Jurate Valciukiene ^1* Egle Lastauskiene ² Aida Laurinaviciene ³ Matas Jakubauskas ¹ Marius Kryzauskas ¹ Ruta Barbora Valkiuniene ^3,4 Renaldas Augulis ^3,4 Ausra Garnelyte ^3,4 Justinas Kavoliunas ² Ugne Silinskaite ⁵ Tomas Poskus ¹

• ¹ Clinic of Gastroenterology, Nephro Urology and Surgery, Faculty of Medicine, Vilnius University, Vilnius, Lithuania
• ² Institute of Biosciences, Life Sciences Center, Vilnius University, Vilnius, Vilnius, Lithuania
• ³ Department of Pathology, Forensic Medicine and Pharmacology, Institute of Biomedical Sciences, Faculty of Medicine, Vilnius University, Vilnius, Lithuania
• ⁴ National Center of Pathology, Vilnius University Hospital Santaros Clinics, Vilnius, Lithuania
• ⁵ Faculty of Medicine, Vilnius University, Vilnius, Lithuania

The current understanding of colorectal carcinogenesis is based on the adenoma-carcinoma sequence, where genetics, intestinal microbiota changes and local immunity shifts play the key roles. Despite the emerging evidence of dysbiotic intestinal state and immune-cell infiltration changes in patients with colorectal adenocarcinoma, early and advanced adenoma, as well as carcinoma in situ are rather less studied. The newly colon-site adapted AI-based analysis of immune infiltrates is able to predict long-term outcomes of colon carcinoma. Though it could also facilitate the pathologic evaluation of precancerous lesion’s potential to progress. The purpose of this prospective cohort study (MIMICA-1) is, firstly, to identify the intestinal microbiota and immune infiltration patterns around the normal bowel tissue, early and advanced adenoma, carcinoma in situ, and adenocarcinoma, and secondly, to analyze the immune – microbiome interplay along the steps of conventional colorectal tumorigenesis. 40 patients (10 per group) with confirmed colorectal dysplasia undergoing endoscopic polypectomy, endoscopic mucosal resection for colorectal small (≤1cm), and large (>1cm) adenoma or carcinoma in situ, or biopsy and subsequent colon resection for invasive colorectal cancer, and 10 healthy patients undergoing screening colonoscopy will be prospectively recruited. Stool samples will be collected prior to bowel preparation for the analysis of fecal microbiota composition. Biopsy specimens will be taken from the terminal ileum, right colon, left colon, and a pathological lesion in the colon (if present) to assess mucosa-associated microbiota composition and intestinal immunity response. DNA will be extracted from all samples and sequenced using the Illumina MiSeq platform. Unifrac and Bray-Curtis methods will be used to assess microbial diversity. The intestinal immune infiltration will be examined using digital image analysis where primarily immunohistochemistry procedures for CD3, CD8, CD20 and CD68 immune cell markers will be performed. Thereafter, the count, density and distribution of immunocompetent cells in epithelial and stromal tissue compartments will be evaluated using AI-based platform. The interaction between the gut microbiota and immune response in adenoma-carcinoma sequence and the healthy patients will be examined. In addition, fecal samples will be explored for gut microbiota’s composition, comparing fecal- and tissue-derived bacterial patterns in healthy gut and along the adenoma-carcinoma sequence.