P307 Study of the proteomic profile in patients with ulcerative colitis, its correlation with diagnosis and disease activity

• Published in: Journal of Crohn's and colitis 2018-01, Vol.12 (supplement_1), p.S256-S256
• Main Authors: Ahmed, E, Badr El-Din, S, Header, D, Moez, P, Ibrahim, M
• Format: Article
• Language: English
• ISSN: 1873-9946; 1876-4479
• DOI: 10.1093/ecco-jcc/jjx180.434

Abstract Background Inflammatory bowel disease (IBD) comprises primarily 2 disorders: ulcerative colitis (UC) and Crohn’s disease (CD). The hallmark of IBD is chronic, uncontrolled inflammation of the intestinal mucosa. The diagnosis is based on a combination of disease history, colonscopy, inflammatory biomarkers, radiological and histological evaluation. Most biomarkers used are not reliable and not disease specific, but reflect generalised inflammation. The Aim of the work is to identify plasma proteomic profiles of UC cases and correlating this profile with the other diagnostic markers and activity of the disease Methods We performed a study with 70 plasma samples collected from patients classified in 2 groups (37 UC, 33 healthy controls) according to accredited criteria. They were subjected to: complete history taking, thorough clinical examination, Laboratory investigations (Erythrocyte sedimentation rate ESR, C- reactive protein CRP, fecal calprotectin, Perinuclear antineutrophil cytoplasmic antibodies P-ANCA), ileocolonscopy, histopathology, imaging were done. Plasma proteomic pattern of UC patients and control subjects was determined using matrix-assisted laser desorption/ionisation (MALDI) time of flight (TOF) mass spectrometer (MS) analysis, all plasma samples were subjected to solid-phase extraction (SPE). We analysed the spectra obtained from all the samples using ClinProTool software. Results There was a statistical significant difference of the plasma proteome profiles of UC group in comparison to health volunteers. Sixty-four signals were identified by the ClinProt software and of these16 peptide peaks were highly significant (sensitivity was 100%, specificity was 84.4%, Positive Predictive Value was 100, Negative Predictive Value was 82). There was a statistical significant difference between active vs. inactive UC group, 5 Integration Regions used for classification between active and inactive UC patients using Genetic Algorithm model (GA) which gave 88.5% cross validation and 100% recognition capability. Markers as ESR, CRP, fecal calprotectin, P-ANCA are statistically correlated to the plasma proteomics found in UC patients. Conclusions Proteomic profile has the potential to improve diagnosis and evaluate UC activity, reducing the need for more invasive techniques. The pattern of these peptides holds the promise of distinguishing disease states and providing clinically important information such as prognosis, response to therapy, or per