Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

P472 Risk of relapse after anti-TNF discontinuation in Inflammatory Bowel Disease: A meta-analysis

J.P. Gisbert*, A.C. Marín, M. Chaparro

Hospital Universitario de La Princesa, IIS-IP and CIBERehd, Gastroenterology Unit, Madrid, Spain

Background

The discontinuation of anti-TNF treatment in Crohn's disease (CD) and ulcerative colitis (UC) patients after achieving remission could be considered due to safety and cost issues. However, the convenience of this strategy is presently unknown.

AIM: To assess the risk of relapse after anti-TNF discontinuation, to evaluate the factors that could affect this risk, and to calculate the response to re-treatment with the same anti-TNF after relapsing in CD and UC patients.

Methods

Inclusion criteria: Studies evaluating the incidence of relapse after anti-TNF [infliximab (IFX), adalimumab (ADA), or certolizumab pegol (CZP)] discontinuation in patients with CD and/or UC that reached, at least, clinical remission at the time of withdrawal of anti-TNF. Search strategy: Bibliographical searches in PubMed, Embase and Congresses up to October 2014. Data synthesis: Percentage of relapse after anti-TNF withdrawal; meta-analyses were performed using the inverse variance method. Subanalyses: according to type of disease, type of CD (luminal or fistulizing), anti-TNF drug, and the reason to stop anti-TNF (clinical and/or endoscopic remission).

Results

26 studies were included. Six studies reported mixed results with ADA-IFX, and 20 of IFX. The risk of relapse after anti-TNF discontinuation was 44% for CD (95%CI=36-51%; I2=76%; 23 studies; 720 patients), and 41% for UC patients (30-53%; I2=53%; 9 studies; 159 patients). In CD, the relapse rate at short-term (6 months after the discontinuation of the anti-TNF) was 35%, at medium-term (12 months) 36%, and at long-term (60-125 months) 49%. These results were highly heterogeneous (I2=71-88%). In UC, there were 33% of relapses 12 months after the discontinuation of the anti-TNF (2-65%; I2=93%; 3 studies; 94 patients). In CD, when the anti-TNF was stopped considering only clinical remission, the relapse rate after 1 year was 39%, decreasing to 26% when endoscopic remission was also required. This subanalysis could not be performed in UC due to lack of data. None of the other studied factors were associated with relapse risk. Re-treatment with the same anti-TNF in those patients who relapsed, induced remission in 79% (66-92%; 8 studies; 255 patients) of them; respective figures for CD and UC were 83% (62-100%) and 85% (62-100%)

Conclusion

One-third of both CD and UC patients in remission under anti-TNF treatment relapsed 1 year after discontinuation of the drug. In CD patients, when the criteria for discontinuation was clinical remission, 39% of relapses were reported, but only 26% when patients had also endoscopic remission. Response to retreatment with the same anti-TNF was generally favourable (80% remission) in patients who relapsed after discontinuation.