Résumé : Background Adalimumab (ADM) has been shown efficacious in ulcerative colitis (UC). In randomized controlled trials, dose escalation from 40 mg ADM every other week to 40 mg every week was required in 20%-25% of patients within 1 year. Real-life data suggest higher escalation rates. Attempts for dose de-escalation have not been studied yet. We assessed the need for, outcome of, and predictors of dose escalation and de-escalation in a large retrospective cohort of UC patients treated with ADM. Methods We included 231 consecutive patients from 10 Belgian centers initiating ADM treatment for active UC before September 1, 2015 (follow-up ≥1 year in each patient). We performed detailed chart review to identify variables associated with short-term clinical benefit (based on physician global assessment and absence of rectal bleeding at week 10), success of dose escalation, and dose de-escalation. Backward Cox regression and Wald Logistic regression were used to identify predictive variables. Results Short-term clinical benefit was achieved in 101 patients (44%) and was less frequent in infliximab failures [37% vs 50%, Odds ratio 0.57 (95% CI 0.34-0.97), P = 0.038]. After a median of 2.8 (1.7-5.1) months, 164 patients (71%) needed ADM discontinuation (n = 35, 15%) or dose escalation (n = 129, 56%). Dose escalation was successful in 77/129 (60%). Dose de-escalation was attempted in 71% (55/77) after a median of 4.3 (2.9-7.2) months and was successful in 80% (43/54). Conclusions In this cohort, 56% of patients with UC required ADM dose escalation with a 60% success rate. Of note, most patients could be successfully de-escalated later on.