Search in the Abstract Database

Search Abstracts 2016

* = Presenting author

P582 Predictive value of infliximab trough levels in quiescent inflammatory bowel disease under maintenance infliximab therapy

M. Bejaoui*, F. Charbit-Henrion, C. Landman, H. Sokol, A. Bourrier, I. Nion-Larmurier, J. Cosnes, L. Beaugerie, P. Seksik

Saint Antoine Hospital, Gastroenterology & Nutrition, Paris, France

Background

Measurement of trough levels and detection of anti-drug antibodies have gained importance in the management of loss of response (LOR) in inflammatory bowel diseases (IBD) patients under anti-tumour necrosis factor (TNF)-alpha therapy. Nevertheless, their role in quiescent IBD is still debated. The aim of this study was to assess predictive value of infliximab trough levels in quiescent IBD under maintenance infliximab therapy.

Methods

We conducted a single-centre prospective study enrolling all consecutive IBD patients in clinical remission under maintenance infliximab therapy between January and March 2011. Clinical remission was defined as CDAI < 150 in Crohn’s disease (CD) and as a simplified Mayo score < 3 in ulcerative colitis (UC). Trough levels of Infliximab in serum (TLI) and antibodies to Infliximab (ATI) were assessed just before infliximab infusion. Loss of response (LOR) was pragmatically defined by the occurrence of either a therapy optimisation, a switch to another biologic, an intestinal surgery, or an IBD complication. Patients were prospectively monitored between January 2011 and March 2014. C-reactive protein (CRP) levels were collected, and CDAI or simplified Mayo score were calculated at each infliximab infusion. Univariate and multivariate analysis using Cox model were performed to identify predictive factors of LOR.

Results

In total, 66 patients with quiescent IBD (29 CD and 37 UC) were included. Average age was 35.5 years [17–86] and sex ratio M / F was 0.94. At baseline, 32 patients had TLI < 3 µg/ml (IFX- group), and 34 had TLI > 3 µg/ml (IFX+ group). The 2 groups were similar regarding age at IBD onset, gender distribution, smoker status, and IBD clinical features according to the Montreal Classification and concomitant immunosuppressive therapy. ATI were found in 5 (15.1%) patients from the IFX- group but in none of the IFX+ group. LOR was significantly higher in IFX- group when compared with the IFX+ (51.7% vs 22.2%; p = 0.02). Rates of complications, therapy optimisation and surgery requirements were similar in the 2 groups. Multivariate analysis showed that 2 factors were associated with LOR within 3 years: TLI < 3 µg/ ml at baseline (OR = 3.67; IC 95% [1.39–9.69]; p = 0.009) and perianal lesions at baseline (OR = 3; IC 95% [1.20–2.46]; p = 0.018).

Conclusion

This study suggests that low trough levels of infliximab (< 3 µg / ml) in patients with IBD in clinical remission under infliximab maintenance therapy are predictive to a LOR in the medium term.