P333 Early intestinal ultrasound remission is associated with endoscopic healing after one year of infliximab therapy in Crohn’s disease: a multicentre prospective study

Abstract Background Endoscopic healing (EH) is currently the consensual long-term treatment target to achieve in Crohn’s disease (CD). However, optimization of therapy based on an endoscopic response implies a delay as endoscopy is an invasive procedure and cannot be used longitudinally, stressing the need for non-invasive predictors of EH. We aimed to assess the predictive role of biomarkers and intestinal ultrasound (IUS) parameters after induction therapy to predict EH one year after infliximab (IFX) therapy. Methods Prospective multicentre cohort study including patients with active CD starting IFX. Harvey-Bradshaw index (HBI), C-reactive protein (CRP), faecal calprotectin (FC), and IUS were performed at week (W) 0, 14, 30, and 54. Ileocolonoscopy was performed at W0 and W54. Clinical remission (CR) was defined as HBI<5 and laboratory remission (LR) as CRP <0.5mg/dL and FC <150ug/g. Transmural healing (TH) was defined as the normalization of bowel wall thickness (BWT≤3mm), doppler sign (≤1), wall stratification, and inflammatory fat in the most affected segment. The IBUS-SAS (International Bowel Ultrasound Segmental Activity Score) was used to assess disease activity and includes BWT, doppler sign, loss of stratification, and inflammatory fat. EH was defined as total SES-CD <3. Results We included 52 patients with CD (54% male; mean age 35±14y). Baseline characteristics are available in Table 1. At W14, 89% were in CR, 46% had LR and 19% had TH. After 1 year of IFX therapy, 90% were in CR, 62% had LR, 43% had TH and 37% had EH. At W14, patients with EH had lower BWT (3.3 vs 4.6mm, p=0.01), lower IBUS-SAS (15 vs 61, p<0.01), and lower FC (44 vs 239ug/g, p<0.01). At W14, both TH (OR 11.3, 95%CI 2.1-61.4, p<0.01) and FC remission (OR 7.4, 95%CI 1.8-30.7, p<0.01) were significant predictors for EH. In a multivariate analysis including significant baseline characteristics and early non-invasive monitoring targets at W14 (table 2), TH was an independent predictor of EH (OR 8.6, 95%CI 1.2-61.7, p=0.03). Even after adjusting for the baseline ultrasonographic severity according to the IBUS-SAS, TH at W14 remained a significant predictor of EH (OR 6.5, 95%CI 1.02-40.9, p=0.047). Using a ROC curves analysis, the IBUS-SAS at W14 was the best predictor for EH, with an AUC of 0.81 and the best cut-off value of 32.2 (Table 3). Conclusion Early assessment of ultrasonographic remission at week 14 can help predict EH after one year of therapy and offers an opportunity for early therapy optimization. TH at week 14 assessed by IUS was an independent predictor for EH after one year of therapy. Our results support the use of early IUS monitoring to predict response to therapy in CD patients.