P162 RAID-Monitor: a new non-invasive method to determine endoscopic activity in inflammatory bowel diseases

Abstract Background Crohn’s disease (CD) and ulcerative colitis (UC) are characterised by episodes of exacerbations and remissions. Monitoring disease activity based on intestinal lesion is mandatory prior to any change in the therapeutic strategy. Colonoscopy is the gold standard technique to monitor the disease activity in IBD patients, but it is usually discarded because of costs and risk issues. Inflammatory faecal biomarkers such as faecal calprotectin (FC) provide a cheaper and non-invasive alternative methodology. However, FC does not always correlate well with endoscopic indexes. RAID-Monitor is a new tool capable to correlate with endoscopic activity in IBD patients. This test is based on a bacterial signature found in faeces. The aim of this study was to evaluate the performance of RAID-Monitor in front of FC and clinical scores, as a reliable indicator for disease activity in IBD. Methods Two cohorts consisting of 34 patients of CD (considering endoscopy activity SES-CD ≥3, 14 active and 20 in remission) and 43 of UC (considering endoscopy activity Mayo >1, 19 active and 24 in remission) are recruited by the Gastroenterology department from four Catalan hospitals. Clinical scores, Harvey–Bradshaw Index (HBI) for CD and Mayo Partial Index (MPI), and a stool sample, to determine FC and RAID-Monitor, are collected prior to colonoscopy. Results RAID-Monitor differentiates the endoscopic activity with sensitivity and specificity values up to 85.7% and 95.0%, respectively, in CD patients. It obtained better results compared with the best results of FC (obtained at cut-off: 200 µg/g). FC displays the same sensitivity (85.7%) but lower specificity values (80.0%). Instead, HBI obtains the worst values of sensitivity and specificity (42.9% and 75.0%, respectively). RAID-Monitor allows a substantial increase of Positive Predictive Value (PPV) (92.3% vs. 76.9%, respectively) and Negative Predictive Value (NPV) in comparison with FC (90.5% vs. 88.9%, respectively). In UC patients, RAID-Monitor displays higher sensitivity and specificity (94.7% and 91.7%, respectively) as compared with FC using the best cut-off at 350 µg/g (73.7% and 70.8%, respectively). MPI obtains a low sensitivity (57.9%) but a similar specificity (91.7%). PPV and NPV (90.0% and 95.6%, respectively) are higher than those obtained with FC (66.7% and 77.3%, respectively). Conclusions RAID-Monitor is an accurate bacterial-based biomarker that correlates well with endoscopic activity in both