The aim of our study is to determine the most valuable marker for early detection of dysplasia in colorectal tissues of patients with IBD. These results will contribute for patient classification into different risk groups regarding closer follow-up. Artificial intelligence is being useful during colonoscopy examination (for diagnosis of IBD, detection and chracterisation of dysplasia, quantifying disease severity and integration with digital pathology), as well as in IBD histologic assessment (in p53 evaluation in particular). 

For investigation of IHC expression of several markers (beta-catenin, VDR, NF-kappaB) in IBD patients, colorectal tissue samples from 30 patients who underwent colonoscopy examination were selected. IHC examinations in the tissue slides of patients with IBD were performed at the "Laboratory for Tumor Pathology" - Medical University of Vienna. 

Of all IHC markers we have analyzed, staining intensity of p53 best correlated with disease activity and intensity of inflammation. Positive nuclear p53 expression was found in 52 % of patients with IBD, of which 32% showed low expression, 15% - medium expression, and 6% - high expression).

Our results are in line with the majority of publications in this field, which consider the protein expression of p53 in dysplastic crypts as a good biomarker and a very valuable diagnostic tool in IBD-associated colon cancers. Traditional diagnostic methods, such as IHC for p53 combined with H&E staining, are widely used to identify colitis-associated dysplasia precisely, but remain subject to inter-observer variability – at this point artificial intelligence offers a promising alternative.