P513 Early initiation of anti-TNF is associated with favourable long-term outcome in Crohn’s disease: Ten-year follow-up data from the Swiss IBD cohort study

Abstract

Little is known about the optimal timing of treatment escalation in Crohn’s disease (CD) and the long-term evolution in early vs. late administration of anti-TNF. We comparatively assessed the long-term outcome in Swiss CD with an up to 10-year follow-up in the Swiss Inflammatory Bowel Disease Cohort Study (SIBDCS). Prospectively collected SIBDCS data were retrospectively analysed in early vs. late (i.e. initiation within 24 months after diagnosis, or thereafter) vs. never anti-TNF-treated patients. Amongst a total of 1705 CD patients, 942 patients received at least one dose of anti-TNF treatment, 246 (26.1%), and 696 (73.9%) early and late after CD diagnosis. While age at diagnosis, smoking medication use (including immunosuppressants) and sex were similar, disease duration prior to inclusion within SIBDCS (median: 5, 16, and 13 years, respectively) and interval since last flare (median: 12, 20, and 26 months, respectively) were significantly shorter in early vs. late and never anti-TNF-treated patients. We did not observe any difference in need of perianal surgery after initiation of anti-TNF in early vs. late treated patients. However, we found a reduced risk of developing stenosis in early anti-TNF-treated patients. Kaplan–Meier analysis and log-rank test for freedom from stenosis according to early and late administration of anti-TNF after diagnosis as well as presence or absence of previous complications. This association was seen in patients overall and in the subgroups of CD patients with or without previous complications (log-rank test: p < 0.001). Furthermore, osteoporosis and anaemia were significantly less frequent in early anti-TNF-treated patients, compared with both, patients with late (p < 0.001 and p = 0.046, respectively) or no (p < 0.001 for both) treated with anti-TNF. In addition, patients with early anti-TNF administration were significantly less often seeking medical consultation, including with a gastroenterologist (p = 0.017), ambulatory hospital visits (p = 0.038) and a composite of any medical visit at all (p = 0.001), while hospitalisation rates were not different between early and late treatment (but, not unexpectedly, higher than in patients never treated with anti-TNF). Frequency of medical consultations within the last 3 months is depicted in the three treatment groups. Despite the non-randomised design of our prospective cohort study with high likelihood of selection bias towards milder in the newer and more severe in the early treated patients, we identified early anti-TNF administration to be associated with a favourable long-term outcome in CD patients within our Swiss cohort study.