P203 Potential, novel serum biomarkers in ulcerative colitis
P. Kourkoulis*1, G. Michalopoulos1, H. Katifelis2, I. Papaconstantinou3, G. Karamanolis4, M. Gazouli2
1Tzaneion General Hospital of Piraeus, Department of Gastroenterology, Piraeus, Greece, 2School of Medicine, National and Kapodistrian University of Athens, Department of Basic Medical Sciences, Laboratory of Biology, Athens, Greece, 3School of Medicine, National and Kapodistrian University of Athens, 2 Department of Surgery, Athens, Greece, 4School of Medicine, National and Kapodistrian University of Athens, Gastroenterology Unit, 2 Department of Surgery, Athens, Greece
Despite the advantages in the management of ulcerative colitis (UC), much less have been achieved in the field of diagnosis and monitoring of the disease, where colonoscopy remains the ‘golden’ method. Established serum biomarkers while commonly used, their poor correlation with the endoscopic features and poor performance as screening tools renders them as inadequate biomarkers by themselves. Therefore, the development of novel, objective, reproducible biomarkers with good correlation with disease endoscopic activity would be of great value for the diagnosis and monitoring of UC. The objective of our study was to evaluate the correlation between leucine-rich Α-2 glycoprotein (LRG), high mobility group box 1 protein (HMGB1), Annexin A1 (ANXA1) and matrix metalloproteinase 3 (MMP3) with endoscopic activity and their role as potential serum biomarkers of UC.
Patients with UC, treated with 5-ASA undergoing colonoscopy, were selectively included in our study. Individuals undergoing preventive colonoscopy with no abnormal endoscopic features were also included as control group. A blood sample was obtained from each member of both groups and endoscopic Mayo subscore (Ms) was recorded for the UC patients. Serum LRG, HMGB1, ANXA1, and MMP3 levels were measured in the blood samples. Statistical analysis (Independent-samples t-test) was performed to compare the data collected and ROC curve analysis for the statistically significant differences recorded.
Forty-two UC patients and fourteen controls were included. The patients’ and controls’ median age was 48 and 54 years old, respectively. While there were no statistically significant differences reported for HMGB1 and LRG, different results were recorded for ANXA1 and MMP3 as shown in the following table.
ANXA1 levels were significantly different between controls and UC patients implying that it could be used as a marker for diagnosis of UC. The best cut-off value was 2.043 μg/ml (88% sensitivity, 93% specificity). MMP3 was significantly lower for Ms = 0, Ms = 0/1 vs. Ms = 1, Ms = 2/3, respectively, suggesting that it could be a marker of mucosal healing and endoscopic remission. The best cut-off values were 4.743 ng/ml for Ms = 0 vs. Ms = 1 (100% sensitivity, 67% specificity) and 6.58 ng/ml for Ms = 0/1 vs. Ms = 2/3 (89% sensitivity, 61% specificity).