bwge2021

Authors
B. OOSTERLINCK (1), T. BREUGELMANS (1), J. DE MAN (1), K. GEBOES (2), J. KUPČINSKAS (3), A. LINK (4), B. DE WINTER (1), A. SMET (1) / [1] University of Antwerp, , Belgium, Laboratory of Experimental Medicine and Paediatrics, [2] University Hospital Gent, , Belgium, Pathology Department, [3] Lithuanian University of Health Sciences, Kaunas, Lithuania, Department of Gastroenterology and Institute for Digestive Research, [4] Otto von Guericke University Magdeburg, Magdeburg, Germany, Department of Gastroenterology, Hepatology and Infectious Diseases

Introduction
One of the hallmarks of gastric adenocarcinoma is aberrant mucin expression, with gastric- and intestinal-type mucins widely being expressed in gastric tumors. Their clinical importance is still controversial in relation to disease progression, outcome and microbiome composition.

Aim
Here, we investigated mucin expression in human gastric adenocarcinomas and its correlation with disease outcome and bacterial taxa.

Methods
Gastric biopsies of tumour and adjacent tissue of two independent cohorts of Belgian (n=45) and Lithuanian (n=43) patients were analysed for mucin expression. One additional Belgian cohort (n=24) is currently being analysed. Relative expression of gastric (MUC1, MUC5AC and MUC6) and intestinal mucins (MUC2, MUC4 and MUC13) was determined by RT-qPCR. The adenocarcinomas were classified by expression leading to gastric (predominantly MUC5AC and MUC6), intestinal (predominantly intestinal and MUC1), mixed (all types) and unclassified/null (neither gastric nor intestinal) mucin phenotypes. The overexpression threshold was defined as a 0.2*MNE increase (mean normal tissue expression) in the tumour compared to normal tissue. Correlation with 5-year survival (Kaplan-Meijer analysis) and other clinical traits (tumour stadia, histology, age, gender, …) was tested. The gastric microbiome is currently being determined using 16s rRNA sequencing on the Illumina platform. Read quality will be assessed, adapters trimmed and further processed using the DADA2 R package to determine sequence taxonomy and community composition. To detect biologically significant differences between mucin phenotypes regarding their microbial communities a linear discriminant analysis effect size (LEfSe) will be determined at multiple taxonomic levels.

Results
Microbial community analysis for the full cohort and the mucin phenotype analysis for the additional Belgian cohort (n=24) are still ongoing. Based on the preliminary analysis of the Belgian (n=45) and Lithuanian (n=43) cohorts, the cancers were classified as gastric (14.5%), intestinal (25.0%), mixed (17.1%) and unclassified (43.4%) mucin phenotypes. MUC13 overexpression was observed in 55.3% of the cases. Interestingly, 30% of the cohort -intestinal and mixed type- had a MUC13 expression exceeding a 1.75*MNE threshold. This correlated with decreased survival (p= 0.017, log-rank test). For the other mucins tested, this association could not be established.

Conclusions
Our results obtained so far highlight a key role for MUC13 in gastric cancer progression and survival. More research is required to understand its exact mechanism.