P230 Prediction of loss of response to anti-TNF therapy using SES-CD in patients with Crohn’s disease

Biological therapies enable to set treatment target as mucosal healing in Crohn’s disease (CD) patients. When mucosal healing could be achieved, it has been indicated that loss of response (LOR) rate is significantly decreased. However, the definition of mucosal healing varies widely because of the complexity of previously reported endoscopic scoring systems. Among them, simple endoscopic score for Crohn’s disease (SES-CD) seems applicable in daily clinical practice for the assessment of mucosal healing. We thus evaluated clinical usefulness of SES-CD for predicting LOR to anti-TNF therapy in CD patients. We retrospectively investigated clinical data of 99 CD patients with ileocolitis or colitis type, who were treated by either infliximab (IFX) or adalimumab (ADA) from January 2003 to September 2018. We excluded 61 patients based on the exclusion criteria, including insufficient clinical data, history of intestinal surgery, primary non-response to IFX /ADA, and intolerance to IFX/ADA. We thus included remaining 38 patients with induction of clinical remission whose clinical course could be followed up for more than a year after ileocolonoscopy. We then analysed possible risk factors associated with subsequent LOR to IFX/ADA. As for mucosal healing, two types of definition were set in the present study; one to be ≤10 based on SES-CD, and the other to be the absence of ulceration ≤ 5 mm in size. The cumulative risk of LOR was calculated by Kaplan–Meier method. Risk factors associated with LOR were examined by univariate and multi-variate analyses using Cox proportional hazard model. Median duration from IFX/ADA initiation to endoscopic evaluation was 13 months (range: 1–105 months). A significantly higher rate of LOR was observed in patients with SES-CD of >10 than in those with SES-CD of ≤10 (p = 0.0032). However, no difference was observed between patients with ulceration (>5 mm) and those without ulceration with respect to LOR rate (p = 0.50). Under multi-variate analysis, duration from IFX/ADA initiation to endoscopic evaluation <5 month (p = 0.0016), serum albumin < 4.2 g/dl (p = 0.0074), and SES-CD >10 (p = 0.014) were the factors associated with the risk of LOR to IFX/ADA. Although further prospective studies with a larger number of cases are necessary, SES-CD can be useful for predicting LOR to anti-TNF therapy.