Y-ECCO Literature Review: Sandra Bohn Thomsen

Sandra Bohn Thomsen

27 Sandra Thomsen 8867 2
Sandra Bohn Thomsen 
© Sandra Bohn Thomsen

Evolution after anti-TNF discontinuation in patients with Inflammatory Bowel Disease: A multicentre long-term follow-up study

Casanova MJ, Chaparro M, García-Sánchez V, Nantes O, Leo E, Rojas-Feria M, et al.
Am J Gastroenterol. 2017;112:12031



Although anti-tumour necrosis factor-α (anti-TNFα) has been in use for Inflammatory Bowel Diseases (IBD) since the 1990s, solid knowledge on the optimal duration of treatment is sparse. A recent meta-analysis, comprising 1150 patients, demonstrated overall relapse rates of 38% for Ulcerative Colitis (UC) and 44% for Crohn’s Disease (CD) after withdrawal of anti-TNFα; for CD the risk was 36% after 12 months, 43% after 16–24 months and 49% ≥ 25 months, whereas for UC the risk was 28% at 12 months and 36% after 16-24 months (no long term follow-up have been assessed) [1]. It would be of great value to be able to predict which patients will achieve remission after withdrawal of anti-TNFα therapy. However, data on predictors of relapse are still insufficient [2]. In this large retrospective multicentre study, the incidence of relapse after discontinuation of anti-TNFα therapy and the predictors of relapse were assessed.

Key findings

A total of 1055 IBD patients were included from 78 Spanish centres. Patients were in clinical remission at the time of withdrawal of anti-TNFα: Harvey Bradshaw Index ≤4 or Partial Mayo Score ≤2. The scores were retrospectively estimated from patient records. The patients were followed up for a minimum of 6 months after discontinuation. The overall incidence of relapse after withdrawal was 44%: 4% at 1 year, 38% at 2 years, 46% at 3 years and 56% at 5 years. Among CD patients, those who discontinued anti-TNFα following an “elective decision” (taken by the patient or physician) were at higher risk of a relapse than those who stopped anti-TNFα after an early combination of anti-TNFα and thiopurine, a “top-down” strategy [hazard ratio (HR)=1.90; 95% confidence interval (CI)=1.07–3.37]. Likewise, discontinuation of anti-TNFα in CD patients because of adverse events increased the risk of relapse compared to the “top-down” strategy (HR=2.33; 95% CI=1.27–2.02). Treatment with adalimumab (ADA) before discontinuation also increased the risk of relapse in CD patients as compared to treatment with infliximab (IFX) (HR=1.29; 95% CI=1.01–1.66). Colonic localisation and structuring disease were further predictors of risk of relapse. Treatment with immunomodulators and older age protected against relapse in CD patients after discontinuation of anti-TNFα (HR=0.67; 95% CI= 0.51–0.87 and HR=0.98; 95% CI=0.97–0.99, respectively). None of these predictors of or protectors against relapse could be demonstrated for UC. Likewise, a sub-analysis of 460 patients with either CD or UC in endoscopic/radiologic remission did not show any predictors of or protectors against relapse. Sixty-nine percent of patients who relapsed were retreated with the same anti-TNFα and 75% of them achieved clinical remission once again. Only 1% underwent surgery.


This study is noteworthy because of the large number of patients included. However, caution must be exercised when interpreting the results, given the retrospective design whereby clinical scores were estimated from patient records. The demonstration that half of the patients sustained remission during a 5-year period underlines previous findings indicating that discontinuation of anti-TNFα is very relevant [1]. However, a combined assessment of risk factors must be taken into consideration for every patient. The study demonstrated that caution must be exercised in the case of young CD patients with colonic localisation or stricturing disease and that CD patients receiving ADA treatment may be at higher risk of a relapse. However, if a relapse occurs, retreatment with anti-TNFα is likely to be successful.

Importantly, the study demonstrated that immunomodulators play a pivotal part in improving outcomes after anti-TNFα withdrawal. Previous studies have failed to demonstrate a beneficial effect of immunomodulators, likely because of insufficient cohort sizes [3,4]. Another explanation may be sub-optimal thiopurine metabolite levels [5], which were unaccounted for in Casanova et al.’s study. In our own prospective study on patients with CD in clinical remission and optimised thiopurine metabolite levels, relapse rates as low as 10% were achieved 1 year after the withdrawal of anti-TNFα therapy [6].

In addition to the importance of immunomodulators, Casanova et al.’s study demonstrated that the “top-down” strategy, using early anti-TNFα and thiopurines, improved outcomes after withdrawal. However, it was not evaluated whether this strategy increased mucosal healing, as seen in the SONIC trial [7]. A part of the explanation for the additive efficacy of immunomodulator/anti-TNFα co-therapy is the prevention of anti-drug antibody formation [8] but more is still to be learned regarding the drug interactions and long-term effects. Prospective studies are needed to investigate the role of immunomodulators during co–therapy and after anti-TNFα withdrawal.

Overall, more studies are warranted in the field of anti-TNFα withdrawal. A randomised multicentre trial with a placebo group is still lacking. The STOP-IT study is a Danish/Scandinavian multicentre study randomising CD patients in deep remission to continue infliximab or placebo. The results of this study are highly anticipated. Studies investigating the predictive value of protectors against disease progression, including deep remission, histological mucosal healing and optimised thiopurine metabolite levels, would also be of great value. 


1. Gisbert JP, Marín AC, Chaparro M. The risk of relapse after anti-TNF discontinuation in inflammatory bowel disease: Systematic review and meta-analysis. Am J Gastroenterol. 2016;111:632–47.

2. Gisbert JP, Marín AC, Chaparro M. Systematic review: factors associated with relapse of inflammatory bowel disease after discontinuation of anti-TNF therapy. Aliment Pharmacol Ther. 2015;42:391–405.

3. Steenholdt C, Molazahi A, Ainsworth MA, et al. Outcome after discontinuation of infliximab in patients with inflammatory bowel disease in clinical remission: an observational Danish single center study. Scand J Gastroenterol. 2012;47:518–27.

4. Kennedy NA, Warner B, Johnston EL, et al. Relapse after withdrawal from anti-TNF therapy for inflammatory bowel disease: an observational study, plus systematic review and meta-analysis. Aliment Pharmacol Ther. 2016;43:910–23. 

5. Moreau AC, Paul S, Del Tedesco E, et al. Association between 6-thioguanine nucleotides levels and clinical remission in inflammatory disease. Inflamm Bowel Dis. 2014;20:464–71.

6. United European Gastroenterology Journal; 2015: 2 (Supplement 1) 

7. Colombel JF, Sandborn WJ, Reinisch W, et al. Infliximab, azathioprine, or combination therapy for Crohn’s disease. N Engl J Med. 2010;362:1383–95.

8. Roblin X, Boschetti G, Williet N, et al. Azathioprine dose reduction in inflammatory bowel disease patients on combination therapy: an open-label, prospective and randomised clinical trial. Aliment Pharmacol Ther. 2017;46:142–149. https://doi.org/10.1111/apt.14106


Sandra Bohn Thomsen is currently undertaking a PhD at Copenhagen University Hospital Hvidovre, Copenhagen, leading a national RCT testing azathioprine–allopurinol co-therapy. At the moment she is finalising a prospective study on withdrawal of anti-TNFα therapy in Crohn’s patients with optimised thiopurine metabolite levels. 27 Sandra Thomsen 8867 2
Sandra Bohn Thomsen 
© Sandra Bohn Thomsen

Posted in Y-ECCO Literature Reviews, ECCO News, Committee News, Y-ECCO, Volume 12, Issue 4