P739 Sequential treatment is effective to achieve remission and avoid colectomy in moderate-to-severe Ulcerative Colitis refractory to anti-TNF therapy

Farkas, B.(1)*;Borsos, M.(2);Resál, T.(1);Bacsur, P.(1);Bor, R.(1);Bálint, A.(1);Fábián, A.(1);Szepes, Z.(1);Molnár, T.(1);Farkas, K.(1);

(1)Albert Szent-Györgyi Medical School- University of Szeged, Department of Internal Medicine, Szeged, Hungary;(2)AdWare Research, Department of Biostatistics, Balatonfüred, Hungary;


The development of anti-TNF agents has brought advances in the treatment of ulcerative colitis (UC). However, many patients require biological treatment sequencing. We aimed to evaluate clinical outcomes, the rate of treatment persistence and colectomy-free survival at second-, third-, and fourth-line therapy after switching from TNF-inhibitor to another treatment and to identify factors predicting the need for colectomy.


In this retrospective study, we included active UC patients refractory to at least one anti-TNF agent and needed treatment sequencing. The efficacy of therapies of the different lines was assessed by endoscopic or clinical scoring systems and laboratory markers. Kaplan-Meier analysis was performed to investigate the colectomy-free survival. Possible predictive factors were examined by logistic regression.


Data of 90 UC patients were analyzed. The most frequently used first-line biologic therapy was infliximab (83.3%), which was discontinued after an average of 21.3 months. In the second-line vedolizumab (VDZ) was administered in 48.9% of cases, adalimumab in 36.7%. Average time between initiation and discontinuation was 15.7 months. In the third line, VDZ was used in 50%, tofacitinib in 29% of the patients, until an average of 8.7 months. In the fourth line, 53.1% of patients received tofacitinib and it was terminated after an average of 8.9 months. Loss of response, as a reason for treatment discontinuation was more common in the first-, and second-line therapies (58.9% and 46.2%), while primary non-response was more often observed in third-, and fourth-line therapies (52.6% and 63.6%). Fifteen patients underwent colectomy (16.7%). There was a significant increase in the probability of having surgery within 2 years after therapy initiation (p=0.0239) independently from treatment lines. Presence of deep ulcer showed inverse correlation with colectomy-free survival (p=0.0189; OR: 10.414 (95% CI: (1.472; 73.669)). Patients with high partial Mayo score at therapy initiation were more likely to undergo colectomy (p=0.0436, OR: 1.463 (95% CI: (1.011; 2.118)). In contrast, gender, age and disease extent at the time of the diagnosis, and the use of concomitant immunmodulators were not shown to be significant in predicting the subsequent surgical need.


Our results suggest that the remission rate does not decrease considerably as the sequential treatment progresses, more than half of the patients receiving fourth-line therapy were still in sustained remission. The more severe disease activity and deep ulcerations proved to be predictive for colectomy. Further studies in a larger patient population would be required to confirm our data and reveal predictive markers.