P181 A retrospective analysis of the clinical application of the multigene analysis test DiBiCol® to differentiate between ulcerative colitis and Crohn’s disease and Non-IBD

• Published in: Journal of Crohn's and colitis Vol. 12; no. supplement_1; p. S190
• Main Authors: Admyre, C, Knittel, T, Ost, A, Lofberg, R, Zargari, A
• Format: Journal Article
• Language: English
• Published: UK Oxford University Press 16.01.2018
• ISSN: 1873-9946; 1876-4479
• DOI: 10.1093/ecco-jcc/jjx180.308

Abstract Background DiBiCol® is a diagnostic test that can differentiate between ulcerative colitis (UC), Crohn’s disease (CD) and non-inflammatory bowel disease (Non-IBD). Since 2009 the DiBiCol® test has been used in Sweden in clinical routine practice at a large number of clinics. In this retrospective analysis the clinical application and the outcome of the DiBiCol® test was studied in 1110 patients. Methods DiBiCol® is a PCR-based method that monitors seven proprietary biomarkers from a single colonic biopsy. By applying a specially designed algorithm, a diagnosis of UC, CD and Non-IBD can be made with a high degree of certainty. The algorithm was established in three clinical studies including more than 300 patients and since 2009 a total of more than 1200 colonic biopsy samples were analysed. Results Based on the data compiled from over 30 IBD units in Sweden over 2 years the causes for requesting the DiBiCol® test were in 52% cases of unclassified IBD, in 16% patients presenting IBD like symptoms for the first time, in 13% cases of indeterminate colitis, in 8% cases before surgery to confirm diagnosis, in 6% cases of changing symptoms and in 5% other reasons. In 994 out of 1110 samples a conclusive diagnosis with more than 70% probability could be made, UC in 492 cases, CD in 431 cases and 71 cases of Non-IBD, while in 116 cases the test result was inconclusive for different reasons. Duplicate colonic biopsy samples taken at the same time were available in 300 patients and provided the same test results in 83% of the cases. In 52/300 patients the test result was different between the 2 biopsies but only in 24 patients a different diagnosis with ≥90%probability was evident meaning a clearly different test results only in 8% of cases (24/300). In 34 patients DiBiCol® was tested at two-time points four weeks apart and in 94% of cases the test results were in alignment with each other. In a subset of 79 patients the clinical and histological diagnosis were compared with the DiBiCol® test results and the data showed that 93% of these cases, in which the clinical diagnosis matched with histopathology, were correctly diagnosed by DiBiCol® with ≥70% probability. The sensitivity / specificity of the DiBiCol® test was calculated to be 97% / 92% for UC diagnosis and 78% / 88% for CD diagnosis. There was a better correlation between the DiBiCol® result and the histopathology evaluation than the clinical diagnosis in this subset. Conclusions Real-world data on the use of DiBiCol® demonstrate that this multigene analysis test is a useful diagnostic tool and can support physicians to diagnose IBD and to differentiate CD from UC.