CI: -0.9 to -0.2; p=0.009) and HOMA-IR from 8.9 to 2.5 (A -5.9 (95% CI -9.9 to -2.6); p=0.006). PDFF improved significantly from 8.1 to 4.6% (A -3.7% (95% CI -6.6 to -0.); p=0.016). BMI and total body fat percentage decreased significantly. At 12 months, 56% (9/16) of patients was still off insulin therapy with a median HbA1c of 6.7 %. Glycaemic and metabolic parameters maintained significantly improved (Table 1). in non-responders, the median insulin dose decreased from 36 units per day at study entry to 17 units per day at 12 months.

[Overview of glycaemic and metabolic secondary endpoints. Values are expressed as medians (Q1-Q3). Paired Wilcoxon signed-rank tests were used.]

	Intention to treat population (n=16)			Responders
	Baseline	6 months, p-value	12 months, p-value	Baseline	6 months (n=12), p-value	12 months (n=9), p-value
Glycaemic parameters
HbA1c [%,mmol/mol]	7.5 (7.1-7.9) 58 (54-63)	7.0 (6.7-7.9), 53 (50-63), p=0.178	7.3 (6.6-8.2), 56 (49-66), p=0.69	7.4 (7.1-7.6), 57. (54 - 60)	6.7 (6.6-7.3) 50 (49-56), p=0.009	6.7 (6.5-7.2) 50 (48-55), p=0.024
HOMA-IR	8.1 (4.3-11.8)	2.7 (1.6-4.7), p=0.002	3.9 (2.8-8.1), p=0.023	8.9 (4.5-13.3)	2.6 (1.4-4.1), p=0.004	7.1 (6.7-7.7), p=0.008
FPG [mmol/l]	10.1 (8.9-12.0)	8.0 (6.6-9.5), p=0.011	7.1 (6.6-9.5), p=0.01	10.5 (9.2-12.0)	7.6 (6.5-8.8), p=0.003	3.6 (1.8-6.7), p=0.015
Metabolic parameters
BMI [kg/m2]	29.3 (26.5 - 32.0)	25.8 (24.3 - 29.8), p=0.000	26.4 (22.7-29.5), p=0.000	29.8 (26.5 - 34.2)	27.2 (24.3 - 31.9), p=0.002	25.5 (22.1-29.5), p=0.008
Total body fat [%]	31.2 (29.3 - 41.0)	28.3 (24.9 - 38.3), p=0.000		32.6 (29.0 - 41.0)	31.1 (24.5 - 38.3), p=0.002
PDFF [%]	8.1 (4.0 - 13.5)	5.3 (3.9 - 11.4) ‡, p=0.053		8.1 (5.1 - 13.2)	4.6 (2.4 - 11.0) ‡, 0.016

Open in a new tab

Conclusion

A single endoscopic DMR procedure, combined with GLP-1 RA and lifestyle counselling can effectively eliminate the need for insulin therapy in the majority of our T2DM patients while improving glucose regulation and overall metabolic health.

Disclosure

This study was funded by Fractyl Laboratories Inc. Declaration of Interests ACGVB has no competing interests. SM has no competing interests. PS has no competing interests. TV has no competing interests. FH reports speaker fees from Sanofi, Bioton, Astra Zeneca, and Boehringer Ingelheim. MB has nothing to disclose. NM has no competing interests. JGPT has no competing interests. MRS has no competing interests. MN has no competing interests. JJGHMB received research support from Fractyl for IRB-based studies and received a consultancy fee for a single advisory board meeting of Fractyl in September 2019.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.151

OP161 Development of An Original Automated Method of Three-Dimensional Reconstruction of An Extended Field of View of The Gastric Antrum

T Bazin ^1,^✉, T-B Phan ², DH Trinh ², F Marzani ³, D Wolf ², C Daul ², D Lamarque ¹

Introduction

The detailed description of the digestive mucosa by endoscopy suffers from a lack of inter- and intra-observer reproducibility because the procedure is operator-dependent, depends on examination conditions and images quality. The three-dimensional (3D) reconstruction of a mucous surface with its texture from only endoscopic images would improve the reproducibility of the examination and its reinterpretation, while ensuring the monitoring of the evolution of lesions over time. 3D reconstruction of segments of the digestive tract presents many technical difficulties. Until now, no method of extending the 3D field of view has ever been described for the digestive tract.

Aims & Methods

The aim of this study, using an artificial intelligence (AI) algorithm, consists of a reconstruction of an extended 3D field of view visualizing in detail the mucosa of the gastric antrum from a recording of the endoscopic examination in white (WL) and blue-green (BGL) light. in this study, we collected eight high-definition videos in WL or BGL (NBI Olympus) during the endoscopic exploration of the gastric antrum of patients, after obtaining their consent to participate in the study. The acquisition was made by helicoidal progression of the endoscope into the antrum. The 3D reconstruction was carried out by the so-called “structure from movement” method (Structure-from-motion, SfM) which determines the shape of surfaces only from 2D images from different camera points of view. This method required a prior processing of the images to correct the distortion of the camera optics (correction of barrel effects due to short focal lengths of gastroscopes). Reconstruction of the mucosal surface involves three stages.

Step 1. The 2D homologous points between images correspond to the projection of the same 3D point located on the surface of the mucosa in images acquired from different viewpoints of the camera. Obtaining many groups made up of a maximum of homologous points is decisive for obtaining a precise surface. This homologous point search is able to deal with strong illumination changes and images including few texture information.

Step 2. A point cloud located on the surface to be reconstructed and the trajectory of the camera are simultaneously calculated. These calculations are based on finding the position and orientation of the camera at the time of the acquisition of each image, as well as on triangulation techniques which make it possible to find the position of 3D points from 2D homologous points.

Step 3. This point cloud is used to build a mesh surface. As the position and orientation of the camera are determined for each acquisition, the colors and textures contained in the images can be projected onto the surface which is thus covered with the information seen in the video-sequences. This original method is fully automated, works for a stomach in motion, and can deliver a 3D surface of the antrum about one hour after the end of the recording.

Results

We obtained a precise 3D reconstruction of the surface of the antral mucosa observable from all points of view and comprising no artifact of light or the interposition of bubbles or deposits.

Conclusion

We describe an original method of image analysis by AI allowing to obtain a precise and visually coherent 3D reconstruction of the surface of a digestive structure like the gastric antrum. This technique marks the beginning of a decisive progress allowing for the presentation, reinterpretation and comparison of endoscopic examinations independently of the operator.

Disclosure

Nothing to disclose

References

Miranda Rosbet, Walter Blondel Christian Daul, Yahir Her-nandez-Mier and Didier Wolf. A simplified method of video-endoscopic image barrel distortion correction based on grey level registration. IEEE Int. Conf. on Image Processing, pages 3383–3386, Singapore, 2004. [Google Scholar]
Trinh Dinh-Hoan, and Daul Christian On illumination-invariant variational optical flow for weakly textured scenes, Computer Vision and Image Understanding, Vol. 179, Issue C, pages 1–18, 2019. [Google Scholar]
Phan Tan-Binh, Trinh Dinh-Hoan, Lamarque Dominique, Wolf Didier, Daul Christian Dense Optical Flow for the Reconstruction of Weakly Textured and Structured Surfaces: Application to Endoscopy, IEEE International Conference on Image Processing (ICIP), Tapei, Taiwan, October 2019.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.152

OP162 Highly Accurate Ai Systems To Predict The Invasion Depth of Gastric Cancer: Efficacy of Conventional White-Light Imaging, Non-Magnifying Narrow-Band Imaging and Indigo-Carmine Dye Contrast Imaging

S Nagao ^1,^✉, Y Tsuji ¹, Y Sakaguchi ¹, Y Takahashi ¹, C Minatsuki ¹, K Niimi ¹, H Yamashita ², N Yamamichi ¹, Y Seto ², T Tada ^3,^4,⁵, K Koike ¹

Introduction

Endoscopic resection has become the standard method of treatment for gastric cancer (GC) worldwide. However, one of the most important preoperative criteria for curative endoscopic resection is the tumor invasion depth; curative endoscopic resection can be achieved for most of the intramucosal cancers and a part of the cancers with submu-cosal invasion <500 μm, while surgical resection is required for GC with deeper invasion. Although the efficacy of evaluating macroscopic features and endoscopic ultrasonography has been reported, there is a need for more accurate and objective methods. Artificial intelligence (AI) has recently attracted much attention, but there is still little evidence concerning its efficacy in predicting the invasion depth of GC.

Aims & Methods

The aim of this study was to develop AI systems to more accurately predict the invasion depth of GC using training datasets with multiple images from different angles and distances. Moreover, while previous reports have focused only on white-light imaging (WLI), the effect of non-magnifying narrow-band imaging (NBI) or indigo-carmine dye contrast imaging (Indigo) on the diagnostic ability of AI remains to be elucidated.

Therefore, we also investigated the diagnostic ability of AI diagnosis with these modalities. A total of 16557 images from 1084 cases of GC for which endoscopic resection or surgery was performed at the University of Tokyo Hospital between January 2013 and June 2019 were extracted. Cases were randomly divided into training and test datasets at a ratio of 4:1. Through transfer learning leveraging a convolutional neural network architecture, ResNet50, 3 independent AI systems were developed, to predict the invasion depth of GC using WLI, non-magnifying NBI and Indigo, respectively. The primary endpoint was the lesion-based accuracy for predicting sub-mucosal invasion ≥ 500 μm or deeper.

Results

There were no significant differences between the background factors of the training and test datasets for WLI, including the male percentage (75.9% vs 74.6%, p = 0.67), mean age (70.4 years vs 71.7 years, p = 0.10). The area under the curve of the AI system using WLI was 0.9590. The newly-developed AI systems in this study each achieved a high diagnostic accuracy for the prediction of invasion depth of GC; the trained AI system (WLI) could diagnose the invasion depth of GC with as high as 94.4% image-based accuracy. The lesion-based sensitivity, specificity, accuracy, positive predictive values and negative predictive values of the AI system using WLI were 84.4%, 99.4%, 94.5%, 98.5%, and 92.9%, respectively. The lesion-based accuracy of the AI system using WLI, NBI, and Indigo were 94.5%, 94.3% and 95.5%, respectively, with no significant difference.

Conclusion

We developed new high-quality AI systems predicting the invasion depth of GC using a training dataset of multiple images from different angles, distances, leading to a highly accurate AI depth diagnosis. While the lesion-based accuracy of the AI systems using WLI, NBI, and Indigo were not significantly different, it remains to be elucidated whether AI systems using NBI or Indigo might achieve a higher accuracy after using a larger training dataset.

Disclosure

Tomohiro Tada MD, PhD: AI Medical Service Inc. CEO and Shareholder

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.153

OP163 Usefulness of The Algorithm of All-In-Focused Images in Imaged Enhanced Endoscopy For Colorectal Neoplasm

T Yamamura ^1,^✉, T Kuno ², M Nakamura ², K Maeda ³, T Sawada ¹, Y Mizutani ⁴, E Ishikawa ², T Ishikawa ², K Furukawa ¹, N Kakushima ², E Ohno ², H Kawashima ⁵, M Fujishiro ⁶

Introduction

Magnifying endoscopy is useful for diagnosis of quality assessment and invasion depth and colorectal neoplasms with image-enhanced endoscopy (IEE). It is sometimes difficult to focus on the whole target because the image includes focused and defocused areas owing to the depth of the target area or irregular surface. in addition to it, the target lesion will easily move by peristaltic movement in endoscopic observation of the colon. A new innovative assistance system for overcoming these problems is demanded.

We applied the all-in-focus (AIF) technique which was developed in the engineering field to the images of colorectal neoplasms with magnification. The aim of this study was to evaluate magnifying endoscopic image with AIF algorithm.

Aims & Methods

Ten colorectal neoplasms were examined by magnification of IEE, and four cases were examined by magnification of pit pattern imaging. Images were acquired at about 80x magnification for about 60 s, using an endoscope with a magnifying function. One still image was synthesized by selecting a pixel value of the most highly evaluated image based on individual pixels, using the image quality measurement (IQM) value. The original and AIF images at certain situation were extracted one by one and a total of 28 images were randomly rearranged. The adjustment of focus range and recognition of surface pattern, vessel pattern and JNET (The Japan NBI Expert Team) classification itself were evaluated on a 5-point Likert scale defined as 5 to excellent.

The original image and the image processed by the AIF technique were compared by four endoscopists, two experts and two non-experts, who were blinded to the patient characteristics. We also evaluated the previous two type of images stained Crystal violet by pit pattern diagnosis in the same manner.

Results

The Likert scale of AIF image with adjustment of focus range and recognition of vessel pattern and JNET was significantly improved compared to the original image with using JNET classification (focus range 4.0Å4.3, vessel pattern 3.9Å4.4, JNET 4.0Å4.4), but recognition of surface pattern was not changed. in the same manner, the Likert scale of AIF image with adjustment of focus range and the recognition of pit pattern was also improved compared to the original image (focus range 2.4Å3.4, pit pattern 2.8Å3.7). When we evaluated the amount of time required to diagnose, there was no significant difference between the original image and the focus adjustment image in both JNET diagnosis and pit pattern diagnosis.

Conclusion

Endoscopic images with AIF algorithm gave a better image quality and improved the facility of evaluation and diagnosis of colorectal neoplasms as subjective evaluation. This technique may be useful to resolve the difficulties with magnifying endoscopic observation.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.154

OP164 Development and Regulatory Approval of An Artificial Intelligence-Assisted Detection System For Colonoscopy

T Matsuda ^1,^✉, S Kudo ², M Misawa ², Y Mori ², K Hotta ³, K Ohtsuka ⁴, S Saito ⁵, H Itoh ⁶, M Oda ⁶, K Mori ⁶

Introduction

One-quarter of neoplastic polyps may be missed in colonoscopy reading, and more than half of post-colonoscopy colorectal cancers might arise from these missed lesions. Computer-aided detection (CADe) systems using artificial intelligence are attracting widespread attention, because they can help to reduce the possibility of missing polyps during colonoscopy as a result of perceptual errors. However, few CADe systems have obtained regulatory approval.

Aims & Methods

The aim of this study was to develop a CADe system and to conduct a benchmark test to obtain regulatory approval. First, we developed a CADe system based on deep learning algorithms that emits an alert with sound and color if it detects colorectal lesions. We collected colonoscopy videos as learning samples from five academic centers in Japan. Second, to obtain regulatory approval, we conducted a benchmark test to evaluate the performance of the CADe system with the supervision of the Japanese regulatory body. We prospectively collected colonoscopy videos from October 2018 to March 2019 at Showa University Northern Yokohama Hospital. We excluded all recorded subjects who (i) were diagnosed with inflammatory bowel disease, (ii) were diagnosed with polyposis disease, (iii) had non-epithelial lesions, (iv) had polyps recorded only on low-quality frames with artifacts, and (v) did not have lesions recorded with white-light endoscopy. Every frame of all videos was annotated regarding the presence or absence of polyps in the frame. The main outcome was the CADe system's sensitivity and specificity for colorectal lesions. We also evaluated the sensitivity and specificity of the artificial intelligence system in finding diminutive (< 5 mm) polyps, protruded lesions, and flat polyps. We calculated the sample size required to make the lower limit of the 95% confidence interval (95% CI) of the system's sensitivity > 90%. This resulted in an estimated sample size of 298, assuming alpha error = 0.05 and 1-beta = 0.80. Therefore, we randomly extracted 300 lesions from the aforementioned recorded video database. We randomly extracted polyp-negative videos from the database to make the number of polyp-negative frames 10 times larger than that of polyp positive videos.

Results

During the study period, we recruited 1731 patients. Among them, 1405 patients were eligible, and their colonoscopy videos were recorded. of the eligible patients, 797 had at least one colonic lesion, and 1635 total lesions were registered in the database. We randomly extracted 300 lesions from all of the included subjects’ combined data as the test set. The developed CADe detected 295 of the 300 test lesions (sensitivity: 98.3%). in per-frame analysis, the sensitivity and specificity were 89.0% and 93.1%, respectively. Among the study samples, the numbers of diminutive, protruded, and flat lesions were 204, 176, and 124, respectively. The per-lesion sensitivity values for diminutive, protruded, and flat lesions were 98.0%, 99.4%, and 96.8%, respectively.

Conclusion

We developed a CADe system for colonoscopy that has sufficient sensitivity and specificity. The developed CADe system is under review for regulatory approval in Japan.

Disclosure

Shin-ei Kudo, Masashi Misawa, and Yuichi Mori received lecture fees from Olympus Corp. Kazuo Ohtsuka reports personal fees and nonfinancial support from Olympus outside the submitted work. Kensaku Mori reports grants from Cybernet Systems during the time of this study, and grants from Olympus Corp., Morita Corp., and NTT outside the submitted work. The other authors have no conflicts of interest.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.155

OP165 Artificial Intelligence Using Convolutional Neural Networks For Detection of Early Barrett'S Neoplasia

M Abdelrahim ^1,^✉, M Saikou ², Y Masaike ³, S Ohtsuka ⁴, N Maeda ⁴, E Hossain ¹, S Arndtz ¹, H Ikeda ⁴, P Bhandari ¹

Introduction

Endoscopic detection of early Barrett's neoplasia remains very challenging, with significant inter-observer variation between endoscopists. Artificial intelligence is proposed to play a role in this field and could have significant clinical and cost implications. We aim to develop and validate a deep learning (DL) algorithm using Convolutional Neural Networks (CNN) for detection of Barrett's neoplasia.

Aims & Methods

We collected 621 high definition white light endoscopy images from 43 lesions (patients) of histologically confirmed Barrett's neoplasia. These images were marked and annotated using specially designed software, and reviewed by two experts. Another 23183 images of non dysplastic were prepared from 44 patients and used as control. Both dysplastic and non dysplastic images were divided into three datasets and used separately for training, validation and testing of CNN algorithm. We used VGG16 architecture for classification and SegNet architecture for delineation. These algorithms were trained to classify and detect images and videos as dysplastic or non dysplastic. Localisation of Barrett's dysplasia was shown in a heat map fashion. Graphic processing unit used was “GeForce RTX 2080 Ti”. We collected metrics on processing speed, sensitivity, specificity and global accuracy.

Results

The model was tested on 107 images and 20 videos of dysplastic Barrett's, and 364 images and 14 videos of non dysplastic. This testing dataset was completely different from training and validation data. Image processing speed of the classification model and the delineation model was 5ms/image and 33ms/image, respectively. This is much faster than the average human visual response latency which is estimated at 70-100ms. The classification algorithm was able to detect Barrett's neoplasia with sensitivity of 96.3%, specificity of 97.0% and global accuracy of 96.8%.

Conclusion

Our model was able to localise and classify Barrett's neoplasia with high accuracy, with sensitivity and specificity meeting PIVI criteria. The ultra short processing time makes this algorithm suitable for real time detection of Barrett's neoplasia.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.156

OP166 Using Large Sample Real-World Data To Study The Progression of Low Risk Pancreatic Cysts: New-Onset Diabetes As A Potential Biomarker of Malignant Transformation

A Schweber ^1,^✉, C Brooks ², E Agarunov ¹, T Gonda ¹

Introduction

Although the risk of the transformation of pancreatic cystic neoplasms to pancreatic cancer (PC) has been extensively investigated, prior study has been significantly limited by small sample sizes and selection bias, yielding highly variable progression results. Furthermore, due to these study limitations, it has been difficult to reliably identify novel risk factors and biomarkers for malignant transformation, such as patient metabolic status. Real-world data, with its large sample sizes, has become a powerful tool for the study of the natural history of disease.

Aims & Methods

Using a large, real-world sample of patients diagnosed with a low risk pancreatic cyst, quantify the risk of progression to PC and assess if metabolic status can be used to risk stratify patients. Patients were sourced from the IBM Marketscan insurance claims databases, covering over 180 million patients from 2008 to 2017. ICD-9/10 codes were used to identify patients with low-risk pancreatic cysts. Patients with a history of worrisome features or prior pancreatic pathology were excluded: PC; acute or chronic pancreatitis; jaundice; steatorrhea; congenital disease; surgery; injury; or other pancreatic disease. Patients were evaluated for at least two years following cyst diagnosis.

Patients who developed PC or had surgery within 6 months of diagnosis were excluded. Kaplan-Meier analysis and Cox proportional hazard modeling were used to assess the risk of progression to PC and the association of progression with metabolic status. Results were standardized by age and insurance coverage.

Results

From the 137,970 patients with a diagnosis of pancreatic cyst, 14,279 low risk patients fulfilled study criteria. At 5 years, the adjusted cumulative progression rate to PC was 2.2% (95% CI:1.8%-2.6%). The cumulative risk of progression increased at a nearly linear rate over 7 years (R =0.979;p< 0.001). A prior history of diabetes (HR=2.01;95% CI: 1.89-2.14) and the development of new onset diabetes (NODM) following cyst diagnosis (HR=3.24;95% CI: 3.03-3.46) were both associated with a greater risk of progression.

The 5 year adjusted progression rates to PC for patients without diabetes, a prior history of diabetes, and NODM were respectively: 1.5% (95%CI:1.1%-1.9%); 3.1% (95%CI:2.7%-3.5%); 4.6% (95%CI:3.1%-6.1%) (Table. 1).

[Table. Projected cumulative risk of progression to pancreatic cancer by metabolic status]

Years Following Cyst Diagnosis	All Patients	No Prior DM	Prior DM	New Onset DM
3	1.4% (1.2-1.7%)	1.1% (0.8-1.3%)	1.9% (1.6-2.2%)	3.1% (2.0-4.2%)
5	2.2% (1.8-2.6%)	1.5% (1.1-1.9%)	3.1% (2.7-3.5%)	4.6% (3.1-6.1%)
7	3.2% (2.5-3.9%)	2.0% (1.2-2.8%)	4.6% (3.8-5.4%)	6.9% (3.7-10.0%)

Open in a new tab

Conclusion

This is the first large scale, real world study of the malignant progression of low risk pancreatic cysts. Using significantly larger sample sizes than prior studies, results suggest that the risk of progression is small among patients with low risk cystic lesions, but increases linearly with time. The finding that new onset diabetes is associated with a significantly higher risk of progression, provides strong support for the inclusion of HbA1C, fasting glucose, and other markers of diabetes mellitus in cyst surveillance algorithms.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.157

OP167 Natural History of Intraductal Papillary Mucinous Neoplasms: A Prospective Study

N Porter ¹, MG Keane ^1,^✉, V Sawant ¹, AV Santharaman ¹, E Afghani ¹, Q Nartey ¹, R Hruban ¹, A Zaheer ¹, E Fishman ¹, M Goggins ¹, MI Canto ¹, S Devlin ¹, A Salamone ¹, L Manos ¹, R Burkhart ¹, W Burns ¹, J He ¹, C Wolfgang ¹, A Klein ¹, AM Lennon ¹

Introduction

Intraductal papillary mucinous neoplasms (IPMNs) are a common clinical finding and are one of three recognized precursors of pancreatic ductal adenocarcinoma (PDAC). PDAC is the 3rd most common cause of cancer mortality with a 5-year survival of only 9%. Surveillance is recommended for patients with an IPMN, however little prospective data exist on the risk of malignant transformation of IPMNs, and the impact of surveillance on early cancer detection.

Aims & Methods

Study aim: We performed a prospective cohort study to evaluate the risk of malignant transformation in participants with clinically defined IPMNs.

Methods: Patients with a clinical diagnosis of an IPMN attending The Johns Hopkins Hospital Multidisciplinary Pancreatic Cyst Clinic between 1^st January 2010 and 31^st September 2019 were prospectively enrolled and followed.

Results

There were 1656 participants enrolled. The median age was 66 (range 18-92) years, of whom 64% were female. The median follow-up of the 1656 participants was 12 (range 0 to 102) months. There were 1010 participants with greater than 6 months follow-up (range 6-103 months. There were 133 (8%) individuals who were diagnosed with high-grade dysplasia (n=49) or PDAC (n=84) during the study period. The majority of these (111; 84%) were identified within 6 months of enrollment. of the 1010 participants with greater than 6 months follow up, 22 (2.2%) were diagnosed with high-grade dysplasia or PDAC. Overall, 118 (89%) of the 133 participants with high-grade dysplasia or PDAC were stage II or less.

Conclusion

In this large prospective cohort study of patients with a clinically defined IPMN, the risk of progression to HGD or invasive PDAC was relatively low. 89% of IPMNs with malignant transformation were identified with high-grade dysplasia or localized PDAC. This contrasts with the SEER data, where only 11% of patients with a PDAC have localized disease.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.158

OP168 Icyst - Implementing The Dissemination of European Evidence-Based Guidelines On Pancreatic Cystic Neoplasms Through An App For Smartphones, Preliminary Results of An Ueg Dissemination of Existing Clinical Practice Guidelines and Standards Project

A Balduzzi ^1,^✉, G Marchegiani ¹, S Andrianello ¹, T Pollini ¹, A Caravati ¹, M Biancotto ¹, D Bonamini ¹, E Secchettin ¹, C Bassi ¹, H Friess ², R Salvia ¹

Introduction

European evidence-based guidelines together with other existing guidelines are well known in the community of pancreatologists. Their limited disseminations in Europe, especially in countries with a low human development index, is a matter of concern as it may cause serious inequalities in terms of treatment.

Aims & Methods

The aim of the study was to implement guidelines dissemination through a digital app. A digital app (iCyst) was released for iOS and Android on October 2019. The app provided both the access to three different guidelines (European evidence-based, International Association of Pancreatology, American Gastroenterological Association) and the possibility to enter data of simulated cases. At the end of each simulation, the app released the recommendation of each guideline and the user indicated which one he would consider (European, IAP, AGA or none).

Results

A total of 840 users from 33 countries downloaded the app worldwide. Among them, 132 users entered a total of 380 simulated cases. Most represented specialties among users were surgery (72%) and gastroenterology (24.9%). Three different tricky scenarios were identified among simulated cases: an IPMN with main duct dilatation from 5 to 9.9mm (case #1), a large (> 40mm) branch duct IPMN (case #2) and a small (< 40mm) branch duct IPMN (case #3). For case #1, 38.4% of user would have followed European guidelines, 26.9% IAP, 6.5% AGA and 19.2% any guidelines. For case #2, 58.8% European guidelines, 0 IAP, 60% AGA and 5.8% any guidelines. For case #3, 52.9% European guidelines, 29.4% IAP, 17.7% AGA and 8.2% any guidelines.

Conclusion

European guidelines seem to be the most followed during case simulations among a digital community of professionals where surgeons are the most represented. However, main pancreatic duct dilatation and cyst size alone still represents a matter of concern which divides those who would choose a more aggressive attitude from those who prefer a conservative one.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.159

OP169 The Diagnostic Incidence of Pancreatic Cysts Nearly Doubled Over The Past Eight Years, Far Outpacing The Growth in The Use of Cross Sectional Abdominal Imaging

A Schweber ^1,^✉, C Brooks ², E Agarunov ¹, T Gonda ¹

Introduction

Despite intensive study of pancreatic cysts, little is known about their epidemiology, with most measures sourced from small studies. Accordingly, the reported prevalence has ranged from 0.21% to 24.3%.

Furthermore, studies regularly cite a recent increase in the incidental diagnosis of asymptomatic cysts, attributing this trend to the increasing use of abdominal imaging. Yet, these claims are based on small cohorts. Real-world data, with its large sample sizes, has become a powerful tool for studying the epidemiology of disease.

Aims & Methods

Using a large real-world sample of patients, to assess trends in the diagnostic incidence and prevalence of pancreatic cysts and estimate the undiagnosed prevalence of asymptotic lesions. Patients were sourced from the IBM Marketscan claims database of over 200 million (M) insured patients. For each year, a study population of over 17M patients was selected, consisting of patients continuously eligible for benefits in that year and the prior two calendar years. Prevalent cases were patients who received a ICD-9/10 diagnosis of cyst in that year. Incident cases lacked a diagnosis of cyst in the preceding 2-year lookback period. The period prevalence denominator included all eligible study patients for that year. The cumulative incidence denominator excluded patients with a lookback period cyst.

To assess the “real-world prevalence” of asymptomatic cysts, the study population was narrowed to imaged patients with no history of cyst or other pancreatic pathology. Using 2017 as the reference, results were directly standardized by age and insurance coverage.

Results

From 2010 to 2017, the standardized cumulative incidence and prevalence grew from 6.3 to 11.4 per 10,000 and 8.9 to 18.7 per 10,000, corresponding with annual percent changes (APC) of 8.7% and 11.2%, respectively (Table 1). This would imply that 609K patients received a diagnosis of pancreatic cyst in 2017, of whom 371K were newly diagnosed. Over this same period, the percent of patients undergoing cross sectional abdominal imaging grew from 8.0% to 9.4%, corresponding with an APC of 2.6% (Table 1).

Table 1.

[Trend Pancreatic Cyst Incidence vs. Trend in Abdominal Imaging]

Year	% of Patients with Abdominal Imaging	Cumulative Incidence of Cyst per 10,000
2010	8.0%	6.3
2011	8.1%	7.1
2012	8.2%	7.4
2013	8.4%	8.5
2014	8.7%	9.1
2015	9.1%	9.7
2016	9.3%	10.6
2017	9.4%	11.4
Annual Percent Change	2.6%	8.7%

Open in a new tab

Among the 722K imaged patients in 2017 without a prior history of pancreatic cyst or pathology, the real-world prevalence was 2.0%, imputing an estimated 6.5 million U.S. patients with a cyst. Prevalence increased exponentially with age until age 65. The diagnosis rate was significantly higher among patients initially imaged with MRI vs. CT.

Conclusion

This is the first large scale and standardized population-based study of the epidemiology of pancreatic cysts. The diagnosis of incident patients nearly doubled in the last 8 years, a growth rate that far outpaced the growth in the use of cross sectional imaging. The estimated real-world prevalence of pancreatic cysts is 2.0% and exceeds that diagnosed clinically by an order of magnitude.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.160

OP170 Exocrine and Endocrine Insufficiency in Autoimmune Pancreatitis: A Matter of Treatment Or Time?

S Nikolic ^1,^2,^✉, I Dahlman ¹, JM Löhr ^3,⁴, M Vujasinovic ^1,⁴

Introduction

Long term sequelae of autoimmune pancreatitis (AIP), such as exocrine and endocrine pancreatic insufficiency, are often described at onset and during the disease course. Studies exploring the implications of pharmacological treatment on exocrine and endocrine pancreatic function are either controversial or offer no firm evidence.

Aims & Methods

The aim of our study was to investigate influence of pharmacological treatment on endocrine and exocrine pancreatic function in patients with AIP. Medical records of all patients diagnosed with AIP at Karolinska University Hospital from January 2004 till December 2019 were retrospectively analysed (N=133). We included 74 patients into final analysis. Diagnosis and decision of treatment was established using International consensus diagnostic criteria (ICDC). Fifty-seven patients (77%) received pharmacological treatment. Screening for diabetes mellitus (DM) and pancreatic exocrine insufficiency (PEI) was performed at time of diagnosis of AIP and during follow up. DM was diagnosed with measurement of plasma glucose (fasting, random or 2 hours after oral glucose tolerance test - OGTT) and/or glycated haemoglobin (HbA1c). Values obtained at diagnosis (first contact) and last contact were used in analysis. Statistics was performed with IBM SPSS 26.0.

Results

Our cohort consists of 54.1% male and 45.9% female patients, median age at diagnosis was 50 years (min 14; max 80). Median follow up was 59.3 months (min 12.1; max 174.2). There were 74,3% definite, 25.7% probable, 74.2% type 1, 18.9% type 2, 6.8% AIP not otherwise specified (NOS). Majority of our patients (97%) had not consumed more than 5 units of alcoholic beverages per week; 61.2% had never smoked, whereas 35.8% used to smoke. We found that 71.7% of our patients were white collar workers. Patients with jaundice had significantly higher prevalence of PEI and DM at diagnosis and at last contact. Patients with either other organ involvement (OOI), IgG4 positive serology or weight loss had statistically higher prevalence of DM at last contact. Our results show statistically higher prevalence of PEI at diagnosis in patients with AIP type 1. There was no significant difference in first and last contact prevalence of PEI (57.1% vs. 57.1%) and DM (6.7% vs. 20%) in patients without treatment. Similarly, no significant difference in first and last contact prevalence of DM (26.8% vs. 35.7%) and PEI (65.1% vs. 56.8%) was observed in patients who received treatment. Stratifying to definite/probable AIP or stratifying to different pharmacological therapies exposed no differences between PEI and DM prevalence at diagnosis and at last contact.

Conclusion

Jaundice is the most important risk factor for the development of PEI and DM. Patients with OOI, positive IgG4 serology or weight loss seem to be at higher risk for developing DM during AIP disease course. Prevalence of exocrine and endocrine consequences in patients with AIP is high and is not related to the type of pharmacological treatment. Thus, long term follow - up of patients with and without treatment is advisable.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.161

OP171 The Transcriptome of Gluten-Specific T Cells in Blood After Gluten Challenge

S Zühlke ¹, EG Lund ¹, O Snir ², S Dahal-Koirala ¹, KEA Lundin ^2,^✉, LM Sollid ¹

Introduction

Gluten-specific T cells possess a distinct phenotype¹, providing the possibility to identify potential drug targets on disease-driving CD4+ T cells in coeliac disease (CeD). Many functional aspects of gluten-specific T cells remain to be studied. Here, we have performed a transcrip-tome analysis of gluten-specific T cells following gluten challenge with the aim to gain insights into characteristic features of these cells and to identify potential drug targets.

Aims & Methods

Ten treated CeD patients underwent a gluten challenge over three days and blood sampling was performed at baseline and on day 6 after initiation of gluten challenge. PBMCs were stained with and enriched for HLA-DQ2.5: gluten tetramers and enriched tetramer-binding gut-homing effector-memory T (tetramer+ integrin β7+ T_EM) cells on day 6 were sorted for subsequent bulk RNA sequencing. Gut-homing effector-memory T cells that did not bind with tetramers (tetramer- integrin β7+ T_EM) were used as control cells. Tetramer-positive gut-homing effector-memory T (tetramer+ β7+ T_EM) cells were FACS-sorted and collected for subsequent bulk RNA.

Results

All the participants responded to gluten challenge with a clear increase of gut-homing gluten-specific CD4+ T_EM cells in blood (median 4 vs. 53 tetramer+ integrin β7+ T_EM / 10⁶ CD4+ cells at baseline vs. day 6). Transcriptome analysis of tetramer+ integrin β7+ T_EM cells after gluten challenge from the five participants with highest response revealed 1249 highly significant differentially expressed genes. CD38, CCR9 and Ki-67 were found among the most upregulated transcripts. Similarly, BIRC5, CD28, CD38, thymosin β-10, CCR9, CXCR3 and CXCR6 as well as metabolic pathways involving components of mitochondrial complex III appears as possible drug target in CeD.

Conclusion

Tetramer+ integrin β7+ T_EM cells from blood after gluten challenge upregulated mRNA for several chemokine receptors, markers of activation and gut homing markers.

They show striking similarity in transcriptional profiles with gut resident cells¹, possibly indicating they are precursors on their way to the gut from a so far undefined origin. Pathway analysis showed a central role for mitochondrial metabolism, which may serve as a future target for drug intervention in CeD.

Disclosure

Nothing to disclose

References

Christophersen A., Lund E.G., Snir O. et al. Distinct phenotype of CD4+ T cells driving celiac disease identified in multiple autoimmune conditions. Nature Medicine 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.162

OP172 Potential Celiac Disease Patients Lack Activating Nk Receptors and Epithelial Stress Markers Up-Regulation Even Prior To Develop Villous Damage

V Discepolo ^1,^✉, M Maglio ¹, L Greco ¹, R Auricchio ¹, R Troncone ¹

Introduction

The events leading to the cytotoxic activation of intraepi-thelial lymphocytes (IELs) and epithelial destruction in active celiac disease (ACD) have not been fully elucidated. The requirement for up-regulation of activating natural killer receptors (NKRs) on IELs (NKG2D and CD94/ NKG2C) and their ligands (MIC-A and HLA-E) in small intestinal epithelial cells was established in ACD¹.

However, the natural history of these changes has not been defined yet. Potential celiac disease (PCD) patients display a normal small intestinal mucosal architecture despite having increased anti-TG2/EMA antibodies in the serum, thus, representing a clinical model of loss of tolerance to gluten in the absence of tissue damage.

Aims & Methods

Taking advantage of a large cohort of PCD patients fol-lowed-up over time², we investigated the expression of activating NKRs and epithelial stress molecules to address whether their up-regulation may predict future development of villous damage. Duodenal biopsies obtained from 16 non-celiac controls, 28 ACD and 63 PCD patients were analyzed by immunohistochemistry. The total number of CD3+, TCR-yö+ and NKG2D+ IELs as well as the epithelial expression of IL-15 in the intestinal epithelial cells were assessed on frozen OCT-embedded biopsies. Epithelial expression of HLA-E, ligand for the CD94/NKG2C, was assessed on formalin-fixed paraffin-embedded sections. Unpaired t-test or Mann-Whitney test were performed to assess statistical significance.

Results

The density of IELs expressing NKG2D was not increased in PCD (mean± SD of positive cells/mm epithelium: 5.9±2.9), but only in ACD (18.7±7.9, p< 0.0001) in comparison to controls (5.9±3.3). Confirming previous findings³. No difference in the density of NKG2D+ IELs was found when comparing subjects with a Marsh 0 (5.8±2.9) to those with a Marsh 1 mucosa (6.4 ± 2.8), nor subjects in those with higher TCR-y5+ IELs counts (6.2±3.0 vs 4.9±2.3), suggesting that the increase in CD3+ or TCR-y5+ IELs does not associate with an activated NKR-phenotype in PCD. in the intestinal epithelial cells the expression of HLA-E was increased in ACD but not PCD as compared to controls, indicating a lack of epithelial stress. Analyzing serial biopsies collected over time from PCD patients who eventually evolved to villous atrophy, we observed no increase in NKG2D⁺ IELs or HLA-E in the epithelium preceding the development of tissue damage, suggesting that those events occur in late stage of CD. An increase in epithelial IL-15 expression was found in a subset of PCD albeit at a lower level than in ACD, suggesting that at these levels it might not be sufficient to elicit NKG2D expression in IELs.

Conclusion

PCD represent a unique model to investigate the events preceding tissue destruction in CD. in the absence of tissue damage no increase in activating NKRs or epithelial stress markers was observed even in PCD subjects who evolved to full-blown disease over time. Cytotoxic activation of IELs and expression of their ligands in intestinal epithelial cells seem to be late events in the natural history of CD. What drives tissue damage in a subset of PCD remains to be elucidated.

Disclosure

Nothing to disclose

References

1 Abadie V., Discepolo V., Jabri B. Semin Immunopathol (2012) 34: 551–566 [DOI] [PubMed] [Google Scholar]
2 Auricchio et al. Gastroenterology 2019; 157: 413–420 [DOI] [PubMed] [Google Scholar]
3 Setty M., Discepolo V. et al. Gastroenterology 2015; 149: 681–691 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.163

OP173 Redefining The Diagnosis of Type 1 Refractory Coeliac Disease Using Urine Gluten Immunogenic Peptides

S Coleman ^1,^✉, A Rej ¹, HA Penny ¹, G Wild ¹, DS Sanders ¹

Introduction

Refractory Coeliac Disease (RCD) is defined as ongoing symptoms or signs of malabsorption with associated villous atrophy (VA) despite strict adherence to a gluten free diet (GFD). However, assessment of dietary adherence is challenging. The utility of urine gluten immuno-genic peptides (GIP) in patients with RCD1 was assessed for the first time in the literature.

Aims & Methods

2553 patients with coeliac disease (CD) were reviewed at Sheffield Teaching NHS Foundation Trust between 1998 and 2019. 4.0% (n=103) of patients had RCD, with 64.1% (n=66) of these being classified as RCD1, and 35.9% (n=37) being classified as RCD2/complicated coeliac disease (CCD). From the RCD1 cohort, 22 patients (33.3%) were successfully treated with budesonide, with 44 patients (66.7%) having ongoing VA. All of these patients had had a dietetic review suggestive of good GFD adherence. RCD1 patients with ongoing VA (n=44) were invited to complete three gluten immunogenic peptides (GIP) tests, using rapid immunochro-matographic testing, following the collection of mid-stream urine samples. Ongoing gluten ingestion was defined as having at least one weak positive/positive urine GIP sample.

Results

At diagnosis, RCD 1 patients were significantly younger than RCD2/CCD patients (p=0.002). 38 RCD1 patients with ongoing VA were recruited (71.1% female [n=27], median age 60 years). 52.6% (n=20) of patients with RCD1 had three negative GIP tests, suggestive of strict GFD adherence. However, 47.4% (n=18) had at least one positive GIP result, suggestive of possible ongoing gluten exposure.

Conclusion

A high proportion of individuals with RCD1 appear to have ongoing gluten exposure despite reported strict GFD adherence, as assessed by urine GIP. Urine GIP may re-define and enable the accurate diagnosis of RCD1 in the future.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.164

OP174 Safety of Bifidobacterium Longum Ncc 2705 and Production of Its Serpin in Patients with Celiac Disease and Non-Celiac Gluten Sensitivity

B Otten ^1,^✉, C D'Urzo ², L Dupuis ², M Aquarius ³, T Klaassen ¹, P Duncan ², S Duboux ², T Nunes ², EF Verdu ⁴, G Bergonzelli ², F Troost ¹

Introduction

Celiac disease (CeD) is a chronic autoimmune enteropathy triggered by ingestion of gluten. Non-celiac gluten sensitivity (NCGS) is an emerging symptom-based condition triggered by ingestion of wheat and related cereals, in the absence of CeD or wheat allergy. The standard treatment of these two conditions is a gluten-free diet (GFD). However, adherence to a strict GFD is challenging due to accidental gluten intake. Expression of the human serine protease inhibitor elafin is decreased in the duodenum of CeD patients [1].

The probiotic Bifidobacterium longum (BL) NCC2705 expresses a bacterial homologue of elafin, serpin [2].

We showed that BL NCC2705 - through the production of serpin - ameliorates gluten immunopathology in a mouse model of CeD [3], but its effects in humans have never been tested.

Aims & Methods

The primary objective of this study was to evaluate the safety and gastrointestinal tolerability of BL NCC 2705 in CeD and in NCGS patients. The secondary objective was to investigate the presence of BL NCC2705 and its production of serpin in the duodenum. This was a double-blinded, randomized, placebo-controlled, cross-over trial in 18 patients with CeD following a GFD for ≥ 1 y and 20 self-diagnosed NCGS patients on a GFD for ≥ 6 wks. BL NCC2705 (1xE10 cfu) or placebo was administered in a capsule twice a day with a meal, over 2 periods of 3 days each. On day 4 of both cross-over periods, a naso-duodenal aspiration catheter reaching the distal duodenum was placed by gastroduode-noscopy.

After catheter positioning, administration of the last dose of probiotic or placebo (T0), and following a 3 g-gluten challenge, duodenal aspirates were collected at 20 min intervals up to 370 min (T370) to assess BL NCC 2705 and serpin amounts. BL NCC2705 and serpin were expressed in genome copies (gc) and pg per mL of duodenal fluid, respectively. Adverse events (AEs) and gastrointestinal (GI) tolerability (visual analogue scales) were recorded all along the study at home. AEs were also evaluated at the clinics during the test.

Results

No serious AEs were recorded. Other AEs and GI tolerability parameter scores (nausea, flatulence, bowel sounds, abdominal cramping, diarrhea, and vomiting) did not differ between probiotic and placebo groups. After probiotics intake on the test day, BL NCC2705 gc were detected in duodenal aspirates, both in CeD and NCGS patients. BL NCC2705 AUC_T0T370 was significantly increased compared to placebo by +3.2log gc/ mLper minute over 370min [95% CI,+2.0log to +4.4log; p=0.016]. in parallel, serpin AUC_T0T370 significantly increased compared to placebo by +36% [95% CI,+1% to +89%; p=0.016]. The concentration peak (C_max) of BL NCC2705 gc appeared 90 min in average after the probiotic intake in both CeD and NCGS patients with a significant rise of +6.7log gc/mL [95% CI,3.9log to 7.5log; p< 0.001] and of +5.6log gc/mL [95% CI,4.3log to 7.2log; p< 0.001], respectively. Similarly, Serpin C_max was reached 90 min in average after probiotic intake in both CeD and NCGS patients with a significant rise of +400%, from 4.9 pg/mL to 24.5 pg/mL [95% CI,104% to 739%; p< 0.001] and of +394%, from 4.9 pg/mL to 24.2 pg/mL [95% CI,110% to 622%; p< 0.001], respectively.

Conclusion

BL NCC2705 is well tolerated and safe for human consumption. The appearance of BL NCC2705 in duodenal aspirates is associated with a concomitant increase in serpin concentration. This provides a mechanistic basis for the potential use of BL NCC2705 as adjuvant treatment to GFD to protect from accidental gluten intake.

Disclosure

Nothing to disclose

References

1.Galipeau H.J., Wiepjes M., Motta J.P., Schulz J.D., Jury J., Natividad J.M., Pinto-Sanchez I., Sinclair D., Rousset P., Martin-Rosique R., Ber-mudez-Humaran L., Leroux J.C., Murray J.A., Smecuol E., Bai J.C., Vergnolle N., Langella P., Verdu E.F. 2014. Novel role of the serine protease inhibitor elafin in gluten-related disorders. Am J Gastroenterol 109: 748–756. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Ivanov D., Emonet C., Foata F., Affolter M., Delley M., Fisseha M., Blum-Sperisen S., Koch-har S., Arigoni F. 2006. A serpin from the gut bacterium Bifidobacterium longum inhibits eukaryotic elastase-like serine proteases. J Biol Chem 281: 17246–17252 [DOI] [PubMed] [Google Scholar]
3.McCarville J.L., Dong J., Caminero A., Bermudez-Brito M., Jury J., Murray J.A., Duboux S., Steinmann M., Delley M., Tangyu M., Langella P., Mercenier A., Bergonzelli G., Verdu E.F. 2017. A Commensal Bifidobacterium longum Strain Prevents Gluten-Related Immunopathology in Mice through Expression of a Serine Protease Inhibitor. Appl Environ Microbiol 83: e01323–17 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.165

OP175 Is Gluten Free Diet Deficient in Selenium? Prospective Interventional Study in Children with Celiac Disease

A Gopan ^1,^✉, M Sarma ¹, SK Yachha ¹, A Srivastava ¹, A Mathias ¹, N Neelu ¹, U Poddar ¹

Introduction

Wheat is a rich source of Selenium (Se), an essential trace element. Though clinically asymptomatic, dietary compliant celiac disease (CD) patients in disease remission may be prone to Se deficiency and its biological effects.

Aims & Methods

1.
To assess the absolute and functional deficiency of Se in gluten free diet (GFD) compliant CD children.
2.
To study the sequential effect of oral Se supplementation and dietary modifications in those deficient.

Asymptomatic CD children on compliant GFD ≥ 3 years in serological (anti-tissue transglutaminase antibody ≤3 times upper limit normal) and mucosal (Marsh grade < 2) remission were prospectively enrolled. Dietary selenium content (3 day recall), plasma Se levels, selenoproteins (SEPP1) and glutathione peroxidase-3 (GPX-3) were estimated at baseline. Se deficiency was defined as plasma Se levels < 85ug/L. Se-deficient patients (those consenting) were supplemented elemental Se (100ug/day) for 3 months followed by Se-rich (≥100ug/day) modified GFD for next 3 months. Response in growth parameters, absolute and functional Se levels were reassessed at 3 and 6 months.

Results

Of 77 screened, 51 CD [duration of compliant GFD 6(3-13)y; age at study enrolment 14(4.7-18)y] children were eligible for the study. Thyroid hypofunction, impaired glucose tolerance and low bone mineral density for age and gender were seen in 16 (39%), 3 (7.3%) & 12 (23.5%) children on baseline endocrine screen. All 51 patients were Se-deficient [plasma Se 34.5(19.3-53)μg/L].

Baseline dietary Se content [26.7(12-99) vs. 40.7(6.5-94)ug/d,p=0.009], plasma Se[33.5(19.3-45.9) vs. 34.7(23.8-53) |g/L,p=0.4], SEPP1 [3.1(2.2-5)vs. 3(1.9-4.4) |g/mL,p=0.7], GPX3[7.5(3.3-60.3) vs. 9.5(0.6-60.6)|g/mL,p=0.3] were present in pure vegans (53%) and ovo/non-vegans (47%). Table 1 shows response to Se supplementation (n=30) vs. untreated patients (n=11). Ten patients with irregular follow-up were excluded for analysis. Significant increase from baseline to 3 months was noted in SEPP-1 [3.0(2.1-5) vs. 3.9(2.4-8.3) μg/mL,p=0.02] but not in GPX-3 levels [7.5(3.4-60.6) vs. 6.3(3.2-40.7) |g/mL; p=0.3].

[Table1. Clinical and biochemical parameters in Se supplemented and non supplemented groups]

Parameter Median (range)	Se supplemented (n=30) Baseline	Se supplemented (n=30) 6mo	Se not supplemented (n=11) Baseline	Se not supplemented (n=11) 6 mo	p value
Plasma Se(μg/L)	34.1 (24.7- 46.9)	41.7(17-100)	35.5 (19.3 -53)	27.9(23.7-62.9)	0.01^*; 0.37^{^}
Delta Se (μg/L)	8.4(-17.4 to 54.1)	-6.4(-22.2 to 22.7)	0.04
SEPP1 (μg/mL)	3.05 (2.2 -5)	3.75(1.3 -5.6)	3.16 (2.8 - 4.4)	3 (1.6-5.1)	0.005^*; 0.48^{^}
Delta SEPP1 (μg/mL)	0.46 (-1.37 to 2.71)	-0.21 (-1.45 to 1.53)	0.04
GPX3 (μg/mL)	7.48 (3.4-60.6)	8.72(3.1 -34.1)	17.8(0.64 -60.3)	12 (1.36-58)	0.52^*; 0.06^{^}
Delta GPX3 (μg/mL)	-0.2(-42.3 to 9.6)	-2.3 (-18.7 to 2.3)	0.14
Weight (Kg)	38.5 (18-68)	41(18-71)	40(21-64)	43(22/78)	<0.001^*; 0.03^{^}
Height (cm)	146 (114- 174)	150 (119 -175)	145 (122 - 167)	147 (122- 175)	<0.001^*; 0.017^{^}

Open in a new tab

Delta levels indicate difference between values at baseline and 6 mo

^*

comparison between baseline and 6 mo in Se supplemented group

^{^}

non-supplemented group; No significant difference between baseline values among the 2 groups; Level of significance p<0.05

Conclusion

GFD-compliant CD have absolute and functional Se deficiency. Short term therapy improves the Se status and growth parameters. Sustained dietary supplementation and its long term clinical impact needs to be studied.

Disclosure

Nothing to disclose

References

1.Rybicka I., Krawczyk M., Stanisz E., Gliszczynska-Swigto A. Selenium in Gluten-free Products. Plant Foods Hum Nutr. 2015; 70(2): 128–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Combs G.F. Jr. Biomarkers of Selenium status. Nutrients 2015, 7, 2209–2236 [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.166

OP176 Gluten-Free Diet Influences The Human Gut Virome Composition

S Garmaeva ^1,^✉, A Gulyaeva ¹, T Sinha ¹, A Shkoporov ², AG Clooney ², SR Stockdale ², KM Daly ², LA Draper ², C Wijmenga ¹, A Kurilshikov ¹, J Fu ^1,³, C Hill ², A Zhernakova ¹

Introduction

The human gut harbors a complex ecosystem of microorganisms, including bacteria and prokaryotic and eukaryotic viruses. Bacterial communities of the gut have been shown to be associated with multiple human-health factors, lifestyle and diet, but the composition and dynamics of the gut virome, and the effect of diet on viral composition has been relatively underexplored.

Aims & Methods

We aimed to study the effect of a gluten-free dietary intervention (GFD) on viral communities using longitudinal analysis of fecal quantitative viromes in healthy adults. A Gluten Free Diet (GFD), that involves the total exclusion of wheat, barley and rye from diet, is currently the only treatment of celiac disease. Individuals with irritable bowel syndrome, gluten intolerance and other gastrointestinal complaints often follow a GFD aiming to improve their gut health, which makes GFD one of the most popular diets worldwide.

We collected stool samples from 11 individuals at 3 timepoints before, during and after the GFD. We performed shotgun metagenomic sequencing of DNA and RNA (cDNA) from virus-like particles (VLP) and DNA of total microbial community without whole-genome amplification. Metagenom-ic reads from VLPs sequences were assembled per sample and obtained scaffolds were subjected to validation as viral. By applying a database-independent approach we were able to obtain the quantitative dynamics of the gut virome, including ssDNA and RNA viruses. Taxonomic profiling of total microbial community was performed using MetaPhlAn2.

Results

In total, 41,014 viral genomes and fragments have been detected for the whole dataset. Taxonomic assignment of contigs revealed 15 viral families of prokaryotic and eukaryotic viruses, with a prevalence of unknown phages belonging to the order Caudovirales and the family Sipho-viridae. On average, each individual carried 2,143 viral genomes or fragments, 13.1% of which was present at all three timepoints. We observed high divergence of viral communities across individuals. The virome was stable at the family level during the dietary intervention, but showed substantial variation at the level of genera and species. in addition, we observed the decrease the abundance of viruses from the family Virga-viridae that are known to infect plants, including gluten-containing species during the gluten-free diet (p-value = 0.03). Additionally, pre-diet between-individual distances could explain 42% of the variation of post-diet between-individual distances (Mantel test with 9999 permutations, r=0.65, p-value=4e-04). This suggests the role of the viral composition prior to the diet in the similarity of the diet effect on the gut virome across individuals. Individuals with lower viral diversity before the dietary intervention showed more changes in their virome composition under the dietary intervention. This is consistent with earlier observations on bacteriome level, which suggest that rich bacteriome remains more stable during dietary intervention.

Conclusion

Individual gut virome is highly diverse across individuals. Gluten exclusion from the diet influences the virome composition, these changes are reversible after returning to the baseline diet. The effect of the diet on the human gut virome is dependent on the initial viral diversity.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.167

OP177 The Role of The Gut Microbiome in Mediating Lactose Intolerance Symptoms

Gois MDF Brandao ^1,^✉, Vila A Vich ^1,², T Sinha ¹, L Bolte ^1,², C Wijmenga ¹, J Fu ¹, A Kurilshikov ¹, A Zhernakova ¹

Introduction

Lactose Intolerance (LI) is a common condition that manifests through gastrointestinal complaints, caused by the malabsorption of lactose. The disease is characterized by the low expression and activity of lactase enzyme, resulting in a difficulty or inability to digest lactose throughout adulthood ¹². The inability to express lactase is inherited in a recessive autosomal manner. Nonetheless, the severity of gastrointestinal complaints after lactose intake varies greatly among patients and has been linked to numerous intrinsic and external factors. Recent studies have demonstrated a correlation between the gut microbiome composition and lactose metabolism^1,3. However, these studies do not acknowledge the potential relatedness among other variables of lactose metabolism. Furthermore, lactose metabolism association analysis have not been performed in a single large cohort, thus, limiting the understanding of potential interactions.

Aims & Methods

Here, we aim to investigate the interplay between genetics, gut microbiome, diet and lactose intolerance. We assessed the gut microbiome composition, genetic variants, dairy products intake and oc-curance of gastrointestinal complaints in 959 participants from the Dutch general population cohort LifeLines Deep. Microbiome composition data was obtained through metagenomic sequencing. Consumption of dairy products was assessed through food frequency questionnaires. Gastrointestinal symptoms were retrieved from health questionnaires. Among the participants of this study, 78 presented the recessive genotype for lactase deficiency. Using the described data, we performed correlation tests, linear regressions and mediation analyses.

Results

Within participants with the recessive trait for lactase deficiency, we observed significant positive correlations between the abundance of Bifidobacterium genus and dairy intake (p-value: 0.037) and between the abundance of Bifidobacterium genus and the occurance of gut complaints (p-value: 0.032). Interestingly, we did not find any significant associations between dairy intake and the occurance of gut complaints (p-value: 0.284). Subsequently, a mediation analysis between dairy intake and gut complaints was performed, using Bifidobacterium abundance as a mediator. The mediation analysis showed a trend towards statistical significance (p-value: 0.094), suggesting that the gut microbiome mediates the occurance of gut complaints upon dairy intake.

Conclusion

Altogether, our results indicate that the gut microbiome plays a key role in lactose metabolism. Thus, elucidating that, in lactose intolerance, the occurance of gastrointestinal complaints is influenced by interactions between host genetics, dairy intake and gut microbiome. Moreover, the mediative role of the gut microbiome associated with oc-curance of symptoms in patients with LI emphasizes the complexity of lactose metabolism. Our study provides a basis for further investigation of the colonic adaptation by the microbiome in patients with LI and with other functional bowel disorders.

Disclosure

Nothing to disclose

References

1.Szilagyi A. et al. Differential impact of lactose/lactase phe-notype on colonic microflora. Canadian journal of gastroenterology 24, 373–379 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Misselwitz B., Butter M., Verbeke K., & Fox M. R. Update on lactose malabsorption and intolerance: pathogenesis, diagnosis and clinical management. Gut 68, 2080–2091 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Bonder M. J. et al. The effect of host genetics on the gut microbiome. Nature Genetics 48, 1407–1412 (2016). [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.168

OP178 The Dutch Microbiome Project: Defining The (Un)Healthy Gut Microbiome and Its Determining Factors in 9000 Metagenomes

R Gacesa ^1,^2,^✉, A Kurilshikov ², A Vich Vila ^1,², V Collij ^1,², S Hu ^1,², L Bolte ^1,², T Sinha ², L Chen ², S Andreu-Sánchez ^2,³, MAY Klaassen ^1,², VC Lenters ^4,⁵, F Imhann ^1,², D Jansen ^1,², J Gelderloos-Arends ², RCH Vermeulen ^4,⁵, S Sanna ^2,⁶, H Hansen ⁷, C Wijmenga ², J Yang-Fu ^2,³, A Zhernakova ², RK Weersma ¹; LifeLines Cohort Study

Introduction

While the gut microbiome has been associated with numerous diseases, the definition of a healthy or unhealthy microbiome remains elusive. It is also unclear how host genetics and phenotypic traits affect the gut microbiome, and how this interaction is influenced by the environment in health and disease.

Aims & Methods

The aims of the Dutch Microbiome Project are to establish gut microbiome patterns associated with health, to identify microbial pathobionts in disease, either specific or shared across diseases (such as gastrointestinal, metabolic and neurological disorders), and to determine how genetics, host characteristics, the exposome including early-life exposures, lifestyle, and environment shape healthy, and disease-associated microbiota.

We collected stool samples and used metagenomic sequencing to characterize composition and function of gut microbiota of ∼9,000 Dutch individuals (age range 8-84, 57.4% female), of which 2756 clustered in families. A total of 257 microbial taxa, 347 metabolic pathways, 47 virulence factors and 97 antibiotic resistance genes were correlated with 252 host and environmental factors including anthropometrics, diet, socioeconomics, physical and mental health, medication use, childhood and current exposure to pets, smoking, air pollutants, and urbanicity. Unsupervised clustering and machine-learning were used to define healthy microbiome and pathobi-onts associated with specific diseases as well as shared across chronic disorders. in addition, we quantified the variance in microbiome explained by phenotypes, and used family pedigrees and co-housing information to quantify the amount of variance explained by genetics, co-housing, and environment using the variance components model of heritability.

Results

We identified a total of 2950 associations between microbial features and physical and mental health (1240 associations with taxa and 1710 with pathways at FDR < 0.05). There were common patterns of associations across multiple diseases (including IBD, IBS, asthma, diabetes and mental disorders), which allowed us to define clusters of health- and disease-linked gut microbes and functions. The microbiome associated with diseases was found to be characterized by a significant increase in prevalence and abundance of opportunistic pathogens of genera Clos-tridium, Gordonibacter and Eggerthela, by a reduction in carbohydrate catabolism, synthesis of amino-acid and vitamins, and by an increase in synthesis of long-chain fatty acids. The healthy microbiome showed high abundances of butyrate-producing commensals from genera Alistipes, Roseburia, Faecalibacterium and Butyrivibrio. The variances of health- and disease-associated microbiota were found to be significantly explained (FDR < 0.05) by co-housing, host factors (such as age, sex, BMI and blood lipid profiles), and current as well as early-life exposome (including diet, socioeconomic, environment and exposure to pets, cigarette smoke and rural environments). Genetics was found to explain a minor part of the variance, with 26 microbial taxa significantly heritable (H² mean = 0.21, sd = 0.04, FDR < 0.05), none of which was an opportunistic pathogen.

Conclusion

This study demonstrates that unrelated diseases share common patterns which can be used to define core healthy and unhealthy gut microbiomes. The microbiome is primarily shaped by the environment and lifestyle, although the core healthy microbiome has a heritable component. Microbiome alterations through improving diet, lifestyle and the environment, and use of probiotics are a promising strategy for improvement of general health.

Disclosure

RKW received unrestricted grants from Takeda, Johnson and Johnson, Tramedico, Ferring, and acted as consultant for Takeda.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.169

OP179 Lifelines Next Pilot: Early Development of The Gut Bacteriome and Virome and Their Interactions

T Sinha ^1,^✉, S Garmaeva ¹, L Chen ^1,², A Gulyaeva ¹, S Jankipersadsing ¹, J Dekens ^1,³, J Spreckels ¹, R Gacesa ^1,⁴, A Kurilshikov ¹, C Wijmenga ¹, J Fu ^1,², AP Zhernakova ¹

Introduction

The establishment of a stable gut microbiome appears to play a crucial role in health, both during childhood and later in life. Disturbances in the intestinal microbiome during infancy have been associated with the development of metabolic and immune-mediated disorders. For example, infants with a gut microbial dysbiosis demonstrate a greater incidence of food allergies and gut complaints. Though recent studies have focused on the bacterial composition of the microbiome, the viral composition remains largely ignored. Furthermore, there is lack of knowledge regarding the influences of maternal and environmental factors on bacterial strains and their evolution.

Aims & Methods

We aimed to investigate the role of the virome and bacterial strains in the early development of the gut ecosystem and their relationship with infant gut health. For this, we used Lifelines NEXT, a prospective birth cohort in the Northern Netherlands aiming to include 1,500 pregnant mothers, their newborns and partners. Repeated sampling of maternal stool commences at 12 weeks of gestation and stool collection from mothers and babies occurs at 7 time points from birth until at least 12 months. Various other biomaterials like blood and breast milk, and extensive environmental and medical data are collected. For determining infant gut health, the ROME III criteria are used along with regular food frequency questionnaires (for both mother and infants). in this Lifelines NEXT pilot, we isolated DNA from total fecal microbiome and from viruslike particles from 30 mother-infant pairs with longitudinally collected samples (n=217) during pregnancy until three months after birth. We performed shotgun sequencing and investigated the abundances of ds-DNA viruses (mainly bacteriophages) along with their bacterial hosts. We further compared bacterial strains across and within mother-infant pairs.

Results

In a principal coordinate analysis, the gut bacteriome and virome of mothers and infants formed significantly different clusters (p< 0.001). Shannon diversity of both bacteriome and virome were significantly higher in mothers than in infants with infants showing greater bacterial diversity with increasing age. Mode of delivery, place of birth, birth weight, intake of antibiotics, probiotics and ranitidine, gestational age and maternal BMI did not significantly influence the infant gut bacterial composition at month 1, 2 and 3 after birth, although this analysis was limited by the relatively small sample size. We found evidence for strain transmission from maternal to infant gut for various species including Bifidobacterium bifidum. Furthermore, we found the differential acquisition of strains of Bifidobacterium breve and Escherichia coli in infants based on feeding pattern (breast milk versus formula-feeding). These strains were absent in the corresponding mothers, suggesting that the diet and the environment are important factors contributing to the infant gut microbiome development at the strain level.

Conclusion

Our results provide an insight into early bacterial and viral colonization of the infant gut. Our high-resolution data enables us to identify patterns of transmission from mothers as well as from the environment. We plan to expand this pilot to the entire Lifelines NEXT cohort and relate infant gut bacteriome and virome to a wide variety of phenotypes including infant functional gut problems. These results will pave the way for the prevention and treatment of infant gut disorders and future diseases by early life microbiome alterations through changes in maternal lifestyle, diet and probiotics intake.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.170

OP180 Disruption of The Gut Microbiome in Renal Transplant Recipients: Results From The Transplantlines Cohort and Biobank Study

JC Swarte ^1,^2,^✉, Y Li ^1,³, S Hu ^1,³, R Gacesa ^1,³, A Vich Vila ^1,³, RM Douwes ², H Blokzijl ¹, HJM Harmsen ⁴, SP Berger ², JSF Sanders ², EAM Festen ^1,³, AP Zhernakova ³, J Fu ^3,⁵, C Wijmenga ^1,³, SJL Bakker ², RK Weersma ¹

Introduction

Renal transplantation is life-changing in many aspects and is currently the only curative treatment for end-stage renal disease. All renal transplant recipients (RTR) use immunosuppressive medication and are frequently exposed to antibiotics which can change the gut mi-crobiome. It has previously been reported in a 16S sequencing study that the gut microbiome of end-stage renal disease patients is different from healthy controls (HC). However, little is known about the gut microbiome of RTR and the effect of immunosuppressive medication on the gut microbiome. Therefore, we used metagenomic sequencing data from 2154 fecal samples to characterize the gut microbiome of RTR.

Aims & Methods

We aimed to define the gut microbiome of end stage renal disease and to study the post-transplantation gut microbiome longitudinally. Furthermore, we aimed to characterize how the gut microbiome of RTR differs from HC and how the use of immunosuppressive medication is related to metabolic pathways. We collected 1,340 fecal samples from a total of 678 RTR patients from the Groningen TransplantLines study, including 369 longitudinal fecal samples of 76 RTR (pre-transplantation and at 3, 6 and 12 months post-transplantation) and >1 year post transplantation samples of 602 RTR. We also collected 1183 fecal samples from matched HC from the Dutch Microbiome Project. Shotgun metagenomic sequencing was performed to analyze gut microbial diversity, metabolic pathways, virulence factors and antibiotic resistance genes. We characterized the microbial change of RTR over time and compared RTR vs HC using linear mixed models including age, sex, BMI and use of PPI, use of antibiotics and use of laxatives. Furthermore, we analyzed which clinical characteristics explained variance in the gut microbiome of RTR using PERMANOVA.

Results

We first analyzed gut microbiome diversity longitudinally and found that microbial diversity significantly decreased post-transplantation compared to pre-transplantation (Wilcoxon-test, P< 0.001). Furthermore, RTR had a significantly lower gut microbial diversity and increased richness of virulence factors and antibiotic resistance genes compared to HC (Wilcoxon-test, P< 0.001). Longitudinally, there were 69 species and 90 metabolic pathways significantly changed over time compared to pre-transplantation (P_FDR< 0.10). Strikingly, case-control analysis including RTR >1 year post-transplantation and HC revealed 229 taxa, 289 pathways, 89 virulence factors and 116 antibiotic resistance genes that were significantly different (P_FDR< 0.10). A large shift in compositionality was observed, characterized by a higher prevalence of Proteobacteria and Actinobacteria and a lower prevalence of Firmicutes and Bacteroidetes (P_FDR< 0.10). The use of immunosuppressive medication significantly explained 3.5% of variance within the gut microbiome and was associated to multiple metabolic pathways (P_FDR< 0.10).

Conclusion

In this study, we characterized the gut microbiome of RTR using shotgun metagenomic sequencing data of 971 RTR and 1183 HC fecal samples and conclude that RTR suffer from dysbiosis. The gut microbiome of RTR is very different from HC and is characterized by a lower microbial diversity and increased richness of virulence factors and antibiotic resistance genes. Immunosuppressive medication changes the composition of the gut microbiome and it leaves a signature in the metabolic pathways of gut microbes in RTR. We show that the gut microbiome of RTR is permanently different from healthy controls and this could have far reaching implications for the outcome of renal transplantation.

Disclosure

RMD is supported by STW/NWO in a partnership with DSM, Animal Nutrition and Health. RKW received unrestricted grants from Takeda, Johnson and Johnson, Tramedico, Ferring, and acted as consultant for Takeda.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.171

OP182 Functional Genomic Screening During Somatic Cell Reprogramming Identifies Dkk3 As A Roadblock of Organ Regeneration

F Arnold ^1,^✉, J Gout ², L Perkhofer ¹, T Seufferlein ², P Hermann ¹, A Kleger ²

Introduction

Somatic cell reprogramming toward induced pluripotency can partly erase disease- and aging-associated phenotypes. Factors relevant to this process might also impact organ regeneration.

Aims & Methods

Here, we identified Dickkopf 3 (Dkk3) as a roadblock of organ regeneration using combined transcription-factor-induced reprogramming and RNA-interference techniques as a screening assay.

Results

Genetic loss of Dkk3 enhanced the generation of induced plu-ripotent stem cells but did not hinder de novo derivation of embryonic stem cells and three-germ layer differentiation. Organoid derivation assays from intestine, liver, and pancreas of Dkk3 knock out mice did not reveal colony formation differences, while in vivo injury induced DKK3 expression in wild type animals, respectively. Accordingly, Dkk3 null mice displayed less liver damage at time of regeneration upon CCl₄-induced acute liver failure. Similarly, recovery from experimental pancreatitis was accelerated albeit the extend of initial damage level was similar. Regeneration onset occurred in the acinar compartment accompanied by virtually abolished canonical Wnt-signaling in the Dkk3 null animals. Mechanistic analysis identified the downregulation of the Hedgehog (Hh) repressor Gli3 as a trigger of increased Hh-signaling upon Dkk3 loss. in contrast, treating knockout mice with a Hh inhibitor worsened improved pancreatic regeneration.

Conclusion

Collectively, our data reveal Dkk3 as a roadblock toward pluripotency and regeneration to establish a direct, previously unacknowledged link between DKK3-/canonical Wnt- and Hh-signaling.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.172

OP183 A Novel Basal Deltanp63 (P40)-Expressing Pancreatic Cell Type Undergoes Expansion and Reorganisation in Chronic Pancreatitis

S Martens ¹, M Van Bulck ¹, K Coolens ¹, F Esni ², G Leuckx ³, H Heimberg ³, L Bouwens ⁴, P Jacquemin ⁵, FM Spagnoli ^e, FX Real ⁷, P In't Veld ⁸, P Lesfevre ⁹, I Rooman ^1,^✉

Introduction

The basal-like/squamous molecular subtype of pancreatic ductal adenocarcinoma (PDAC) is characterised by expression of basal cell markers such as deltaNp63 (p40), an isoform of the p53 homologue p63. in other organs, tumours that display basal features arise from normal epi-thelia in which a subset of cells with a basal phenotype can be identified. These deltaNp63-expressing basal cells have stem cell capacity that can be activated in tissue regeneration and can be the cell of origin of cancer. Hitherto, in the normal pancreatic epithelium, such basal cells have not been reported.

Aims & Methods

We used multiplex immunofluorescence for protein detection and BaseScope for RNA detection and quantified the stainings with HALO software. We assessed (deltaN)p63 expression in histologically normal human pancreas tissue of subjects without a history of pancreatic disease (n=120, obtained for an islet transplant program) or with chronic pancreatitis (n=11).

In addition, we analysed the pancreas of normal mice (n=14) and mice with caerulein-induced acute (n=5) or chronic (n=5) pancreatitis. in addition to deltaNp63, other basal cell markers were assessed in a subset of these human and mouse samples.

We developed in house a method (see results) for sample clearing and 3D protein detection. Samples were imaged with light sheet fluorescence microscopy and analysed with Arivis software.

Results

We detected rare (approx. 1/8000) deltaNp63+ cells in the healthy adult human pancreas, but not in mouse pancreas. in the human samples, their frequency increased approximately 30-fold in chronic pancreatitis. The deltaNp63⁺ cells were located in or near ducts and were positive for cytokeratin (CK) 19 but lacked markers of differentiated pancreatic duct cells such as CA19.9 and lacked MUC6, a marker of pancreatic ductal glands.

Moreover, they had low or undetectable levels of SOX9 and exhibited cytoplasmic expression of HNF1b, in contrast to the strong nuclear expression of these transcription factors in the neighbouring duct cells. in addition, the deltaNp63+ cells co-expressed other typical markers of basal cells such as CK5, CK14 and S100A2. To image this rare basal cell type in 3D at single cell resolution, we developed a method, based on FLASH, CUBIC and ECi principles, that could be completed in two weeks. This method used formalin-fixed paraffin embedded archival human pancreatic specimens and preserved the native 3D pancreatic morphology and volume. We revealed that the deltaNp63+/ CK5+ cells were located as singular cells in the duct lining, between the basal lamina and the luminal duct cells, or as clusters branching from these ducts. Unlike in normal pancreas, in chronic pancreatitis samples these basal cells were organised in more abundant gland-like structures containing several layers around a larger lumen.

Conclusion

For the first time, we identified and imaged in 3D a novel cell population in the adult human pancreas, i.e. basal cells that increase upon chronic inflammation. The identification of a basal cell population in the pancreas raises important questions about their developmental origin, fate, and role in regeneration and disease, including the basal-type PDAC.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.173

OP184 Txnrd1 Modulates The Course of Acute Experimental Pancreatitis By Influencing Nfkb Activity

B Li ^1,^✉, M Neuhofer ¹, M Aichler ², M Brielmeier ³, R Schmid ¹, H Einwächter ¹

Introduction

Acute pancreatitis (AP) is an inflammatory disorder of the exocrine pancreas associated with tissue injury and necrosis (1). The disease causes major morbidity and mortality. The severity of AP can be mild, moderate, or severe, which depends on the extent of local injury in and around the pancreas, and more importantly systemic injury to remote organs (2). A more severe form of the disease seen in approximately 20% of all patients with AP is associated with significant local complications in the form of necrosis and often systemic injury due to systemic inflammation (3).

When a protective role of radical scavengers was found in an animal model of AP, it was postulated that oxidative stress might mediate a central step in the pathogenesis of AP. Since then, the role of free radicals in the development of AP has often been discussed.

In addition, many studies have pointed out that NF-kB activation plays a pivotal role in the early stages of AP, both in experimental models as well as in human AP. Previously, it was shown that deletion of P65/RelA, a member of NF-kB proteins, leads to a dramatically worsened phenotype in experimental AP (4).

The main cellular anti-oxidative defense systes are glutathione (GSH), thioredoxin (Trx) and peroxiredoxin (Prx) (5). GSH and Trx systems are thiol-dependent redox systems, they play complementary and some overlapping roles. Thioredoxin reductase (Txnrd) is the only enzyme known to catalyze the change from inactive oxidized Thioredoxin to the active reduced form, Txnrd1 is the cytoplasmic isoenzyme of thioredoxin reduc-tases. Here, we set out to investigate the role of cytosolic Txnrd1 in the pancreas and more specifically, during acute pancreatitis.

Aims & Methods

Pancreas-specific deletion of Txnrd1 was achieved by crossing Ptf1a-Creex1* mice to mice carrying floxed Txnrd1 alleles. The deletion was validated on the DNA, RNA and protein level. Mice were analyzed under basal conditions at 12 weeks and 24 weeks of age. For gene expression analysis, RNA from unstimulated pancreata of 4 mice per genotype was labeled and hybridized to mouse expression gene chip arrays (Affymetrix Mouse Genome 430A 2.0 Array). For Gene Set Enrichment Analysis (GSEA), GSEA software and all genesets were provided by the Broad Institute of MIT and Harvard (6). For induction of pancreatitis, animals underwent 8 injections of cerulein 50 μg/kg at hourly intervals. Pancreatic tissue was harvested at 0 h, 3 h, 8 h and 24 h after induction of AP. Further analyses included western blot, histology, immunohistochemistry and RNA-Seq.

Results

Mice were born at the expected mendelian rations and did not show any phenotypical abnormalities. Surprisingly, gene expression analysis revealed a significant downregulation of NFkB related genesets in Txnrd1^Δpanc compared to control mice. After experimental induction of pancreatitis, Txnrd1^Δpanc mice showed increased edema, infiltration and massive necrosis compared to control mice.

Conclusion

Redox regulation of NFkB has been proposed for several decades. There has been a multitude of research on redox regulation of NF-kB, but so far, the exact mechanisms by which this regulation occurs, are unclear. in addition, most findings connecting redox regulation and NFkB activity were obtained in vitro. Our model shows a worse pancreatitis phenotype together with reduced NFkB activity. We therefore speculate that redox regulation of NFkB might be relevant in vivo. However, the exact mechanism by which Txnrd1 modulates NFkB activity has yet to be identified.

Disclosure

Nothing to disclose

References

1.Vege S.S., DiMagno M.J., Forsmark C.E., Martel M., Barkun A.N. Initial Medical Treatment of Acute Pancreatitis: American Gastroentero-logical Association Institute Technical Review. Gastroenterology. 2018. Mar; 154(4): 1103–39. [DOI] [PubMed] [Google Scholar]
2.Banks P.A., Bollen T.L., Dervenis C., Gooszen H.G., Johnson C.D., Sarr M.G. et al. Classification of acute pancreatitis–2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013. Jan; 62(1): 102–11. [DOI] [PubMed] [Google Scholar]
3.Garg P.K., Madan K., Pande G.K., Khanna S., Sathyanarayan G., Bohidar N.P. et al. Association of extent and infection of pancreatic necrosis with organ failure and death in acute necrotizing pancreatitis. Clin Gastroenterol Hepatol. 2005. Feb; 3(2): 159–66. [DOI] [PubMed] [Google Scholar]
4.Algul H., Treiber M., Lesina M., Nakhai H., Saur D., Geisler F. et al. Pancreas-specific RelA/p65 truncation increases susceptibility of acini to inflammation-associated cell death following cerulein pancreatitis. J Clin Invest. 2007. Jun; 117(6): 1490–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Sies H., Berndt C., Jones D.P. Oxidative Stress. Annu Rev Biochem. 2017. Jun 20; 86: 715–48. [DOI] [PubMed] [Google Scholar]
6.Subramanian A., Tamayo P., Moo-tha V.K., Mukherjee S., Ebert B.L., Gillette M.A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005. Oct 25; 102(43): 15545–50. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.174

OP185 The Role of Orai1 Ca²+ Channel in The Pancreatic Ductal Epithel Cells Under Pathophysological Conditions

P Pallagi ¹, M Görög ², A Varga ³, T Madacsy ¹, N Papp ¹, A Balazs ⁴, B Tél ⁵, P Hegyi ⁶, J Maléth ^7,^✉

Introduction

In non-excitable epithelial cells, Ca²+ signaling is one of the major signaling pathways determining crucial functions, such as ion and fluid secretion. On the other hand, sustained intracellular Ca²⁺ overload, which is largely determined by the extracellular Ca²+ influx via Orai1 channels, is a hallmark of acute pancreatitis. The impaired fluid and HCO₃^- secretion of pancreatic ductal epithelial cells (PDEC) increases the severity of AP. Our previous works showed that the toxic Ca2+ elevation has a major contribution to the damage of the ductal cells, however it's development and the role of Orai1 in the process is not understood yet.

Aims & Methods

We aimed to clarify the relevance of Orai1 in PDEC. Interlobular pancreatic ductal fragments were isolated from FVB/n mice and human pancreatic organoid cultures were used for RT-PCR, immu-nostaining and fluorescent Cl^-, Ca²+ and pH measurements. Experimental AP was induced by cerulein (7X50 μg/bwkg), or ethanol/fatty acid (1,75 g/ kg ethanol and 750 mg/kg fatty acid) injection or retrograde infusion of 4% sodium-taurocholate (Na-TC) into the pancreatic duct. Orai1 inhibitor CM-C (20 mg/kg) was administrated after the induction of AP.

Results

Orai1 is expressed on the apical membrane of mouse and human PDEC and mediates Ca²+ influx upon physiological ER Ca²+ depletion. Inhibition of Orai1 prevented bile acid (BA) induced toxic Ca²+ overload, prevented BA induced inhibition of HCO₃^- secretion and reduced the forskolin stimulated HCO₃^- secretion. Orai1 inhibition decreased the forskolin-stim-ulated HCO₃^- secretion and reduced the activity of cystic fibrosis transmembrane conductance (CFTR) as suggested by the Cl^- extrusion upon extracellular Cl^- removal. in in vivo experiments, post-treatment with the Orai1 inhibitor CM-C significantly reduced the severity of cerulein-, BA and alcohol-induced AP and abolished the decrease in in vivo pancreatic fluid secretion during AP.

Conclusion

Our results suggest that the Orai1 mediated Ca²+ entry plays a major role in pancreatic ductal physiology and pathophysiology. The inhibition of Orai1 potentially restores the function of pancreatic ductal cells and improves the outcome of AP.

Disclosure

There is no conflict of interest.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.175

OP186 Prospective Association Between Ultra-Processed Food Consumption and Risk of Non-Alcoholic Fatty Liver Disease: A Nationwide Long-Term Follow Up Study

YI Choi ^1,^✉

Introduction

Even consumption of ultra-processed food, one of food type which have undergone significant food processing, has been associated with the all cause of mortality, there is limited data on its relation with the risk for non-alcoholic fatty liver disease (NAFLD).

Aims & Methods

We evaluated the prospective association of ultra-processed food consumption and the risk of NAFLD using nationally representative long term follow up study.

We used the data from the Korean_health examinee study (HEXA study) obtained during the periods from 2004-2013(baseline) and 2012-2017(fol-low up study). Ultra-processed food intake was assessed through in-per-son interviews with a validated food-frequency and amount questionnaire. NALFD was defined by fatty liver index (FLI) (>60). To reveal the association of ultra-processed food intake at baseline and the risk of newly development of NAFLD and NAFLD associated fibrosis during follow up period, univariate and multivariate analysis were done.

Results

Of 177,231 participants (40-70 year), after exclusion of participants with excess alcohol intake, with baseline FLI >60, history of chronic liver disease, or without follow up data, we finally enrolled 40,224 participants. During median 7.1 year of period, total of 5,160 incidental cases with NAFLD were developed. Ultra-processed food consumption was associated with the risk of NAFLD (hazard ratio(HR)= 1.49 [1.21-1.98], p< 0.001) in univariate analysis. The Multivariate adjusted HR of the risk of NAFLD for the upper intake quintile of ultra-processed food consumption compared with those of the lowest quintile were 1.41 (95% confidence interval: 1.10-1.81) adjusting for total calorie intake per day, total protein intake per day, total lipid intake per day, total carbohydrate intake, drink habit, smoking status, physical activity, and stress perception

Conclusion

In this nationwide population based long-term follow up study, we revealed the ultra-processed food intake had adverse effect for the risk of the NAFLD. Further studies are needed to find complex mechanisms underlying the association between ultra-processed food intake and adverse health outcomes.

Disclosure

This work was supported by the Gachon University Gil Medical Center (Grant number: FRD2019-14). However there is not any conflict of interest in this study.

References

1.Lawrence M.A., Baker P.I. Ultra-processed food and adverse health outcomes. BMJ 2019; 365: l2289. [DOI] [PubMed] [Google Scholar]
2.Gibney M.J., Forde C.G., Mullally D., Gibney E.R. Ultra-processed foods in human health: a critical appraisal. Am J Clin Nutr 2017; 106: 717–724. [DOI] [PubMed] [Google Scholar]
3.Monteiro C.A., Cannon G., Moubarac J.C., Levy R.B., Louzada M.L.C., Jaime P.C. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutr 2018; 21: 5–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Srour B., Fezeu L.K., Kesse-Guyot E., Alles B., Mejean C., Andrianasolo R.M., Chazelas E. et al. Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Sante). Bmj 2019; 365: l1451. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.176

OP187 Study of The Necessity of Cholecystectomy After Treatment of Choledocholithiasis For Elderly Patients

R Kodama ^1,^✉, Y Koh ¹, H Midorikawa ¹, Y Yokota ¹, H Saegusa ¹, H Ushimaru ¹

Introduction

Patients with gallbladder stones have been shown to have a high incidence of acute cholecystitis after endoscopically treating bile duct stones, and cholecystectomy is often recommended after treatment for common bile duct stones.

However, there are many cases, especially in the elderly, in which it is difficult to decide whether to recommend cholecystectomy from the viewpoint of the underlying disease and prognosis.

Aims & Methods

To clarify the necessity of cholecystectomy for gallstones after treatment of choledocholithiasis in elderly patients over 80 years of age. of the 314 patients with ERCP for common bile duct stones in our hospital during the 5 years from April 2011 to March 2016, 197 cases had gallstones.

We retrospectively compared 106 cases aged 79 years or younger (young group) and 91 cases aged 80 years or older (elderly group). Furthermore, we studied the rate of biliary troubles and mortality in each group according to the presence or absence of cholecystectomy.

Results

The patient background was different between the young and elderly groups in terms of sex ratio and mean common bile duct diameter of 67:39, 10.4 mm and 40:51, 11.5 mm, respectively. The mean observation period was not different in both groups. Cholecystectomy was performed in 51% of the young group and 9% of the elderly group, and the rate of surgery was significantly lower in the elderly group. Postoperative biliary troubles occurred in 18% of the young group and 30% of the elderly group, and tended to be more frequent in the elderly group. According to the Kaplan-Meier analysis, cholecystectomy significantly reduced the risk of biliary problems in the young group, but there was no significant difference in mortality. There were no significant differences in biliary trouble and death in the elderly group.

Conclusion

The presence or absence of cholecystectomy does not affect biliary tract problems or death in elderly people over 80 years old, and follow-up may be appropriate without surgery. The number of cases is small and further studies are needed.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.177

OP188 Hepatocellular Carcinoma Surveillance At A District General Hospital in The Uk - Can The Surveillance Interval Be Increased During The Covid- 19 Pandemic?

A Halim ¹, M Olsen ^1,^✉, P Youd ¹

Introduction

Patients with chronic liver disease (CLD) have an increased risk of developing hepatocellular carcinoma (HCC) in comparison to the normal population. Current EASL guidance advises patients with CLD should undergo six- monthly HCC surveillance with an ultrasound scan (USS) and serum alpha- fetoprotein (AFP) measurement. As a result of the global COVID- 19 pandemic, most medical outpatient services in the United Kingdom (UK) have indefinitely been suspended or have significantly reduced patient capacity to minimise the risk of COVID-19 transmission. Many patients are also postponing their planned hospital appointments now for fear of contracting COVID- 19.

Aims & Methods

The objective of this retrospective study is to assess the impact of delaying HCC surveillance in patients with CLD and what could be considered an acceptable increase in surveillance interval during an extraordinary circumstance such as the global COVID- 19 pandemic. A list of patients with CLD that were admitted to hospital between 2017-2019 was collated by the hospital's coding department. The key words used for the search were ‘alcohol liver disease’, ‘NASH’, ‘chronic liver disease’, ‘hepatitis’ and ‘cirrhosis’. Data was collected by scrutinising the patients’ clinic letters, pathology results and radiology reports that were available through the hospital's electronic database.

Results

Our search generated an initial list of 196 patients. Application of the inclusion criteria led to the exclusion of 128 patients from the study. Data for the study was collected from a total of 68 patients with CLD. We found the average interval for surveillance USS was 7.5 months. Overall, 40 patients (59%) had their USS performed later then the recommended window of surveillance, ranging from seven months up to 18 months. in 72% (49/68) of patients, no change was detected on USS while 28% (19/68) showed progression of CLD. 12% (8/68) were found to have a new lesion and were referred for urgent computed tomography (CT). None of these patients were found to have HCC on CT.

One patient was referred for urgent CT after developing worsening ascites and found to have an early HCC despite having regular 6 monthly USS which suggested stable CLD.

One patient who missed their six month follow-up and re- attended clinic after a 12- month interval were noted to have an elevated AFP and found to have stage 4 HCC on subsequent CT.

Conclusion

This study has demonstrated that encouraging patients with CLD to extend their surveillance USS would overall be detrimental, even against the risk of COVID-19 infection.

A significant proportion of patients in this relatively small study were noted to have changes on their USS over a six- month interval. Therefore, regular USS assessment remains not only an important tool in screening for HCC but also for screening for other potential complications of CLD, such as ascites and portal vein thrombosis.

Urgent cross- sectional imaging should be sought where patients manifest clinical signs of disease progression even if their surveillance USS were unremarkable.

Where patients choose to postpone their surveillance USS during the COVID-19 pandemic they must be informed of the risks of doing so.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.178

OP189 Abp (Albumin, Bilirubin and Platelets) Criteria To Predict The Absence of High-Risk Esophageal Varices: Validation in A Portuguese Tertiary Hospital

MM Estevinho ^1,^✉, C Fernandes ¹, M Sousa ¹, JC Silva ¹, AC Gomes ¹, E Afecto ¹, J Correia ¹, J Carvalho ¹; Group of Hepatology I.M.D.²

Introduction

In order to avoid screening upper endoscopy Baveno consensus defined criteria to predict the absence of high-risk esophageal varices (HR-EV) in chronic compensated liver disease. Nevertheless, considering that transient elastography it is not widely available, the ABP criteria¹ was built as an alternative.

Aims & Methods

This study aimed to validate ABP criteria (albumin > 4.0g/dL, bilirubin < 22μmol/L and platelet > 114000/μL) in our center and to compare it with Baveno VI criteria. A retrospective study of all patients who performed an upper endoscopy between 2010 and February 2020 was performed. Inclusion criteria were patients with Child-Pugh A cirrhosis, with transient elastography and laboratory tests performed within a maximum of 12 months of the endoscopy. The sensitivity (true rate of absence of HR-EV), specificity, positive and negative predictive values (PPV and NPV), Cohen's kappa coefficient (K) and area under the receiver operating characteristic curve (AUROC) were calculated.

Results

A total of 146 patients (76.7% male, mean age 63.8 years) were included; high risk varices (large or small with red wales) were present in 19.9%.

[Characteristics of the study cohort (n=146)]

Age (years) - mean ± SD	63.8 ± 10.7
Sex (masculine) - n (%)	112 (76.7)
Etiology - n (%)
Viral (HBV or HVC)	48 (32.9)
Alcohol	42 (28.8)
Mixed etiology	31 (21.2)
Nonalcoholic fatty liver disease	20 (13.7)
Liver stiffness (kPa) - mean ± SD	26.29 (16.3)
Esophageal varices - n (%)
No	81 (55.5)
Yes	65 (44.5)
Large	19 (13.1)
High risk	29 (19.9)

Open in a new tab

The sensitivity and specificity of the ABP criteria for the absence of HR-EV were 99.1% (95% confidence interval [CI] 95.33-99.9) and 44.8% (95%CI 26.5-64.3%); the NPV was 99.6% and the PPV was 31.6%. Regarding Baveno VI criteria, the sensitivity and specificity were 96.6% (95%CI 82.2-99.9) and 64.1% (95%CI 54.7-72.8), while NPV and PPV were 98.7% and 40.0%. The AUROC for Baveno VI was slightly higher than that obtained for ABP: 0.803 versus 0.720. The agreement between the two scores was fair (62.2%, κ=0.2). ABP criteria selected more patients than Baveno VI criteria (132 versus 76) to avoid screening endoscopy and only one of those presented with HR-EV. ABP revealed a better concordance with endoscopy (0.546 versus 0.396 of Baveno VI).

Conclusion

ABP, a score developed in a multi-ethnic Asian cohort and easily applicable in any clinical setting, revealed a great ability to identify patients with compensated cirrhosis that can safely avoid screening endoscopy in the Portuguese population.

Disclosure

Nothing to disclose

References

1.Kew G. S. et al. (2019). Identifying Patients with Cirrhosis Who Might Avoid Screening Endoscopy Based on Serum Albumin and Bilirubin and Platelet Counts. Clinical Gastroenterology and Hepatology, S1542-3565(19) 31270–4. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.179

OP190 Circulating Microrna-21 and Microrna-122: Prognosis Prediction and Correlation with Hif-1α in Hepatocellular Carcinoma Patients Treated with Transarterial Chemoembolization

F Pelizzaro ^1,^✉, R Cardin ¹, C Carlotto ¹, M Minotto ¹, A Sartori ¹, A Imondi ¹, B Penzo ¹, C Aliberti ^2,³, A Ponzoni ², A Vitale ⁴, U Cillo ⁴, F Farinati ¹

Introduction

MicroRNA-21 (miR-21) and microRNA-122 (miR-122) play an important role in hepatocellular carcinoma (HCC) progression and have been identified as promising circulating biomarkers for this tumor.

Aims & Methods

We aimed at evaluating these two miRNAs in HCC patients treated with transarterial chemoembolization (TACE) as predictors of progression-free survival (PFS) and investigating their correlation with TACE-induced ischemia as assessed by the determination of Hypoxia inducible factor-1α (HIF-1α) serum levels.

For these purposes, 12 healthy volunteers, 28 cirrhotics and 54 HCC patients (tested both before and four weeks after TACE, at the time of the imaging control) were prospectively enrolled. The miRNA level before TACE and the miRNA ratio (miRNA after/before TACE) were evaluated as potential progression-free survival (PFS) predictors, with cut-offs established with the ROC curve method and the Youden J test. The Spearman's rank correlation coefficient was used to correlate miRNAs with HIF-1α. miR-21 and miR-122 levels were measured by qRT-PCR on whole blood and expressed as 2^-MCt, while HIF-1α was assessed in serum by an ELISA test.

Results

Cirrhotics had the highest level of miR-21, compared both to controls (p=0.009) and HCC patients (p=0.047). A statistically significant drop of miR-21 levels after TACE was observed in HCC patients (p=0.03). miR-122 progressively increased in the three groups, with the highest levels in HCC patients as compared to controls (p=0.02) and cirrhotics (p=0.04). Patients with an HCC developed on a viral chronic liver disease had higher levels of miR-122 in comparison to patients with alternative etiologies (p=0.006). miR-21 levels were significantly correlated with HIF-1α both before (r=0.34; p=0.045) and after TACE (r=0.35; p=0.035). Patients with miR-21 ratio (after/before TACE) and miR-122 below their respective cutoffs had a statistically significant longer PFS. in particular, patients with miR-21 ratio below the cut-off (1.64 fold change) had a significantly longer PFS compared to those with higher levels (median PFS 5.6 months vs. 1.4 months, respectively; p=0.0002). A significantly longer PFS was also demonstrated in patients with miR-122 below the identified cut-off (10.22 fold change) as compared to patients with miRNA levels above this threshold (5.6 months vs. 2.5 months; p=0.02). The multivariate Cox regression analysis identified miR-21 ratio, AFP levels and radiological response as independent prognostic predictors. The combined assessment of AFP and miR-21 ratio identified early progressors among patients with complete or partial radiological response, as those with one or both markers above their respective cut-offs had a statistically significant shorter PFS (p=0.001).

Conclusion

miR-21 ratio and miR-122 propose themselves as useful prognostic biomarkers in TACE-treated HCC patients. in particular, miR-21 ratio is associated to PFS at univariate and multivariate analysis and identifies (together with AFP) early progressors among the patients achieving a radiological response to treatment. The finding of a link between circulating miR-21 and HIF-1α in HCC indicate a potential role of miR-21 in modulating the angiogenetic process.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.180

OP191 Utility of Routine Esophageal Biopsies in Patients with Refractory Reflux Symptoms

Nijhuis RAB Oude ^1,^✉, WL Curvers ², M van der Ende ², T Herregods ¹, JM Schuitenmaker ¹, A Smout ¹, A Bredenoord ¹

Introduction

In the most recent update of the Rome criteria, it is recommended to obtain esophageal biopsy samples in all patients with refractory reflux symptoms, in order to rule out eosinophilic esophagitis (EoE). However, obtaining biopsies in every suspected reflux patient would indicate an enormous volume of biopsies, while the prevalence of EoE in this patient group is thought to be extremely low. Moreover, biopsies will not help in the differentiation of functional heartburn from reflux disease.

Aims & Methods

The main objective of this study was to assess the additional diagnostic yield of esophageal biopsy sampling in patients with PPI-refractory reflux symptoms. Consecutive patients with reflux symptoms refractory to standard dose PPI therapy were prospectively enrolled. All patients underwent upper endoscopy with biopsy sampling in accordance with current guidelines. Endoscopic features suggestive of EoE and other endoscopic abnormalities were routinely recorded by the endoscopist. The Reflux Disease Questionnaire, the Straumann Dysphagia Index, and a complete medical history were carried out to assess comorbidities and esophageal symptoms prior to endoscopic evaluation. EoE was defined as the presence of >15 eosinophils per microscopic high-power field in at least one esophageal biopsy specimen, accompanied by symptoms of esophageal dysfunction.

Results

Of the 301 included patients, fourteen patients met the clinico-pathological diagnostic definition of EoE, corresponding to a prevalence of 4.7%. All patients with refractory reflux symptoms that eventually ended up with a final diagnosis of EoE, presented with concomitant symptoms of dysphagia, versus 49% of the patients without EoE (p< 0.001). in line with this, symptoms of food impaction, history of bolus impaction that required endoscopic removal, atopic background and presence of endoscopic features compatible with EoE were more frequently identified in EoE patients (all p< 0.01). of these characteristics, dysphagia had the highest sensitivity (100%), whereas atopic background (OR 23.7; 95%-CI 5.2-108.8), endoscopic features compatible with EoE (OR 13.0; 95%-CI 4.1-40.9) and a history of endoscopic dislodgement of food bolus impaction (OR 12.8; 95%-CI 2.8-57.9) were identified as the factors with the strongest association and the highest diagnostic accuracy for EoE diagnosis (all p< 0.001). Diagnostic yield in patients lacking symptoms of dysphagia or endoscopic features was negligible (0% and 1.9%, respectively).

Conclusion

Routine esophageal biopsy sampling in patients with refractory reflux symptoms has a very low diagnostic yield (4.7%). Our data shows that biopsies should only be obtained in those patients that have dysphagia accompanied by one or more specific clinical characteristics suggestive of EoE. Diagnostic yield of biopsy sampling in patients lacking typical EoE hallmarks was negligible and should therefore be omitted.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.181

OP192 Accuracy of The Ppi Test For Gastro-Esophageal Reflux Disease Diagnosis: Comparison with Objective Wireless Ph Study Data

R-I Rusu ^1,^✉, P Woodland ², SS Zeki ¹, J Jafari ¹, M Fox ³, T Wong ¹

Introduction

Proton pump inhibitors (PPIs) are widely prescribed for gastro-esophageal reflux disease (GERD) symptoms. The “PPI test” is frequently used in lieu of formal testing. A placebo response in patients without true GERD will lead to ongoing, unnecessary use, and an inadequate response to PPI in patients with true GERD may result in missed opportunities for alternate treatment strategies. It has been shown previously that, with 48hr pH monitoring, a 2-week PPI trial has limited accuracy for GERD diagnosis. However, it is possible that restricting to 48 hrs pH monitoring could “miss” true GERD diagnoses, and a 2-week PPI trial may not be long enough for adequate symptom relief.

Aims & Methods

We aimed to assess the accuracy of response to an 8-week PPI trial in diagnosis of GERD when using 96hr pH monitoring as gold standard.

We first established 96hr normal values in a cohort of 40 asymptomatic healthy volunteers

(age 28±9 years, 72% F). Upper limits of normal were defined as 95^th percentile values for mean total acid exposure, worst day acid exposure, and mean DeMeester score.

We studied 86 patients (age 48.4±13. years, 71.7% F) for testing of troublesome heartburn symptoms. All patients completed a RESQ-7 questionnaire off PPI, then had wireless pH capsule investigation for 96hrs. Total acid exposure%, worst day acid exposure% and mean DeMeester score were recorded and compared to our normal values. After completion of the investigation, all patients started PPI, at least standard dose, for 8 weeks. At 8 weeks the RESQ-7 was repeated. Percentage improvement in heartburn intensity at 8 weeks compared to baseline was measured.

A successful PPI response was defined as >50% improvement in heartburn intensity at 8 weeks compared to baseline.

Results

Normal value cut-offs were determined as 4.8% for mean acid exposure, 7.4% for worst day acid exposure, and 9.2 for mean DeMeester score. There was no difference in % heartburn improvement after 8 weeks PPI between patients with pathological and physiological acid exposure whether pathological reflux was determined by mean acid exposure (heartburn improvement 15.9%+65.6 vs. -1.8%+79.2, mean+s.d, p=0.1), worst day acid exposure (13.3%+72.1 vs. 1.5%+71.9, p=0.3), or by mean DeMeester score (12.5%+74.1 vs. 3.8%+63.2, p=0.4). Successful PPI response had only a 27.0% positive predictive value for GORD, and a 81.8% negative predictive value.

Conclusion

Wireless pH study parameters in a large group of patients with heartburn referred for reflux testing show that PPI response is inadequate to make a diagnosis of reflux disease. Our data show that a large number of patients referred despite PPI response do not have reflux disease. This suggests widespread over-prescribing of PPI, and testing can inform a need to wean off the medication. On the other hand, a significant proportion of patients do not have PPI response yet have pathological reflux on testing. These patients could be offered alternate management strategies.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.182

OP193 Effectiveness of A Novel Medical Device Containing Hyaluronic Acid, Alginate and Camellia Sinensis Extract, For The Treatment of Laryngopharyngeal Reflux Symptoms

R De Bastiani ¹, F Ingravalle ², C Tosetti ¹, E Benedetto ¹, G Casella ¹, M Zamparella ¹, G Patrizia ¹, EV Savarino ^3,^✉

Introduction

Association of gastroesophageal reflux disease (GERD) with extraesophageal manifestations, such as chronic cough and laryngitis, are commonly reported in clinical practice. Laryngopharyngeal Reflux (LPR) are usually treated with proton pump inhibitors (PPIs) as firstline approach. Recently, a novel medical device, an oral spray containing Camellia Sinensis, hyaluronic acid and alginate, has been introduced on the Italian market for the treatment of LPR. Data on its effectiveness in clinical practice are not available yet.

Aims & Methods

We aimed to investigate the effectiveness of the oral spray in primary care patients presenting with chronic laryngeal symptoms suggestive of LPR. in a prospective multicentre, open label, randomized, controlled, pilot study, consecutive patients presenting with chronic laryngeal symptoms were enrolled by primary care physicians. All subjects completed the validated reflux symptom index (RSI) questionnaire and in case of a score >13, a symptom-based diagnosis of LPR was made. Thereafter, all individuals with a diagnosis of LPR were randomized to receive PPI treatment (double-dose) or PPI treatment (double-dose) plus the oral spray (2 puff after breakfast, lunch and dinner and at night-time) for 8 weeks. During the treatment period, all the patients repeated the RSI at week 4 and 8. in case of an RSI < 13 the treatment was considered effective.

Results

Between March and August 2018, a total of 55 patients were enrolled. Thirty-four patients (13 males / 21 females, mean age ± SD 54.41 ± 14.14 year) were randomized to receive PPI therapy plus the oral spray, whereas twenty-one controls (6 males/15 females, 59.43±11.57) received PPI treatment only. Compared to controls, patients under PPI therapy plus the oral spray reported less frequently throat clearing (p=0.003), excess throat mucus or postnasal drip (p=0.002), troublesome or annoying cough (p=0.034), and sensation of something sticking or a lump in your throat (p=0.099). Details are reported in Table 1. Overall, 16 out of 17 patients under PPI therapy plus the oral spray achieved a remission status (RSI < 13), whereas only 7 out of 14 controls reported similar effect (p=0.017). No safety issues have been highlighted.

[Table of RSI Symptoms]

	T0		T1		T2		P-value (T-test)
Table of Symptoms	Case	Controls	Case	Controls	Case	Controls	T0	T1	T2
Hoarseness or voice problem	2.47 ± 1.54	3 ± 1.3	1.68 ± 1.43	2.26 ± 1.48	1.23 ± 1.21	1.78 ± 1.21	0.211	0.219	0.232
Throat clearing	3.59 ± 1.11	3.4 ± 0.97	2.14 ± 1.39	2.26 ± 1.29	1.47 ± 0.97	2.79 ± 1.26	0.540	0.770	0.003
Excess throat mucus or postnasal drip	3.21 ± 1.58	3.25 ± 1.51	2.04 ± 1.61	2.63 ± 1.42	1.29 ± 1.12	2.86 ± 1.41	0.921	0.238	0.002
Difficult swallowing	0.97 ± 1.36	1.3 ± 1.45	0.45 ± 0.94	0.78 ± 1.15	0.29 ± 0.66	0.71 ± 1.38	0.414	0.323	0.293
Coughing after eating or lying down	2.57 ± 1.63	1.35 ± 1.52	0.95 ± 1.3	0.57 ± 0.99	0.7 ± 1.27	1.43 ± 1.63	0.010	0.321	0.191
Breathing difficulties or choking episodes	1.76 ± 1.61	1.45 ± 1.5	0.68 ± 1.06	0.68 ± 0.92	0.23 ± 0.42	0.86 ± 1.19	0.501	0.994	0.061
Troublesome or annoying cough	2.76 ± 1.71	2.3 ± 1.64	1.3 ± 1.45	1.68 ± 1.45	0.76 ± 1.06	1.79 ± 1.42	0.342	0.426	0.034
Sensations of something sticking or a lump in your throat	2.76 ± 1.79	2.6 ± 1.91	1.77 ± 1.78	1.58 ± 1.63	0.94 ± 1.05	1.86 ± 1.81	0.756	0.726	0.099
Heartburn. chest pain. indigestion. or stomach acid coming up	2.21 ± 1.66	2.9 ± 1.37	0.72 ± 1.09	1.79 ± 1.47	0.29 ± 0.75	0.79 ± 1.08	0.127	0.013	0.160
Total RSI	21.97 ± 6.14	21.05 ± 5.11	11.36 ± 6.63	14.16 ± 6.62	7.06 ± 3.75	14.57 ± 7.86	0.581	0.196	0.002

Open in a new tab

Conclusion

Our pilot study supports the effectiveness and safety of the oral spray containing Camellia Sinensis, hyaluronic acid and alginate as add-on therapy to PPIs for the treatment of LPR. Further larger and controlled studies are required to confirm our preliminary results.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.183

OP194 Low-Fodmaps Diet For The Treatment of Refractory Gastroesophageal Reflux Disease. A Randomized Controlled Trial

B Vauquelin ¹, P Riviere ¹, E Rolland ¹, C Melchior ², S Roman ³, S Bruley Des Varannes ⁴, F Zerbib ^1,^✉, G Gourcerol ⁵, S Sacher-Huvelin ⁶

Introduction

Dietary advices are often proposed to patients with gastro-esophageal reflux disease (GERD). The low-FODMAPs diet (Fermentable Oligo-, Di-, Mono-saccharides and Polyols) improves lower gastro-intestinal symptoms. Colonic fermentation of alimentary carbohydrates has been shown to impact oeso-gastric motility and the occurrence of reflux episodes., The aim of this study was to evaluate the efficacy of a low-FOD-MAPs diet in patients with proton pump inhibitors (PPI) refractory GERD.

Aims & Methods

This was a multi-center, randomized, open-label study comparing the efficacy of a 4-week low-FODMAPs diet (LFD) to the usual dietary advice (standard diet, SD) for GERD. Patients with symptomatic PPI-refractory GERD, defined by a Reflux Disease Questionnaire (RDQ) score >3 and abnormal pH-impedance monitoring on PPIs were included. The primary endpoint was the percentage of responder patients (RDQ ≤ 3) at the end of the 4-week period. The secondary endpoints were to evaluate the effect of the diet on pH-impedance parameters, and associated functional symptoms (dyspepsia, irritable bowel syndrome), using different scores such as Gastrointestinal Quality of Life Index (GIQLI) and IBS-SSS score (irritable bowel syndrome severity scoring system).

Results

Thirty-one patients (55% women, median age 45 years) were included in 4 French centers, 16 randomized in the LFD group, 15 in the SD group. There was no significant difference at inclusion between the 2 groups in terms of age, sex, weight, consumption of FODMAPs per day, pH-impedance parameters, RDQ, IBS-SSS, and GIQLI scores. Adherence to the assigned diet was good, with a significant difference in the amount of FODMAPS consumed per day between the LFD (2.5 g) and the SD group (13 g), (p< 0.001). The caloric intake decreased in both groups (from 1600 to 1400 on average, p< 0.001) without significant difference between the two groups. There was no difference in response rates (RDQ score < 3) between the LFD (6/16, 37.5%) and SD (3/15, 20%) groups (p = 0.20). Total RDQ score and dyspepsia subscore decreased significantly over time in both groups (p=0,002), with no difference according to the assigned diet group (p=0,85). IBS-SSS decreased in both groups significantly for the FODMAPS group only (p=0.04). There was no difference in GIQLI score. There was no significant difference in the pH-impedance parameters between LFD and SD.

Conclusion

A low FODMAPs diet did not show any benefit over a standard diet to improve symptoms in patients with PPI-refractory GERD.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.184

OP195 Pneumatic Dilation For Persistent Dysphagia After Antireflux Surgery, A Multicenter Randomized Sham-Controlled Clinical Trial

JM Schuitenmaker ^1,^✉, F van Hoeij ¹, MP Schijven ², J Tack ³, JM Conchillo ⁴, EJ Hazebroek ⁵, A Smout ¹, A Bredenoord ¹

Introduction

Laparoscopic fundoplication is the most effective treatment for proton pump inhibitor-refractory gastro-esophageal reflux disease (GERD). Postoperative dysphagia, a common side effect after fundoplication, is usually self-limiting and generally disappears after 2-6 weeks. in 3.0-25.6% of patients however, dysphagia persists for more than three months postoperatively. There is no evidence-based treatment for persistent dysphagia after anti-reflux surgery, retrospective data suggest that pneumatic dilation may be efficacious. The aim of this study was to evaluate the effect of pneumatic dilation on persistent dysphagia after fundoplication in a sham-controlled study.

Aims & Methods

We performed a multicenter, single blind, randomized sham-controlled trial of patients with persistent dysphagia (> 3 months) after fundoplication. Patients with an Eckardt symptom score ≥ 4 were randomly assigned to pneumatic dilation (PD) using a 35-mm balloon or sham dilation. Primary outcome was treatment success, defined as an Eckardt score < 4 and a minimal reduction of 2 points in the Eckardt score after 30 days.

Secondary outcomes included change in stasis on timed barium esopha-gogram, change in high-resolution manometry parameters and questionnaires on quality of life, reflux and dysphagia symptoms.

Results

Forty-two patients were randomized. in the intention to treat analysis, the efficacy of pneumatic dilation (7/21 patients (33%)) and sham dilation (8/21 patients (38%)) were similar after 30 days (p=0.747). There was no significant difference in change of stasis on the timed barium esophagogram after 2 minutes (PD vs sham: median 0.0 cm, IQR 0.0 -4.3 cm vs median 0.0 cm, IQR 0.0 - 0.0; p=0.122) or change in lower esophageal sphincter relaxation pressure (PD vs sham: 10.56 ± 6.25 vs 14.60 ± 6.17 mmHg; p=0.052). Quality of life, reflux and dysphagia symptoms were not significantly different between the two groups.

Conclusion

Pneumatic dilation with a 35-mm balloon is not more effective as sham dilatation for the treatment of persistent dysphagia after fun-doplication.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.185

OP196 Artificial Intelligence To Predict Lymph Node Metastasis in T2 Colorectal Cancer For Endoscopic Full-Thickness Resection

K Ichimasa ^1,^✉, S Kudo ¹, K Nakahara ¹, B Villard ², Y Mori ¹, M Misawa ¹, Y Maeda ¹, H Nakamura ¹, N Ogata ¹, T Kudo ¹, T Hayashi ¹, K Wakamura ¹, H Miyachi ¹, N Sawada ¹, H Itoh ², M Oda ², K Mori ², F Ishida ¹

Introduction

Endoscopic full-thickness resection is an emerging technique for removal of colorectal lesions, mainly early (T1) colorectal cancer (CRC) or non-lifting colorectal lesions.¹ ² Although surgical resection with lymph node dissection is a standard strategy for T2 CRC, they actually present 25% of lymph node metastasis (LNM)-positivity. Thus, the remaining 75% of LNM-negativity patients can choice endoscopic full-thickness resection without lymph node dissection.

The aim is to investigate whether artificial intelligence (AI) can predict LNM-positivity of T2 CRCs.

Aims & Methods

A total of 511 consecutive patients with T2 CRCs that were surgically resected from 2001 to 2016 were included. We excluded those with ulcerative colitis, Lynch syndrome, FAP, synchronous advanced carcinoma and missing data. We divided patients into two groups according to date: data from 411 patients (2001-2014) were used for machine learning for the AI model, and the remaining 100 patients (2014-2016) were included for model validation.

The AI model analyzed 7 clinicopathological factors (patient age, sex, tumor size, location, tumor differentiation, lymphatic invasion, vascular invasion) and artificial neural network was used to predict the likelihood of LNM occurrence. Operative specimens were used as the gold standard for the presence of LNM. The AI model was validated by calculating the AUC (area under the curve) of ROC (receiver operator characteristics) for prediction of LNM.

Results

The rate of LNM-positivity in training and validation dataset was 26% (106/411) and 28% (28/100), respectively. The AUC of the AI model was 0.96. The AI model showed the sensitivity of 94%, specificity of 96% and accuracy of 95%, respectively.

Conclusion

The AI model can accurately predict LNM-positivity in T2 CRCs. This model will enable the patients of T2 CRCs to have an option of endoscopic full-thickness resection in the near future. UMIN clinical trial: 000038257

The ethics committee of Showa University Northern Yokohama Hospital: No. 17H-066

Disclosure

Nothing to disclose

References

1.Guillaumot M.A., Barret M., Jacques J. et al. Endoscopic full-thickness resection of early colorectal neoplasms using an endoscopic submucosal dissection knife: a retrospective multicenter study. Endosc Int Open 2020; 8: E611–E616. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Kuellmer A., Mueller J., Caca K. et al. Endoscopic full-thickness resection for early colorectal cancer. Gastrointest En-dosc 2019; 89: 1180–1189 e1. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.186

OP197 Microbial Community Heterogeneity Within Colorectal Neoplasia and Its Correlation with Colorectal Carcinogenesis

W Liu ^1,^✉, X Zhang ¹, HCH Lau ¹, L Zhao ¹, H Chen ¹, J Yu ¹

Introduction

Gut microbial dysbiosis has pivotal involvement in colorectal cancer (CRC). However, the intratumoural microbial mapping and its association with CRC progression remains elusive. Hence, we aimed to determine the microbial community architecture within a neoplasia and its contribution to colorectal carcinogenesis.

Aims & Methods

We collected 266 tissue biopsies from patients with CRC or adenoma, and 4-6 biopsies from a neoplasia plus 2 biopsies from adjacent normal mucosa per each individual. Microbial profiling was performed with subsequent investigation on microbiota diversity index and heterogeneity. The correlation of intratumoural microbial dysbiosis with host genetic alterations of KRAS mutation and microsatellite instability (MSI) in all CRC biopsies was also analysed.

Results

Upon normalising microbial abundance among lesion samples with their paired controls, we discovered that the intratumoural micro-bial community was heterogeneous and abundances of some microbes including well-documented CRCassociated pathobionts (Fusobacterium, Bacteroides, Parvimonas and Prevotella) were highly varied within a single neoplasia. When correlating such heterogeneity with CRC development, we identified progressive taxonomic alteration of microbes with high intra-neoplasia abundant variation. Moreover, we found that each microbe had its own altering pattern of intra-neoplasia abundant variation along the adenomacarcinoma sequence. We further revealed the significant difference in intratumoural microbiota between samples with and without KRAS mutation (p < 0.001) or MSI (p < 0.001), and illustrated the association of intratumoural microbial heterogeneity with host genetic alteration.

Conclusion

We demonstrated that the intra-neoplasia microbiota is heterogeneous, which displays its correlation with colorectal carcinogenesis. Our findings have provided new insight for the contribution of gut microbiota heterogeneity in CRC progression.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.187

OP199 A Free Resection Margin Less Than 1 Mm After Local Resection of T1 Crc Is Associated with A Low Risk of Disease Recurrence

KM Gijsbers ^1,^2,^✉, L van der Schee ², T van Veen ², A-M van Berkel ³, F Boersma ⁴, KJC Haasnoot ², K Kessels ⁵, RWM Schrauwen ⁶, R-M Schreuder ⁷, T Seerden ⁸, BWM Spanier ⁹, J Terhaar sive Droste ¹⁰, FP Vleggaar ², W De Vos Tot Nederveen Cappel ¹¹, MM Lacle ¹², F ter Borg ¹, LMG Moons ²; T1 CRC Working Group

Introduction

Several prediction models use a free resection margin < 1 mm as risk predictor of intramural cancer recurrence after local excision of T1 colorectal cancer (CRC). However, the evidence for using the < 1 mm threshold is under discussion. The aim of this study was to evaluate the risk of residual cancer after local resection of T1 CRC in patients with a free resection margin of 0.1-1 mm, in the absence of other histological risk factors.

Aims & Methods

A multicenter retrospective observational cohort-study was performed, identifying all consecutive T1 CRCs diagnosed between 2014 and 2017 in 11 Dutch hospitals.

Data on patient and polyp characteristics, histological risk factors, treatment approach and follow-up (imaging and endoscopy) were collected. Within this cohort, patients were identified meeting the following inclusion criteria: 1) local resection (endoscopic or TAMIS) of a T1 CRC, 2) a free resection margin of 0.1-1mm, 3) absence of poor differentiation and lym-phovascular invasion.

Main outcomes were the proportion of local intramural residual cancer (composite of local recurrence during follow-up, or cancer cells in the local resection scar in adjuvant resection specimens) and proportion of adverse outcomes (composite of aforementioned local intramural residual cancer together with lymph node- and distant metastases). All cases with adverse outcome were reviewed by an expert gastrointestinal pathologist (histological risk factors and resection margins on the hematoxylin-eosin staining slides of the original specimens).

Results

Out of 1616 T1 CRC patients, 169 met the inclusion criteria (mean age 68 years (SD 7.5 years), 31% female, 93% ASA I-II). Median follow-up time was 27 months (IQR 18-39 months), 66% of T1 CRCs were detected within the national bowel cancer screening program. Thirty-seven percent were located in the rectum, and 28% had a pedunculated morphology. Median size was 15 mm (IQR 12-20 mm), 91% was removed en-bloc. Three different treatment strategies were used, 98 (58%) underwent periodic endoscopic evaluation of the scar (Group A), 26 (15%) received an adjuvant full-thickness resection of the scar (Group B), and 45 (27%) had an oncological resection (Group C). Overall, local intramural residual cancer was observed in 5/169 (3.0%, 95% CI 1.0-6.8%). This was 4/98 (4.1%) in group A, 1/25 (4.0%) in group B, and none in group C.

Histological review showed high-grade tumor budding in 4 out of 5 (80%) patients with local intramural residual cancer. in the case without highgrade tumor budding, lymphovascular invasion was not assessable due to fragmentation. An adverse outcome was observed in 12 patients (7.1%, 95% CI 3.7-12.1%). This was in 4/98 (4.1%) in group A, 2/25 (8.0%) in group B, and 6/46 (13.0%) in group C.

All cases of adverse outcome were seen in non-pedunculated T1 CRCs. Despite the original histology report being negative for histological risk factors, poor differentiation and/or lymphovascular invasion was observed through histological review in 73% of patients with an adverse outcome. Including high-grade tumor budding, 82% of patients with adverse outcome had a risk factor present.

Conclusion

A free resection margin < 1 mm does not seem to be an independent predictor for intramural residual cancer, especially in the absence of high-grade tumor budding. Identifying histological risk factors seems to be a crucial challenge making double reading essential in T1 CRCs.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.188

OP200 The Interplay of Atg16L1 and Xbp-1 Crucially Coordinates Epithelial Dna Damage Responses and Protects From Intestinal Carcinogenesis

G Laue ^1,^✉, N Kakavand ¹, L Welz ¹, JP Bernardes ¹, F Tran ^1,², S Schreiber ², P Rosenstiel ¹, K Aden ^1,²

Introduction

The unfolded protein response (UPR) and macro-autopha-gy are key cellular homeostatic pathways and mutations of genes involved in these pathways are associated with IBD^1,2. Macro-autophagy is a cellular catabolic pathway executing degradation of cellular components in response to various stress stimuli. The UPR is a three-way program responding to unfolded proteins in the endoplasmic reticulum (ER) and controls protein translation, increases ER-capacity, promotes autophagy³ and can induce mitochondrial mediated apoptosis^4,5 as last resort. We have previously shown that deficiency in the enzyme RNaseH2b^6,7 leads to impaired ribonucleotide excision repair and induced spontaneous DNA damage with subsequent suppression of stem cell proliferation. The aim of this study was to test the role of autophagy (ATG16L1) and unfolded protein response (XBP-1) and their interplay in coordinating epithelial stem cell fate in the context of DNA damage.

Aims & Methods

H2b^dIEC mice, ATG16L1^dIEC mice and XBP-1^dIEC mice were established with the aid of Cre-loxP system and crossed to generate ATG16L1/ RNaseH2b^dIEC mice, XBP-1/RNaseH2b^dIEC mice and ATG16L1/XBP-1/RNas-eH2b^dIEC mice. To study the impact of ATG16L1 and XBP-1 on chronic DNA damage in vivo we performed basal phenotyping of young and aged mice using Immunohistochemistry (IHC) and Immunofluorescence. ModeK-cells were used to study the impact of ER-stress and autophagy on DNA damage. We generated organoids from small intestinal crypts to describe in vitro the direct molecular effects of ATG16L1 and XBP-1 on epithelial DNA damage with Western Blot and quantitative real time PCR. Propidium iodide (PI) staining, flow cytometry and colony forming assay were carried out to measure epithelial stem cell suppression and cell death.

Results

Induction of macro-autophagy is part of the intestinal epithelial response to DNA-damage as shown in ModeK cells under AraC stimulation by increased LC3 conversion and AMPK-phosphorylation in Western Blot. Upon inhibition of autophagy epithelial cell death was increased under DNA-damage conditions.

Using IHC, we observed that ATG16L1/RNaseH2b^dIEC mice and ATG16L1/ XBP-1/RNaseH2b^dIEC mice displayed increased DNA damage (yH2Ax) and epithelial cell death (Tunel) compared to RNaseH2b^dIEC mice and XBP-1/ RNaseH2b^dIEC mice.

Intestinal organoids and siRNA-transfected ModeK-cells with deficient autophagy were more prone to cell death in response to DNA-damage as indicated by PI staining. Despite ongoing cell death, BrdU- and Ki67- IHC revealed that defect autophagy significantly increased epithelial proliferation in ATG16L1/RNaseH2b^dIEC mice compared to RNaseH2b^dIEC mice. Prevalence of malignancies was two times increased from 50% of all aged XBP-1/RNaseH2b^dIEC mice (n=12) to 90% of all aged ATG16L1/XBP-1/RNas-eH2b^dIEC mice. (n=10).

Conclusion

We conclude that dysfunctional macro-autophagy, which is associated with IBD, can promote DNA-damage induced cell death but also increase epithelial proliferation.

In addition we show that UPR and autophagy mutually fine tune p53-de-pendent DNA damage response to protect from intestinal carcinogenesis.

Disclosure

Nothing to disclose

References

1.Kaser et al. XBP1 links ER stress to intestinal inflammation and confers genetic risk for human inflammatory bowel disease. Cell (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Cheng et al. T300A polymorphism of ATG16L1 and susceptibility to inflammatory bowel diseases: A meta-analysis. World J Gastroenterol (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Ogata et al. Autophagy is activated for cell survival after endoplasmic reticulum stress. Mol Cell Biol (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Ma et al. Two distinct stress signaling pathways converge upon the CHOP promoter during the mammalian unfolded protein response. J Mol Biol (2002). [DOI] [PubMed] [Google Scholar]
5.McCullough et al. Gadd153 sensitizes cells to endoplasmic reticulum stress by down-regulating Bcl2 and perturbing the cellular redox state. Mol Cell Biol (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Bartsch et al. RNase H2 Loss in Murine Astrocytes Results in Cellular Defects Reminiscent of Nucleic Acid-Mediated Autoinflammation. Front Immunol (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Aden et al. Epithelial RNase H2 Maintains Genome Integrity and Prevents Intestinal Tumorigenesis in Mice. Gastroenterology (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.189

OP201 The Gut Microbiome Follow-Up Analysis Reveals Its Fingerprint Potential and Clinical Relevance

L Chen ^1,^✉, D Wang ², S Garmaeva ³, A Kurilshikov ², A Vich Vila ⁴, R Gacesa ⁴, T Sinha ³, RK Weersma ⁴, C Wijmenga ⁵, AP Zhernakova ⁶, Study Fu J, LifeLines Cohort ²

Introduction

The gut microbiome can undergo dynamic changes over the course of an individual's life. However, we still know little about long-term temporal shifts in the human gut microbiome, nor about what drives these changes or their potential influence on host health and disease.

Aims & Methods

To understand the temporal changes of the gut micro-biome and its linkage with host health and disease. We have followed 338 participants in the Dutch population-based Lifelines-DEEP cohort for 4 years. Each individual has been deeply phenotyped for various lifestyle, clinical and physiological factors at two time points. Metagenom-ics sequencing of faecal samples at both time points allowed us to assess changes in microbial taxonomic and functional composition, as well as the abundance of antibiotic resistance and virulence genes. Apart from microbial compositional changes, indeed the gut microbiome can also undergo genetic evolution or microbial strain replacement that potentially contribute to their functional changes, which may in turns affect its host health status.

Thus, we have also compared microbial genomic variability at strain level by calling single-nucleotide polymorphisms (SNPs) i.e., SNP haplotype differences in species specific genome and structure variants, including copy number variations and presence/absence of genomic regions in the bacterial genomes. We have also evaluated the potential of utilizing mi-crobial genetic information as its host's fingerprint based on clustering approach.

Results

We observed that the gut microbial composition shows higher long-term stability in people with high microbial diversity when compared with people with low diversity. We also found that the gut microbiome showed higher compositional and genomic variability between- than within-individuals. Interestingly, a set of genetically stable microbes showed personal specificity and can uniquely distinguish their hosts with accuracy up to 94% e.g., Methanobrevibacter smithii, which highlights the gut microbiome can sever as its host's fingerprint. Moreover, individual phenotypic changes over time associated with microbial changes. For instance, we observed that the increment of plasma levels of haemoglobin A1c (HbA1c) was associated not only with the decreased abundance of a glucose metabolism pathway but also the absence of a genomic region in Bacteroides uniformis that encodes starch metabolism gene (P< 10^-5). in addition, we observed high presence of mi-crobial antibiotic resistance in the follow-up (P_Wilcox< 10^-8), particularly for tetracycline resistance that was positively associated with meat consumption (r=0.18, P=9.2x10^-4). This finding suggests that residual antibiotics from animal husbandry products may contribute to microbial antibiotic resistance in the human gut.

Conclusion

Our microbiome follow-up analysis highlights microbial changes can be driven by environmental exposures and may affect host metabolic health. and the gut microbiome can sever as its host's fingerprint.

Disclosure

None

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.190

OP202 Gpr35 Signaling in Epithelial Cells Facilitates Goblet Cells Maturation and Limits Citrobacterrodentium-Induced Colitis

H Melhem ^1,^✉, B Kaya ¹, JH Niess ²

Introduction

To maintain colonic mucosal barrier integrity, the gut epithelium constantly regenerates by rapid proliferation and differentiation from colonic stem cells to specific epithelial cell types. The identification of G protein-coupled receptor (GPR) 35 polymorphisms as a risk factor associated with inflammatory bowel disease (IBD) indicates that functional defects of GPR35 may be related to the pathogenesis of IBD [1,2]. GPR35 has been shown to promote intestinal epithelial cell turnover during wound healing and tumorigenesis [3-5].

Aims & Methods

Although GPR35 is highly expressed in the colon, its importance for the differentiation and maturation of epithelial cells remains largely unknown. Here, we examined the relevance of GPR35 for the differentiation of stem cells into mature goblet cells and colonocytes with single cell RNA sequencing (scRNA-seq) in mice lacking GPR35 specifically in the epithelium.

Results

To identify the GPR35-expressing cells, we generated a GPR35-tdTomato reporter mouse line which we crossed with the macrophage CX3CR1-GFP reporter mice. Ex vivo imaging and flow cytometry analysis of the double reporter mice revealed that GPR35 is localized in epithelial cells and in lamina propria CX3CR1+ phagocytes of the ileum, proximal and distal colon. Because epithelial cells and macrophages express GPR35, both cells type are thought to be contributors to GPR35-mediated host defense.

To identify the importance of epithelial GPR35 on mucosal barrier, we generated a mouse line in which we specifically deleted GPR35 from the epithelium by crossing GPR35^f/f mice with Villin^cre mice. in a steady state, epithelial GPR35 deletion altered the epithelium composition of the proximal colon as indicated by decreased mRNA expression levels of genes related to stem cells (Lgr5, Axin2 and ASCL2), colonocytes (Epcam, Villin) and goblet cells (Muc2).

Further, goblet cell numbers in histological sections showed a significant decrease in GPR35^f/fVil^cre mice compared to WT littermates. Dysregulation of goblet cell maturation in GPR35 deficient mice was reflected by a transcriptional imbalance of genes involved in goblet cell differentiation and maturation Atoh1 and GIF1. Cultured organoids from GPR35^f/fVil^cre mice showed low budding/crypt formation as compared to organoids from WT mice at day eight. The absence of mature goblet cells in organoids from GPR35^f/fVil^cre mice was verified by decreased mRNA expression levels of Muc2, Atoh1 and GIF1. scRNA-seq indicated that epithelial deletion of GPR35 alters the development of stem cells towards matured epithelial cell subtypes.

Mechanistically, scRNA-seq analysis showed upregulation of oxidative phosphorylation and tricarboxylic acid cycle genes in colonic stem cells lacking GPR35 compared to WT stem cells suggesting mitochondrial dysfunction in these cells. Loss of goblet cells in GPR35^f/fVil^cre mice indicated by AB/PAS and Muc2 staining resulted in increased susceptibility to Citro-bacter rodentium (C. rod.)-induced colitis as indicated by increased fecal bacterial abundance and dissemination of C. rod. to the liver or mesenteric lymph nodes in the early phase of the infection.

Conclusion

In this study, we discovered that GPR35 expressed by epithelial cells is necessary for the differentiation of stem cells into mature goblet cells and for protection against C. rod. infection. Epithelial expression of GPR35 directs proper stem cell differentiation and mucin production, which enforces a sterile inner-mucus barrier and, during infection, facilitates defense against enteric pathogens.

Disclosure

Nothing to disclose

References

1.Ellinghaus D., Folseraas T., Holm K., Ellinghaus E., Melum E., Balschun T., Laerdahl J.K., Shiryaev A., Gotthardt D.N., Weismuller T.J. et al. (2013). Genome-wide association analysis in primary sclerosing cholangitis and ulcerative colitis identifies risk loci at GPR35 and TCF4. Hepatology 58, 1074–1083. [DOI] [PubMed] [Google Scholar]
2.Farooq S.M., Hou Y., Li H., O'Meara M., Wang Y., Li C., and Wang J.M. (2018). Disruption of GPR35 Exacerbates Dextran Sulfate Sodium-Induced Colitis in Mice. Dig Dis Sci 63, 2910–2922. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Ji S.G., Juran B.D., Mucha S., Folseraas T., Jostins L., Melum E., Kuma-saka N., Atkinson E.J., Schlicht E.M., Liu J.Z. et al. (2017). Genome-wide association study of primary sclerosing cholangitis identifies new risk loci and quantifies the genetic relationship with inflammatory bowel disease. Nat Genet 49, 269–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Schneditz G., Elias J.E., Pagano E., Zaeem Cader M., Saveljeva S., Long K., Mukhopadhyay S., Arasteh M., Lawley T.D., Dougan G. et al. (2019). GPR35 promotes glycolysis, proliferation, and oncogenic signaling by engaging with the sodium potassium pump. Sci Signal 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Tsukahara T., Hamouda N., Utsumi D., Matsumoto K., Amagase K., and Kato S. (2017). G protein-coupled receptor 35 contributes to mucosal repair in mice via migration of colonic epithelial cells. Pharmacol Res 123, 27–39. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.191

OP203 Role of Dna Methyltransferase 3A (Dnmt3A) in Intestinal Epithelial Cells During Homeostasis and Inflammation

A Fazio ^1,^✉, N Mishra ², J Kuiper ², S Stengel ², D Bordoni ², P Arnold ³, F Tran ², B Messner ², G Ito ², P Rosenstiel ²

Introduction

DNA methylation is one of the epigenetic modifications which regulates gene transcription in a cell-type and tissue-specific manner. DNA methyltransferases (DNMTs) are responsible for the establishment and maintenance of the DNA methylation profile. The effects of DNA methylation in intestinal development and homeostasis have been extensively investigated, the specific role of DNMT3A in intestinal homeostasis and inflammation is still not clear.

Notably, genetic variants in the human DNMT3A gene have been associated with an increased risk of IBD and aberrant DNA methylation has been extensively observed in several diseases, including IBD. However, it is unknown if DNMT3A may play a role during the mechanism involved in this abnormal methylation pattern and consequently the development of diseases.

Aims & Methods

To assess the function of DNMT3A in IECs, conditional deletion of DNMT3A in IEC mouse model has been used. Gene expression analysis and DNA methylation have been investigated via RNA-seq and RRBS technology. in order to unveil the biological function of DNMT3A under inflammatory condition, acute and chronic DSS-induced colitis models have been performed.

To further depict how DNMT3A contributes on intestinal homeostasis, intestinal organoid from Dnmt3a^DIEC mice and control were generated and employed for further in vitro investigation on the functional and molecular levels.

Results

We find that mice that lack Dnmt3a in IECs are more susceptible to chronic chemically-induced colitis. Indeed, Dnmt3a deficient mice had increased weight loss and reduced expression of the intestinal stem cell marker Olfm4, suggesting a delay during regeneration process. To determine the role of DNMT3A during the early regeneration phase, we treated the mice with a short course of DSS in drinking water followed by normal drinking water. Dnmt3a^DIEC mice were characterised by a significant increase in fecal blood, damaged epithelial area and reduced wound healing. Reduced regeneration property was associated with strongly decreased epithelial proliferation and reduced expression of intestinal stem cell marker Lgr5 and the mucosal wound healing marker TFF3. Decreased gene expression of the adherens junctional proteins, E-cadherin and b-catenin, and the tight junctional protein occludin was observed in both the early stage and chronic inflammation in Dnmt3a^DIEC mice. Furthermore, we studied the barrier function in a 3D system ex vivo, using murine colonic organoid. DNMT3A KO organoids were more permeable to the FITC-4D dye compared to the WT control. Taken together, our data suggest that Dnmt3a deficiency in the intestinal epithelium affects the structure of adherens junctions along with reduced expression of beta-catenin and E-cadherin, suggesting a compromised mucosal barrier mechanism which might affect regeneration and permeability.

Conclusion

Our results suggest a protective role of DNMT3A during intestinal inflammation. We showed a functional role of DNMT3A in regulating intestinal permeability via transcriptional dysregulation of the apical-junctional complex proteins. in conclusion, our study provides insights into the role of DNMT3A in intestinal epithelium homeostasis, with a particular emphasis on gene expression, DNA methylation and mucosal barrier properties. These findings indicate a critical importance of DN-MT3A transcriptional regulation during homeostasis and inflammation. Thereby, altered expression of DNMT3A or its target genes could be used as alternative biomarkers to screen intestinal epithelial barrier defect and develop new potential therapeutic molecules.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.192

OP204 An Autophagy Protein Atg16L1 Protects Against Intestinal Gvhd By Regulating Interferon Signaling and Epithelial Cell Death

Y Matsuzawa ^1,^✉, A Hine ², J Axelrad ², M van den Brink ³, K Cadwell ¹

Introduction

Intestinal graft-versus-host disease (GVHD) is a life-threatening complication of allogeneic hematopoietic cell transplantation (allo-HCT). Mechanisms of susceptibility to intestinal GVHD are poorly understood. We previously demonstrated that the autophagy gene ATG16L1 is protective during allo-HCT¹. A common ATG16L1 variant (ATG16L1^T3m) was initially identified as a susceptibility factor for the inflammatory bowel disease (IBD) Crohn's disease. Both intestinal GVHD and Crohn's disease frequently involve the distal small intestine, but can involve any part of the gastrointestinal tract, and are characterized by overproduction of Th1 cytokines TNFa and IFNg and epithelial barrier disruption. More recently, we showed that Atg16L1 deletion in intestinal epithelial cells (IECs) in mice is sufficient to confer increased lethality following allo-HCT². Moreover, we and others independently demonstrated that ATG16L1 and other autophagy components inhibit a form of programmed necrosis termed necroptosis in murine intestinal organoids^2-4, a 3D cell culture system in which IEC lineages are differentiated from epithelial stem cells. However, it remains unclear how inhibiting ATG16L1 disrupts intracellular signaling to decrease viability of IECs, and the relevance to human disease requires further investigation.

Aims & Methods

In this study, we apply a preclinical mouse model of allo-HCT to demonstrate intestinal GVHD in an Atg16L1 mutant setting can be ameliorated by blocking necroptosis signaling. We then try to develop an ex vivo platform, initially with samples from mice and then with human specimens, to recreate genetic susceptibility to T cell-mediated damage by co-culturing intestinal organoids with peripheral T cells. Based on our previous findings indicating that the autophagy protein ATG16L1 protects intestinal epithelial cells against necroptosis, we chemically and genetically inhibited this cell death modality in Atg16L1 deficient mice in an improved chemotherapy-based allo-HCT model. We perform RNA-Seq analysis of intestinal organoids to examine how ATG16L1-defi-ciency leads to the impaired viability of IECs. We then developed a novel in vitro model in which intestinal organoids differentiated from mice or human biopsies were co-cultured with peripheral T cells to further examine the effect of ATG16L1 inhibition.

Results

Intestinal GVHD in Atg16L1 deficient mice was reversed by inhibiting necroptosis, and intestinal organoids derived from these mice displayed aberrant interferon signaling and loss of viability in the presence of allogeneic T cells. Pharmacologically inhibiting necroptosis or interferon signaling protected human organoids derived from individuals homozygous for the ATG16L1^T300A variant from allogeneic T cells.

Conclusion

Using information from a preclinical disease model, we developed an ex vivo platform that recreates genetic susceptibility to intestinal GVHD. Our results suggest that similar co-culture models incorporating organoids and immune cells can be applied towards elucidating mechanisms and drug targets specific to patient subsets, such as individuals who harbor the common ATG16L1 variant.

Disclosure

Nothing to disclose

References

1.Hubbard-Lucey V.M., Shono Y., Maurer K. et al. Autophagy gene Atg16L1 prevents lethal T cell alloreactivity mediated by dendritic cells. Immunity 2014; 41: 579–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Matsuzawa-Ishimoto Y., Shono Y., Gomez L.E. et al. Autophagy protein ATG16L1 prevents necroptosis in the intestinal epithelium. J Exp Med 2017; 214: 3687–3705. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Aden K., Tran F., Ito G. et al. ATG16L1 orchestrates interleukin-22 signaling in the intestinal epithelium via cGAS-STING. J Exp Med 2018; 215: 2868–2886. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Xie Y., Zhao Y., Shi L. et al. Gut epithelial TSC1/mTOR controls RIPK3-dependent necroptosis in intestinal inflammation and cancer. J Clin Invest 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.193

OP205 P53 Orchestrates Dna Damage-Induced Stem Cell Repression Via Sting/Ifn-I Mediated Epithelial Necroptosis

H Xiang ^1,^✉, G Rios ¹, F Wottawa ¹, J Berrnardes ¹, F Tran ^1,², N Kakavand ¹, L Welz ¹, K Aden ^1,², P Rosenstiel ¹

Introduction

RNase H2 is responsible to remove ribonucleotides from incorporated RNA: DNA hybrids and absence of RNase H2 leads DNA damage and formation of dsDNA containing micronuclei, which are released into the cytosol to activate endogenously cGAS/STING. (1, 2). After dsDNA sensing of cGAS, STING can recruit TBK1 and active IRF3, and induce the expression of INF-I, which leads to downstream events to clear dsDNA(3-5). Our previous work showed that chronic DNA damage due to knock out of RNase H2 can leads to stem-cell suppression in IEC in a P53 dependent manner (6). To which extent p53 is involved in the DNA-damage induced activation of the cGAS/STING pathways is not known. Here, we investigated the impact of cGAS/STING in mediating the p53 dependent stem-cell suppression in RNase H2b (H2b)^iIEC mice.

Aims & Methods

We aim to study the role of STING in response to DNA damage and figure out the potential function of STING on intestinal stem cell. We hypothesized that STING acts downstream of p53 and thereby mediates stem cell repression in response to DNA damage. To confirm our hypothesis, we created H2b^iIEC /Tmem173^/- and H2b/p53^iIEC, which were used for intestinal organoid generation and baseline phenotyping. We assessed STING expression using in-vitro and in-vivo approaches in the context of DNA damage using H2b^fl/fl, H2b^aIEC, H2b^iIEC/Tmem173^/- and H2b/ p53^iIEC small intestinal organoids and small intestinal immunohistochem-istry. The impact of STING on p53-driven stem cell proliferation was assessed using intestinal organoid colony formation assay and EdU-stain-ing. Findings were validated in IHC of small intestinal sections from mice, using Ki67, BrdU and Tunel staining. DSS Colitis was performed to test the impact of STING deficiency on colitis susceptibility in H2b/Tmem173^AIEC mice. RNA sequencing of intestinal organoids of indicated genotypes were performed to identify gene expression signatures underlying the proliferation phenotype.

Results

Using intestinal organoid we show that epithelial DNA damage induces STING expression in a p53 dependent manner, as H2b^AIEC organoids showed strong STING expression, which was completely abrogated in H2b ^iIEC /Tmem173^-/- and H2b/p53^iIEC intestinal organoids. Colony forming assays and EdU staining of intestinal organoids revealed that epithelial proliferation was restricted by the expression of STING as deficiency of STING showed restored proliferation in response to DNA damage. In-vivo we confirmed this restriction of intestinal epithelial proliferation using Ki67 and BrdU IHC staining. Surprisingly aged mice (52 weeks) presented with spontaneous intestinal carcinogenesis (42%, n=3/7) in H2b^aIEC /Tmem173^-/-mice, which was due to restored stem cell function in crypt stem cell niche. in DSS colitis model we observed that DNA damage - induced intestinal inflammation was mediated by STING, as suppression of epithelial proliferation was completely restored in H2b^iIEC/Tmem173^/-. Mechanistically RNA sequencing revealed that STING facilitated DNA damaged - induced stem cell restriction via induction of the IFN-I/MLKL pathway and subsequent necroptosis induction.

Conclusion

We provide evidence that STING/INF-I pathway is regulated by P53 in response to DNA damage and thereby coordinates subsequent stem cell restriction in intestinal epithelial. These data can be used to understand the role of STING between inflammatory and carcinogenesis in intestinal.

Disclosure

Nothing to disclose

References

1.Vaisman A., Woodgate R. Redundancy in ribonucleotide excision repair: Competition, compensation, and cooperation. DNA Repair (Amst). May 2015; 29: 74–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Bartsch K., Knittler K., Borowski C. et al. Absence of RNase H2 triggers generation of immunogenic micronuclei removed by autophagy. Hum Mol Genet. Oct 15 2017; 26(20): 3960–3972. [DOI] [PubMed] [Google Scholar]
3.Motwani M., Pesiridis S., Fitzgerald K.A. DNA sensing by the cGAS-STING pathway in health and disease. Nat Rev Genet. Nov 2019; 20(11): 657–674. [DOI] [PubMed] [Google Scholar]
4.Ishikawa H., Ma Z., Barber G.N. STING regulates intracellular DNAmediated, type I interferon-dependent innate immunity. Nature. Oct 8 2009; 461(7265): 788–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Ishikawa H., Barber G.N. STING is an endoplas-mic reticulum adaptor that facilitates innate immune signalling. Nature. Oct 2 2008; 455(7213): 674–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Aden K., Bartsch K., Dahl J. et al. Epithelial RNase H2 Maintains Genome Integrity and Prevents Intestinal Tumorigen-esis in Mice. Gastroenterology. Jan 2019; 156(1): 145–159 e119. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.194

OP206 The Diagnostic Value of Dilated Intercellular Spaces of Esophageal Epithelium in Patients with Gastroesophageal Reflux Disease

V Gorgulu ^1,^2,^✉, P Ergun ^2,³, S Kipcak ^2,⁴, B Doganavsargil ⁵, S Bor ^2,⁶; Ege Reflux Study Group

Introduction

Dilated intercellular spaces (DIS) are one of the earliest and important histopathological change of esophageal epithelium in gastro-esophageal reflux disease (GERD). It is not clear whether DIS has a diagnostic value for GERD. Many studies have shown that erosive esophagi-tis patients have more pronounced DIS than non-erosive reflux patients (NERD), whereas NERD patients have more pronounced DIS than healthy controls (HC). in this study, we aimed to test the diagnostic value of DIS in different phenotypes of GERD in an extensively evaluated group of patients and compared healthy controls.

Aims & Methods

25 HC and 117 GERD patients were evaluated. By using upper gastro-intestinal endoscopy, high resolution esophageal manometry and 24-h pH-impedance monitoring, GERD patients were separated to subgroups as erosive reflux disease A+B (ERD A+B) n=48, erosive reflux disease C+D (ERD C+D) n=12, NERD n=31, reflux hypersensitivity (HS) n=11, functional heartburn (FH) n=15. We evaluated our DIS findings according to Porto classification according different pH cut-off values for ERD and NERD+HS groups. Data also classified in accordance with proton pump inhibitor (PPI) response; nonresponders (< 50%), responders (≥50%), very good responders (≥80%). Biopsies were taken 3-5 cm above the lower esophageal sphincter by one researcher. Specimens were fixed with %10 buffered formalin, dehydrated with alcohol, exposed to xylol and embedded in paraffin. Paraffin blocks were sectioned at a thickness 5μm, and hematoxylin and eosin (H&E) staining was done. 10 different areas without artefact for each patient were photographed at 1000x magnification with oil lens under light microscope. 10 perpendicular measurement were performed continuously around the one epithelial cell that had the widest DIS between upper basal and lower prickle layer. Total 100 measurement were taken for each patient.

Results

When GERD patients evaluated all together (mean: 1,1027 ±0,31081μm) they showed wider DIS than HC (mean: 0,9521 ±0,21301μm) (Independent samples t-test p=0,005). We found 65% sensitivity and 72% specificity of 1.0261μm cut off value from Roc curve analysis (p=0,015). There was a weak relationship between acid exposure and DIS; Pearson correlation coefficient was 0,281 p=0,002 (2-tailed) between total % time pH< 4 at 24 hours and DIS value. No difference has been shown according to Porto classification cut-off values both for ERD patients (One way Anova p=0,715) and NERD+HS (One way Anova p=0,542). According to different PPI response rates, again there was no difference with ERD A+B+C+D (One way Anova p=0,845) and NERD+HS (One way Anova p=0,396).

[DIS Measurements of GERD phenotypes and FH]

Subgroups	HC	ERD A+B	ERD C+D	NERD	HS	FH
Mean (μm)	0,9521	1,2055^c,d	1,2942^a,b	0,97	0,9869	0,9795
SD (μm)	0,21301	0,28126	0,23532	0,31241	0,22927	0,33345

Open in a new tab

SD: standard deviation. One way Anova p=0.000. Bonferroni test: a; p=0,009 vs HC, b; p=0,011 vs NERD, c; p=0,005 vs HC, d; p=0,005 vs NERD

Conclusion

Since our results do not differ statistically between the HC, NERD, HS, FH subgroups, we assume that measurement of DIS is not useful for the differentiation of GERD subgroups except erosive esophagitis. Because ERD patients can already be diagnosed with endoscopy, it is not cost-effective to use laborious and time-consuming DIS measurement for diagnostic purposes. in addition, the sensitivity and specificity of the cut off value might not sufficient enough for the diagnosis.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.195

OP207 The Pathophysiology of Non-Erosive Reflux Disease (Nerd) Is Different in Children Compared To Adults in Regards To Oesophageal Mucosa Integrity and Innervation

K Nikaki ^1,^✉, C Lee ¹, A Ustaoglu ¹, P Woodland ¹, D Sifrim ¹

Introduction

In adults, the mucosa of patients with Non-Erosive Reflux Disease (NERD) shows an increased vulnerability “in vitro” when exposed to refluxate-like solutions (1). Moreover, the normal oesophageal mucosa has deep lying afferent nerve endings in the distal oesophagus and more superficial nerves in the proximal oesophagus. in contrast, NERD patients have more superficial nerves along the oesophageal length (2).

Aims & Methods

The aim of our study was to assess oesophageal mucosal integrity and innervation in children with NERD and controls. We prospectively recruited children undergoing an upper GI endoscopy for clinical reasons and identified a subgroup with normal histology +/-normal impedance pH-metry that served as controls and a patient group with increased acid exposure on impedance pH-metry. We obtained oe-sophageal biopsies from 3-5cm above the lower oesophageal sphincter and from the proximal/upper third. Fresh distal biopsies were placed in Ussing chambers (surface area:0.011cm²) and exposed to pH2 and pH5 solutions containing bile acids and pepsin. The baseline trans-epithelial resistance (TER) and its % of drop when exposed to the refluxate-like solutions were recorded. Distal and proximal biopsies were immunohisto-chemically stained with CGRP. CGRP positive nerve fibres were identified and their position relative to the lumen was determined (expressed as median number of cell layers from the lumen for all sections examined).

Results

For the mucosal integrity studies, 19 children were included (12M:7F, median age: 11 years, range: 0.83-15 years) in the control group and 9 in the NERD group (4M:5F, median age: 6 years, range: 1-15 years). The baseline TER were comparable between the two groups (179 vs 175 Ohms). The drop of the TER when mucosa was exposed to refluxate-like solutions with pH2 and pH5 was -44% and 0% in the control group and -65% and +10% in the NERD group (non-significant differences). For the innervation studies, 19 children were included in the control group (13M:6F, median age: 11 years, range 0.83-15 years) and 15 children in the NERD group (8M:7F, median age: 10 years, range 1-16 years). in the control group, CGRP positive nerve fibres were identified at a median of 19.5 cell layers from the lumen in the proximal and 19 cell layers in the distal oesophagus. in the NERD group, CGRP positive nerve fibres were identified at a median of 17.5 cell layers in the proximal and 17 cell layers in the distal oesophagus (non-significant differences). There is no correlation between the age and the position of the nerve fibres identified.

Conclusion

Unlike adults with NERD, the oesophageal mucosa in children with NERD does not show an increased vulnerability to acid exposure. The oesophageal mucosa innervation in children is similar in the proximal and distal oesophagus with deep lying nerve fibres both in controls and NERD patients, in contrast to adult healthy volunteers who have superficial nerves in the proximal oesophagus and NERD patients who have superficial nerves in the distal oesophagus too. These differences may be due to different composition/pH of the refluxate, duration of exposure to reflux and subsequent inflammation and/or repair mechanisms.

Disclosure

Nothing to disclose

References

1.Woodland P., Lee C., Duraisamy Y. et al. Assessment and protection of esophageal mucosal integrity in patients with heartburn without esophagitis. Am J Gastroenterol 2013; 108(4): 535–43. [DOI] [PubMed] [Google Scholar]
2.Woodland P., Ooi J.L., Grassi F. et al. Superficial esophageal mucosal afferent nerves may contribute to reflux hypersensitivity in non-erosive reflux disease. Gastroenterology 2017. [DOI] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.196

OP208 Lack of Significant Differences in Somatization, Anxiety, Depression and Fear of Pain Levels Across The Gastro-Oesophageal Reflux Disease Spectrum

A Geeraerts ^1,^✉, L Van Oudenhove ², H Geysen ¹, T Vanuytsel ³, J Tack ⁴, A Pauwels ¹

Introduction

Co-morbid somatization, (symptom-specific) anxiety and depression symptoms are prevalent in functional gastrointestinal disorders, such as functional dyspepsia. Based on previous research, it has been hypothesized that patients with functional heartburn (FH) display higher levels of anxiety and depression compared to the other phenotypes in the gastro-oesophageal reflux disease (GORD) spectrum. Furthermore, some studies suggest that psychological aspects play an important role in the pathophysiology of reflux hypersensitivity (RHS). However, it remains to be established that psychosocial comorbidity increases across the GORD spectrum, from true GORD to FH.

Aims & Methods

To investigate levels of somatization, anxiety, depression, and fear of pain among the different GERD phenotypes in a large cohort of patients seen at a tertiary care center. Patients with typical GERD symptoms (‘on’ or ‘off’ proton pump inhibitors, undergoing 24 hour impedance-pH monitoring (MII-pH) were asked to fill out several questionnaires. Patients were classified in 4 categories based on the MII-pH recordings (Lyon consensus): true GERD, borderline GERD, RHS and FH. Depression and somatization were measured using the Patient Health Questionnaire (PHQ-9, PHQ-12, PHQ-15), anxiety was measured using the generalized anxiety module of the PHQ(7) and fear of pain was assessed with the Pain Anxiety Symptom Scale (PASS). Data were analyzed using SAS 9.5 (SAS Institute, Cary, NC, USA). General linear models were used to compare the averages between the 4 groups.

Results

Three hundred and fifty-four patients were included, which were classified as follows: 115 true GORD, 68 borderline GORD, 57 RHS and 114 FH. Patients with RHS were significantly younger compared to patients with true GORD, borderline GORD and FH. Although the percentage of females was slightly higher in RHS and FH, this did not reach statistical significance. Patients with true GORD had a significantly higher BMI compared to patients with RHS and FH (table 1). Most importantly, we could not find any significant differences between groups in PHQ and PASS scores (table 2). The prevalence of depression (PHQ9 sum score ≥10) was 41% in our cohort, which is markedly higher compared to only 5.6% in the general population. 26.5% of our patients displayed a high level of somatization (PHQ15 sum score ≥15), compared to 3.1% in the general population.

Table 1.

[Demographics and results presented as mean ±SEM]

	True GORD (n=115)	Borderline GORD (n=68)	RHS (n=57)	FH (n=114)	p-value
Age in years ±SEM	49 ±1.20	49 ±1.98	42 ±1.84	51 ±1.3	0.001
Sex (number (%))	F: 59 (51.3) M: 56 (48.7)	F: 39 (57.3) M: 29 (42.7)	F: 37 (64.9) M: 20 (35.1)	F: 71 (62.3) M: 43 (37.7)	NS
BMI in kg/m2 ± SEM	26.56 ±0.46	25.09 ±0.68	23.87 ±0.64	24.07 ±0.38	0.003
PHQ7 ± SEM	4.66 ±0.40	4.20 ±0.45	4.75 ±0.43	4.08 ±0.31	NS
PHQ9 ± SEM	6.48 ±0.50	5.93 ±0.60	7.17 ±0.85	5.80 ±0.39	NS
PHQ12 ± SEM	8.54 ±0.45	8.20 ±0.55	9.21 ±0.74	7.70 ±0.44	NS
PHQ15 ± SEM	11.31 ±0.60	11.68 ±0.65	12.74 ±0.89	10.61 ±0.54	NS
PASS (fear) ± SEM	11.20 ±0.90	11.8 ±1.20	11.85 ±1.57	10.77 ±0.80	NS
PASS-cognitive ± SEM	17.48 ±0.88	16.70 ±1.34	17.92 ±1.62	16.78 ±0.84	NS
PASS-physiological ± SEM	10.65 ±0.88	12.51 ±1.29	12.37 ±1.44	10.11 ±0.88	NS
PASS-overt behavioral ± SEM	17.25 ±0.94	18.76 ±1.38	18.31 ±1.19	18.35 ±0.93	NS

Open in a new tab

Conclusion

This study shows that somatization, anxiety, depression and fear of pain was not higher in FH and RHS compared to patients with true and borderline GORD. Overall, our cohort shows a higher prevalence rate for depression and somatization compared to the general population. Therefore, screening for psychological and extraintestinal symptoms should be considered in all phenotypes of refractory GORD symptoms.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.197

OP209 Is Sleeve Gastrectomy A Model To Study Gastroesophageal Reflux Pathophysiology? The Effects On Esophageal Function

G Gualtieri ¹, EV Savarino ², N de Bortoli ³, M Frazzoni ⁴, L Frazzoni ⁵, M Ghisa ⁶, S Parisi ¹, G Terracciano ⁷, Volpe M Lanza ¹, V Savarino ⁸, S Tolone ^9,^✉

Introduction

Exact pathophysiology of gastroesophageal reflux disease (GERD) is still subject to investigation. The interplay between defensive mechanisms and their response to aggressive factors is still to be elucidated. Sleeve gastrectomy (SG) is a bariatric procedure considered at risk for GERD development. Currently, data are scarce about the role of defensive mechanisms after SG, in particular if their modification can influence the pathogenesis of “de novo” GERD.

Aims & Methods

We aimed to assess esophageal motor function, chemical, volume clearance and mucosal integrity in obese without preoperative GERD.

We reviewed data from 35 subjects (19 female, median age 42 years) who underwent a standardized questionnaire for symptom presence and severity, upper endoscopy, high-resolution manometry (HRM) and MII-pH. Any pre-operative evidence of esophagitis (any grade), hiatal hernia and GERD, reviewed according to Lyon consensus, constituted exclusion criteria. Patients with post-operative SG morphological defect (stricture, residual fundus) were excluded.

At HRM motor function was assessed according to Chicago classification v3.0 and EGJ-Ci was also calculated. At MII-pH, distal esophageal acid exposure (AET%), number and quality (acid,weakly-acid and weakly alkaline) of reflux at MII, volume clearance (bolus exposure%) and Symptom Association Probability (SAP) were recorded. Chemical clearance was determined using the post-reflux swallow-induced peristaltic wave index (PSPWi) Mucosal integrity was assessed via the mean nocturnal esophageal basal impedance (MNBI). ANOVA and Wilcoxon test were used for comparisons. A p < 0.05 was considered significant. A group of 15 normal healthy volunteers (BMI 20-25) served as control group.

Results

In the 35 patients enrolled the preoperative (130.8 kg (119-156), Body Mass Index (BMI) =46.1 (38-58)) and postoperative (98 kg (72-110), BMI=34.7 (28-46)) anthropology was statistically different. Before SG none had any features of GERD and the results were similar to HVs. After SG, reflux symptoms incidence was not modified; any increase in perception of heartburn, regurgitation (mean GerdQ 6 vs. 7), and epigastric pain was observed.

Only one patient after SG showed esophagitis grade A. EGJ features were not dissimilar after SG. Motility resulted impaired with 46% of ineffective peristaltic waves vs 10%, p< 0.05 and a decreased DCI. The preoperative median AET% was 1.47 for total time. Postoperatively, DeMeester's score (p=0.041) and the median AET% in recumbent position (p=0.04) significantly increased (from 1 (0-1.2) to 3.1 (0-4.2), p=0.04). The SG produced an increase of total reflux episodes (33 vs. 53; p< 0.0001) detected at MII. A significant increase of postoperative weakly acid reflux episodes in both upright (17 vs. 28; p< 0.0001) and recumbent (4 vs. 8; p<0.0001) position was detected.

Bolus exposure increased in post-operative period from 0.2% to 3.1% (p< 0.05). PSPW index decreased after SG (80 ±10 vs. 50 ±20; p< 0.001), as well as MNBI, that decreased (3016±819 vs 1542 ± 452; p< 0.001). There was an inverse correlation between PSPW, MNBI and the number of reflux, and an inverse correlation between PSPW and AET (p< 0.0001).

Conclusion

Even in absence of EGJ impairment, after SG an increased number of weakly acid reflux is documented. This is associated with impairment in volume and chemical clearance, as well as to impairment in peristalsis vigour. This phenomenon reflects also damage on mucosal integrity. This data seems to support the role of a vicious circle in GERD pathogenesis.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.198

OP210 Inter-Reviewer Variability in Analysis of Reflux Episodes and Post-Reflux Swallow Induced Peristaltic Wave On Ph-Impedance Studies: The Wingate Consensus

B Rogers ^1,^✉, CP Gyawali ², M Frazzoni ³, EV Savarino ⁴, S Roman ⁵, D Sifrim ⁶

Introduction

More reflux episodes are detected with pH-impedance monitoring compared to pH monitoring alone, from identification of reflux episodes independent of pH.The post-reflux swallow induced peristaltic wave (PSPW), represents a primary peristaltic effort that brings saliva to the distal esophagus to neutralize acid as part of refluxate clearance from the distal esophagus. Reflux episodes and PSPW need accurate identification for optimal pH-impedance interpretation. Automated analysis lacks precision for reflux episodes, and is not available for PSPW interpretation.

Aims & Methods

We aimed to evaluate inter-reviewer variability among experts in reflux episode and PSPW identification in healthy controls and GERD patients.

Reflux episodes and PSPW were identified by 5 international expert reviewers in pH-impedance studies from 19 healthy controls (median age 20 years, 52.6% female) before and after a consensus meeting (the Wingate Consensus) to define identification standards for these events. The standards agreed upon were: a) reflux episode: retrograde propagation of ≥50% impedance drop for ≥4 seconds in the distal two impedance channels; b) PSPW initiation of a swallow within 30 seconds of return of impedance to baseline in the distal impedance channel, propagating through the entire esophagus with ≥50% drop in impedance in the distal-most channel. Reviewers subsequently evaluated 19 randomly selected de-identified pH-impedance studies from GERD patients (median age 52 years, 78.9% female) from four countries. Comparisons to automated analysis, and analysis of inter-reviewer agreement were performed within ±10 episodes and ±5 episodes of median reviewer interpretation values. Conditional probability of agreement, which assesses proportional agreement between reviewers, and Cronbach's alpha, which measures internal consistency of results, were utilized. Repeat analysis was performed after exclusion of one outlier value in each patient.

Results

Automated analysis significantly overestimated reflux episodes compared to median reviewer values (p≤0.001). Median reflux episodes identified remained similar (p=0.94), but PSPW decreased (p=0.05) following the consensus meeting. More significantly, inter-reviewer agreement improved for reflux episodes and PSPW within ±10 episodes of ±5-review-er medians in healthy controls (Table); corresponding values were noted in GERD patients. Agreement was lower within 5 episodes of 5-reviewer medians, and higher within both 10- and 5-episodes of 4-reviewer medians after exclusion of one outlier value.

Conclusion

There is significant disagreement between automated analysis and reviewer identification of reflux episodes on pH-impedance monitoring; automated analysis without reviewer oversight is unreliable for pH-impedance interpretation. Utilization of criteria defining reflux episodes and PSPW improves agreement and reduces variability between reviewers.

[Comparison of agreement in identification of reflux episodes and PSPW on pH-impedance studies]

	Conditional probability of agreement		Internal reliability
Degrees of agreement	Within ±10 episodes	Within ±5 episodes	Within ±10 episodes	Within ±5 episodes
Healthy controls: pre-Wingate consensus
Automated analysis	0.263	0.158
Reflux episodes: 5 reviewers	0.768	0.642	0.807	0.862
PSPW: 5 reviewers	0.811	0.621	0.813	0.764
Healthy controls: post-Wingate consensus
Automated analysis	0.316	0.105
Reflux episodes: 5 reviewers	0.800	0.674	0.853	0.890
PSPW: 5 reviewers	0.875	0.737	0.796	0.878
GERD patients
Automated analysis	0.316	0.263
Reflux episodes: 5 reviewers	0.758	0.547	0.570	0.570
PSPW 5 reviewers	0.800	0.568	0.903	0.868

Open in a new tab

Disclosure

CPG: Consulting: Medtronic, Diversatek, Isothrive, Ironwood, Quintiles; BDR: no disclosures; MF: no disclosures; AS: no disclosures; ES: Lecture Fee: Medtronic, Takeda, Janssen, MSD, Abbvie, Malesci; Consulting: Medtronic, Takeda, Janssen, MSD, Reckitt Bencikser, Sofar, Unifarco, SILA, Oftagest; SR: consulting Medtronic, research support Diversatek Healthcare, Medtronic; DS: research grants: Reckitt Benckiser UK, Jinshan Technology China, Alfa Sigma Italy

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.199

VC01 Emergency Endoscopic Treatment of A Massive Upper Gastrointestinal Bleeding of Very Uncommon Origin

S Sferrazza ^1,^✉, MF Maida ², F Vieceli ¹, G de Pretis ¹, A Lisotti ³, P Fusaroli ⁴

Introduction

Acute upper gastrointestinal bleeding is an urgent condition with significant morbidity and mortality. An EGD with prompt recognition and treatment of the lesion is essential for the patient's prognosis.

Aims & Methods

We report our experience in the management of a massive upper gastrointestinal bleeding secondary to an uncommon and potentially fatal condition. A 70-year-old Caucasian man was admitted to our unit in March 2018 for hematemesis with severe anemia (Hb 5.2 g/dL) and hypovolemic shock. His medical history included quadriplegia due to horse-riding accident and absolute dysphagia treated with enteral feeding through percutaneous endoscopic gastrostomy (PEG). There were neither history of liver disease nor intake of NSAIDs, anticoagulants and antiplatelet drugs. The coagulation values were in the normal range. Following patient evaluation and hemodynamic resuscitation with transfusions of 3 units of red blood cell, an EGD was performed showing a massive arterial bleeding from mid-esophagus, without a certain diagnosis. Computed tomography angiogram of the chest showed an aortoesoph-ageal fistula in the presence of a double aortic arch system. Following surgical consultation, the placement of a vascular endoprosthesis was excluded due to the presence of this congenital anatomical variant. A second EGD was performed with placement of a fully covered metal stent (WallFlex, Boston Scientific, USA), with full resolution of bleeding and rapid hemodynamic stabilization.

Results

After a week, the patient was asymptomatic, Hb values rose to 12.2 g/dl. The patient was discharged with indication to follow-up. After 18 months, the patient is alive, no further signs of bleeding were reported, a follow-up EGD confirmed the stent was still placed in situ, and a follow-up CT scan showed the closure of the fistula.

Conclusion

Aortoesophageal fistula is an uncommon and usually fatal condition due to massive gastrointestinal bleeding with emergent endo-vascular stenting representing the first therapeutic strategy. Endoscopic stenting still has an important role as a bridge to vascular treatment and as primary therapy, if endovascular stenting is not feasible due to altered anatomy.

Prompt diagnosis and multidisciplinary management are essential for the selection of the best therapeutic strategy and improvement of patient outcomes.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.200

VC02 Treatment of Large Main Bile Duct Stones in A Patient with 4/5 Gastrectomy and Roux-En-Y Anastomosis Using Anterograde Cholangioscopy and Electohydraulic Lithotripsy After Hepaticogastrostomy

M Schaefer ^1,^✉, A Lamoureux ¹, H Chanty ², W Yacoub ¹, V Laurent ³, A Ayav ², J-B Chevaux ¹

Introduction

Treatment of benign biliary obstruction is a challenging situation in patients with altered anatomy. ERCP with single or double balloon enteroscopy can be difficult, with an overall cannulation success of about 60%.

Aims & Methods

We describe the case of a 86-years old women with history of 4/5 gastrectomy and Roux-en-Y anastomosis and main bile duct symptomatic lithiasis, with failure of two enteroscopies. She previously underwent surgical choledocotomy with incomplete clearance of the main bile duct or early recurrence. Following multidisciplinary discussion, and considering that patient refused a new surgery, a full endoscopic approach was proposed.

Results

The first step of the treatment was an EUS-hepaticojejunostomy with placement of a fully covered self expandable metallic biliary stent. Then, we performed a direct anterograde electrohydraulic lithotripsy (Biliary EHL Probe and Autolith Touch Generator, Boston Scientifics, Marlborough) under endoscopic control, using an ultraslim videogastroscope as a cholangioscope (GIF-XP180N, Olympus, Tokyo). All visible stones were destroyed.

A guidewire was placed during cholangioscopy to place a transpapillary biliary metallic stent to enable complete biliary stones clearance.

Conclusion

Full endoscopic approach using anterograde electrohydraulic lithotripsy through an hepaticogastrostomy is feasible in patients with altered anatomy and large main bile duct stones.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.201

VC03 Endoscopic For Treatment of Complicated Cholecystocolonic Fistula with Oro-Anal Rendezvous

KD-C Pham ^1,^2,^✉, RF Havre ^1,²

Introduction

Cholecystocolonic fistula (CCF) is an uncommon complication of cholecystitis, but yet it is the second most common cholecys-toenteric fistula. Treatment is complex surgery, and most of patients are old and unfit. Unsuccessful treatment may result in sepsis and death. We describe a case of endoscopic treatment for CCF with impacted fecalith in the gallbladder.

Aims & Methods

The patient was an 89-year-old female with multiple admissions for cholecystitis. ERCP with sphincterotomy and stone extraction was performed. Due to co-morbidity, she was considered unfit for surgery, and an external gallbladder drain was placed permanently. However, her condition deteriorated. Computer tomography showed unevenly thickened wall of the gallbladder with mixed solid content and a CCF to the right colon. We hypothesized that cholecystitis was caused by fecal translocation into the gallbladder through the CCF. After consent, we planned endoscopic closure of the CCF and EUS guided drainage of the gallbladder. From the antrum we used a 19G FNA needle to puncture the gallbladder. Contrast with blue dye was injected, and we could see contrast defects and visualize CCF under fluoroscopy.

A long guidewire was passed through the FNA needle into the gallbladder. After some attempts, we succeeded maneuvering the guide wire into the colon. Leaving the EUS scope in place, we performed colonoscopy, and managed to find the guidewire, and pull it out through the anus. There was no obvious fistula opening in the colon, and it would be impossible to find it without the guide wire in place. A colonoscope with an over the scope clip (OTSC) was introduced over the guide wire. The guide wire was useful to find direction during the second colonoscopy with the OTSC. The fistula opening was trimmed with a biopsy forceps to cause inflammation for better healing. By pulling the guide wire from the oral and anal side at the same time, we achieved good anchorage and aim to deploy the OTSC, and closed the CCF from the colon.

After removing the colonoscope, with the guide wire in place, we performed EUS guided cholecysticogastrostomy with a 10 x 10mm hot LAMS over the guide wire. The LAMS was dilated with a 10mm pneumatic balloon. We found a hard mass consistent with a fecalith with many gall stones inside the gallbladder. Endoscopic debridement inside the gallbladder were performed twice until the wall of the gall bladder fundus was seen. For control, we injected contrast into the gallbladder, but didn't see passage of contrast into the colon.

Results

After the second debridement of gallbladder material, the patient was discharged. in the following 10 weeks, she did not develop cholecystitis again.

Conclusion

This case report demonstrates the feasibility of endoscopic treatment of a complicated CCF with a mixed fecalith inside the gallbladder.

Disclosure

Pham KDC has received lecture fee from Boston Scientific and Olympus.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.202

VC04 Clip Muscle Protection (Climp) Method in Muscle-Retracting Sign During Colonic Endoscopic Submucosal Dissection

N Suzuki ¹, A Dhillon ^1,^✉, E Seward ², A Humphries ¹

Introduction

Muscle retracting (MR) sign is a feature occasionally observed during Endoscopic Submucosal Dissection (ESD)¹. The muscle layer can be pulled towards a neoplastic lesion due to the desmoplastic reaction associated with cancer invasion, or it may be due to fibrosis caused by mechanical forces of intestinal peristalsis pulling on the body of the polyp over time. The MR sign indicates a potentially difficult ESD and reduces the chance of a complete resection.

Aims & Methods

We performed a Clip Muscle Protection (CliMP) method, in which clips are attached at the base of the retracting muscle during colonic ESD, for 6 benign polyps. When MR sign was encountered during ESD, the surrounding submucosal layer was dissected to expose retracted muscle and endoclips were applied at the base of the tented area. This sealed the muscle and allowed further resection above the clipped area.

Results

A complete resection was possible in 4 out 6 cases. Two CliMP cases are shown on the accompanying video. Morphologically they were broad based Ip in 5/6 polyps at the sigmoid colon, the final lesion was a LST nodular mixed type at the rectosigmoid junction. The median size of the polyps was 45 mm in diameter (range between 35-75mm). No complications were observed. No electrocautery effect was observed at the clip attachment site. All 6 lesions were found to be tubular or tubulovillous adenomas with high grade dysplasia on histopathological analysis. R0 resection was achieved for all of the four completed cases; two procedures were abandoned due to a broad MR sign in one and an inability to access the whole of the lesion due to sigmoid fixation in the other.

Conclusion

CliMP method appears to allow continuous deeper dissection without complication in lesions demonstrating MR sign during colonic ESD.

Disclosure

Nothing to disclose

References

1. Toyonaga. et al. Clinical significance of the muscle-retracting sign during colorectal endoscopic submucosal dissection. EIO. 2015 doi: 10.1055/s-0034-1391665. [DOI] [PMC free article] [PubMed] [Google Scholar]

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.203

VC05 First Successful Transplantation of Protective Small Intestinal Mucosa To A Former Scar Area in The Cervical Esophagus After Circumferential Scar Excision and Vacuum Sponge Conditioning For Neo-Vascularisation and Better Transplant Acceptance in Man

J Hochberger ^1,^✉, J Bernhardt ², P Koehler ³, S Kopp ⁴, M Loss ⁵

Introduction

A 62 year old man had undergone in july 2015 curative circumferential tubular endoscopic submucosal dissection (ESD) from 20-27 cm aborally due to an early squamous cell cancer (SCC) located in the upper esophageal sphincter area due to poor surgical alternatives, RFA or radio-chemotherapy options

Aims & Methods

Different measures to prevent stricture formation failed and one year later the patient had to return every 10 days to the hospital for dilatation. Due to again poor surgical alternatives an experimental concept was carried out after acute and chronic animal experiments in the pig. As the scar area was likely to be poorly vascularized the scar was first excised in an again tubular fashion from the upper esophageal sphincter over 7 centimeters including the entire scar. Two polyurethan vacuum sponges were used and changed every 3-4 days over 20 days in order to stimulate neovascularization of the area using a continuous vacuum of -125 mm Hg. in a second intervention a 30 cm segment of small intestine was resected surgically and the mucosa specially prepared and transplanted to the pri-orly conditioned scar area. The specimen was temporarily fixed against the wall using a non-covered nitinol stent.

Results

Two month after the second procedure several islands of histo-logically proven vital PAS positive small intestinal mucosa could be observed. in the meantime, villi of small intestinal mucosa can clearly be visualized. in the meantime intestinal mucosa has spread out over 3 cm in the cervical esophagus and is stable. Clinically the patient has recovered completely from the intervention and works full time as engineer.

Conclusion

Our case shows the feasibility of a completely new concept: Small intestinal mucosa is easy to harvest, has a tubular form and a diameter similar to the one of the esophagus. Conditioning of the muscle after endoscopic excision of a scar in the esophagus using polyurethane vacuum sponges similar to their use in plastic surgery is a second new concept demonstrated in this case that may offer hope to patients with chronic severe benign esophageal strictures. A combined primary surgical and endoscopic approach for widespread resection of early esophageal cancers plus subsequent immediate transplantation of protective small intestinal mucosa may offer a new perspective for widespread early SCC with increased risk for secondary stricturing after endoscopic resection. However, a number of further pre-clinical and clinical trials have to be carried out to systematically reproduce this first case.

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.204

VC06 Double-Balloon Enteroscopy (Dbe) As A Stabilising Platform For Saline-Immersion Therapeutic Endoscopy (Site) Facilitated Endoscopic Submucosal Dissection (Esd) of A Large, Intussuscepting Ileal Lipoma

A Skamnelos ^1,^✉, A Murino ¹, R de Sousa Magalhães ^1,², R Chacchi Cahuin ¹, T Luong ³, R Raymond ¹, N Lazaridis ¹, EJ Despott ¹

Introduction

Small bowel lipomas are benign, usually innocuous lesions which more frequently affect the terminal ileum (TI). They are usually asymptomatic but may give rise to obstructive symptoms and gastrointestinal bleeding, especially if ≥2cm in diameter. Although in such cases, surgery has been the mainstay of care, endoscopic resection is being increasingly employed. Endoscopic submucosal dissection (ESD) is emerging as a safe and definitive endotherapeutic, minimally invasive strategy for such lesions.

Aims & Methods

A 69-year-old man with ischaemic heart disease and cerebrovascular disease was referred to our tertiary institution for endo-scopic resection of a large, symptomatic, intussuscepting lipoma of the TI. At retrograde double-balloon enteroscopy (DBE), under conscious sedation, a 35mm semi-pedunculated submucosal lesion was identified within the TI, about 5cm proximal to the ileocecal valve.

Results

Endoscopic morphology was pathognomonic of a lipoma. The lesion's size and its location within the narrow lumen of the TI precluded safe snare resection and we therefore proceeded with ESD. After injection of submucosal-lifting-solution (succinylated gelatin, indigo carmine (0.05%) and dilute epinephrine (1:100,000)), the stalk was dissected using saline-immersion therapeutic endoscopy (SITE) facilitated ESD with a ball-tipped, needle-type ESD knife (Flush Knife-BT (1.5mm), Fujifilm, Tokyo, Japan). The resection site was clipped prophylactically and the lipoma was retrieved with a Roth” net (Steris, Mentor, OH, USA). No immediate, early or late adverse events occurred. Histopathology confirmed complete resection of an ileal lipoma.

Conclusion

Our case demonstrates the usefulness of this technique, particularly when combined with the safe and stable endotherapeutic platform afforded by DBE and the buoyancy facilitated by SITE. To the best of our knowledge this is only the 3^rd reported case of ESD for a TI lipoma and the only case to date combining the use of SITE-ESD and DBE.

Disclosure

Edward J. Despott has acted as a consultant for Boston Scientific and Ambu. He has also received academic grants and speaker honoraria from Fujifilm, Aquilant Endoscopy, Norgine and Olympus. Alberto Murino has acted as a consultant for Boston Scientific and GI supply He has also received academic grants from Fujifilm, Aquilant Endoscopy, Norgine and Olympus. All other authors have no conflicts of interest to declare.

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.205

VC07 Small Bowel Polypectomy For The Management of Peutz-Jegher Associated Recurrent Intussusception

MS Ismail ^1,^2,^✉, S Semenov ^1,², S Sihag ¹, D McNamara ^1,²

Introduction

Intussusception as a consequence of large small bowel polyps is not uncommon in Peutz-Jegher patients. Prior to Device-Assisted Enteroscopy (DAE), surgery with associated morbidity or conservative management was the only treatment option.

DAE and polypectomy in selected cases can offer a less invasive endoscopic solution.

Aims & Methods

We report the successful case of a patient who required multiple repeated large jejunal polyp resection to control recurrent acute small bowel obstruction episodes as a result of intussusception including a demonstration of polypectomy technique, management of post-polyp-ectomy bleeding and retrieval method.

Both gastrointestinal and systemic surveillance recommendations are also reviewed.

Results

LN,18 year old female, referred with a history of Peutz-Jeghers for small bowel assessment and polypectomy. She presents with ongoing intussusceptions episodes with small bowel obstructions. Background history of small bowel resection due to intussusception from large polyps. She initially underwent capsule endoscopy which showed large polyps in the small bowel, shown in the first section of the video. Subsequently, she underwent a series of DAE (second part of the video), which demonstrate injection techniques for small bowel polyps, retrieval techniques including using Roth Net and bring-your-polyp-for-a-walk technique.

All DAE procedures were well tolerated without any side effects. LN remains asymptomatic with ongoing surveillance organised.

Conclusion

DAE is safe, effective and less invasive when compared to surgical management in patients with large Peutz-Jegher polyps

Disclosure

Nothing to disclose

United European Gastroenterol J. 2020 Oct 10;8(8 Suppl):8–142. doi: 10.1177/2050640620927344.206

VC08 Point Blank - Endoscopic Retrieval of An Extraluminal Bullet

K Gurram ^1,^✉, A Al-khazraji ¹, B Singh ¹, M Ahmed ¹, U Syed ¹, H Boinpally ¹, R Sharma ¹, R Bansal ¹, A Walfish ¹, J Baum ², J Aron ²

Introduction

In 2016, more than 250,000 deaths worldwide were associated with guns.1 90% of mortality associated with gunshot wounds (GSW) is from abdominal trauma.2 The severity of the colon injury dictates the type of surgery that must be performed.3 If diverting colostomy is performed depending on the location (especially deep in the pelvis) closure of the defect can be challenging.3 We demonstrate an endoscopic technique for FB retrieval and closure of defect that is especially useful to prevent recurrent surgery or the original surgery at all.

Aims

Demonstrate the feasibility of safely removing the extraluminal bullet/foreign body (FB) endoscopically. Understand the angle of approach to make the retrieval more feasible. Learn to utilize multiple endoscopic procedural skill sets to achieve the removal of extraluminal FB. Appreciate the different types of accessories necessary to remove the FB. Case History: A 36-year-old male with no past medical history presented with multiple GSW. A computer tomography (CT) scan was performed demonstrating a retained bullet in close proximity of the rectum. The patient underwent diverting colostomy and the rectal defect was not treated. Gastroenterology (GI) was consulted to retrieve the bullet for ballistics and to close the defect.

Methods

A gastroscope (HQ190 Olympus scope) with a distal attachment (12.4mm-Olympus) was introduced into the rectum and irrigation with Gentamicin 150 mg in 500 ml of sterile water was performed. A full-thickness defect was visible. A wire was placed through the defect and the defect was dilated up to 10 mm with a balloon. The bullet was not visualized through the defect and on fluoroscopy (FP) the lesion was not en face. Therefore, the decision was made to take a different approach that was more en face. A linear endoscopic ultrasound (EUS) (GF -UTC 180, Olympus) was utilized to identify the bullet. A 19 gauge fine needle aspiration (FNA) needle was used to make a track to the bullet. A guidewire was left in place and the needle was removed. Sequential balloon dilation was performed from 4mm to 10mm. An insulated tip knife (IT- Nano, Olympus) was utilized to perform a full-thickness resection. The bullet was visible through the defect and was removed with the help of a grasping forceps (Raptor, US Endoscopy). The defects were closed with over the scope clips (OVESCO) with no contrast leak.

Results

Bullet retreived and defect closed.

Conclusion

This demonstrates the feasibility of removing extraluminal FB safely and effectively. The importance of using an en face approach is paramount. EUS guided approach is advantageous in forming the track while directly visualizing the vessels with doppler. The utilization of multiple accessories may be necessary to remove the foreign body. Good knowledge of operating over the scope clips and endoscopic suturing is essential before undertaking these types of procedures.

Disclosure

Nothing to disclose

References

1.The Global Burden of Disease 2016 Injury Collaborators. Global Mortality From Firearms, 1990-2016. JAMA. 2018; 320(8): 792–814. doi: 10.1001/jama.2018.10060 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Evaluation and management of abdominal gunshot wounds: A Western Trauma Association critical decisions algorithm. Martin M.J. et al. J Trauma Acute Care Surg. 2019; 87(5): 1220. [DOI] [PubMed] [Google Scholar]
3.Management of Colorectal Trauma. Won Jun Choi, J Korean Soc Coloproctol 2011; 27(4): 166–173 [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr1-2050640620927344.3] 1.Feagan B. G. et al. Treatment of ulcerative colitis with a humanized antibody to the a4β7 integrin. N. Engl. J. Med. 352, 2499–2507 (2005). [DOI] [PubMed] [Google Scholar]

[bibr2-2050640620927344.3] 2.Parikh A. et al. Vedolizumab for the treatment of active ulcerative colitis: A randomized controlled phase 2 dose-ranging study. Inflamm. Bowel Dis. 18, 1470–1479 (2012). [DOI] [PubMed] [Google Scholar]

[bibr3-2050640620927344.3] 3.Rosario M. et al. Exposure-efficacy Relationships for Vedolizumab Induction Therapy in Patients with Ulcerative Colitis or Crohn's Disease. J. Crohns. Colitis 11, 921–929 (2017). [DOI] [PubMed] [Google Scholar]

[bibr4-2050640620927344.3] 4.Fischer A. et al. Differential effects of a4β7 and GPR15 on homing of effector and regulatory T cells from patients with UC to the inflamed gut in vivo. Gut 65, 1642–1664 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

UEG Week 2020 Oral Presentations

OP001 Identification of Emp2 Gene as a New Therapeutic Target to Limit Aiec Colonization in Cd Patients

M Chervy

N Barnich

J Denizot

Introduction

Aims & Methods

Results

Conclusion

Disclosure

OP002 Host-Genetics, Dysbiosis and Clinical History Explains Fecal Metabolic Alterations in Patients with Inflammatory Bowel Disease

Vila A Vich

S Hu

L Bolte

V Collij

BH Jansen

RAA Ruigrok

G Abu-Ali

C Giallourakis

J Schneider

J Parkinson

A Al Garawi

A Kurilshikov

R Gacesa

J Fu

A Zhernakova

RK Weersma

Introduction

Aims & Methods

Results

Conclusion

Disclosure

OP003 Dose-Dependent Differential Effects of Vedolizumab Therapy on Adhesion of Regulatory and Effector T Cells

E Becker

M Wiendl

M Dedden

A Schulz-Kuhnt

I Atreya

R Atreya

MF Neurath

S Zundler

Introduction

Aims & Methods

Results

Conclusion

Disclosure

References

OP004 Systemic Iron Deficiency and Intestinal Inflammation Control Gut Mucosal Hif-1α and Tfrc Activation in Inflammatory Bowel Disease

AR Bourgonje

Rosa R Fagundes

S Hu

R Barbieri

BH Jansen

T Blokzijl

CT Taylor

KN Faber

G Dijkstra

Introduction

Aims & Methods

Results

Conclusion

Disclosure

OP005 Anti-Inflammatory Actions of Butyrate on Macrophages and Epithelial Cells in Intestinal Mucosa of Inflammatory Bowel Disease Patients

Venegas D Parada

MK De la Fuente

T Blokzijl

Y Cui

D Simian

HJM Harmsen

R Quera

G Dijkstra

KN Faber

MA Hermoso

Introduction

Aims & Methods

Results

Conclusion

Disclosure

Acknowledgments