Search in the Abstract Database

* = Presenting author

OP010 Azathioprine dose reduction in patients with inflammatory bowel disease on combination therapy: a prospective study

E. Del Tedesco1, S. Paul2, H. Marotte3, C. Jarlot1, N. Williet4, J. M. Phelip4, L. Peyrin Biroulet5, J. F. Colombel6, X. Roblin*4

1University Hospital, Gastroenterology, Saint-Etienne, France, 2University of Saint-Etienne, Immunology, Saint-Etienne, France, 3University Hospital, Rheumatology, Saint-Etienne, France, 4CHU Saint-Etienne, Saint-Etienne, France, 5CHU de Nancy, Department of Gastroenterology, Vandoeuvre-lès-nancy, France, 6Mount Sinai Hospital, New York, New York, United States


Combination therapy with infliximab (IFX) and azathioprine (A) is the most effective strategy in patients with Crohn’s disease (CD) and ulcerative colitis (UC) naïve to both therapies, but it has raised safety concerns. The main objective was to investigate whether AZA dose can be reduced to improve the risk/benefit profile of combination therapy.


This prospective study included 3 cohorts of patients with IBD treated for at least 1 year with IFX-AZA and being in deep remission (clinical and endoscopic and/or biomarkers remission) for at least 6 months. All patients had an IFX trough level over 2 µg/ml and were on stable doses of AZA (2 to 2.5 mg/kg/d) and IFX (5 mg/kg every 8 weeks). In cohort A, AZA and IFX were continued unchanged; in cohort B, the dose of AZA was halved, with a minimum dose of 50mg/d; in cohort C, AZA was stopped, and IFX continued as monotherapy. Primary endpoint was failure defined as a clinical relapse (Crohn’s disease activity index [CDAI] > 220 with faecal calprotectin > 450 µg/g stools) and/or need to change the original therapeutic regimen because of adverse events. Trough levels of IFX (TRI) and antibodies (ATI) were measured before each infusion.


Included were 81 patients (45 CD and 36 UC; mean age, 29.7 years; mean disease duration, 24 months) (28 in Cohort A, 27 in Cohort B, and 26 in Cohort C). Five patients (17.8%) in Cohort A, 3 in Cohort B (11.5%),and 8 in Cohort C (30%) experienced failure at 1 year (p = 0.1 across group). Three patients in Cohort A had to stop AZA or reduce the dose because of myelotoxicity or digestive intolerance. In cohort A, the mean TRI concentrations were similar at the time of inclusion (3.65 vs 3.45 µg/ml, respectively). In Cohort B, the mean TRI remained stable after the reduction of AZA dose (3.95 vs 3.6 µg/ml, respectively), whereas there was a significant reduction in the mean 6-TGN levels (310 pmoles vs 128 pmoles, respectively; p = 0 .03) at 1 year, whereas in cohort C, there was a significant reduction in TRI (4.2 vs 2.1 µg/ml; p = 0.02). Four patients in Cohort A (14.2%) had an unfavourable evolution of IFX pharmacokinetic defined by a decrease of TRI < 1 µg/ml or undetectable TRI with positive ATI at 1 year; 5 in Cohort B (18.5%); and 14 in Cohort C (53.8%) (p = 0.01). By ROC analysis (AUROC: O.93), a threshold of 6-TGN < 105 pmoles was associated with unfavourable pharmacokinetic (sensitivity, 67%; specificity, 92%; likelihood ratio, 7.67).


AZA dose reduction in patients with IBD on combination therapy is as effective as the maintenance of AZA at the same dose, and it may improve the AZA safety profile. A threshold of 6-TGN < 105 pmoles was associated with an unfavourable evolution of IFX pharmacokinetics.