Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

OP005 Infliximab trough level thresholds vary depending on the efficacy criterion chosen in IBD patients

X. Roblin*1, M. Rinaudo2, C. Jarlot1, E. Del Tedesco1, J.M. Phelip1, L. Clavel1, L. Peyrin Biroulet3, S. Paul2

1University Hospital, Gastroenterology, Saint Etienne, France, 2University of Saint Etienne, Immunology, Saint Etienne, France, 3CHU de Nancy, Department of Gastroenterology, Vandoeuvre-les-Nancy, France

Background

Several studies and meta-analyses have shown a correlation between infliximab trough levels (TLI) and clinical remission. Isolated cut-off points vary according to the techniques used. No study has been carried out to determine whether the variations in this therapeutic cut-off depend on the analysis criterion. The aim of our study was to compare whether there were similarities in this threshold for clinical remission (CR) or biomarkers assessment in IBD.

Methods

we included in a monocentric observational study including all consecutive IBD patients treated with IFX between 2010 and 2013. The TLI measurements (ELISA technique, Theradiag) were performed immediately prior to an infusion of IFX (5mg/kg) simultaneously with the CRP (mg/L) and faecal calprotectin (µg/g of stools). The CDAI or partial Mayo scores were reported for all the patients. Exclusion criteria were primary non-responders to IFX, patients previously treated with another anti-TNF agent, patients optimized at a dose other than 5 mg/kg/8W and exclusive anoperineal CD. Clinical remission was defined for CD by a CDAI score of < 150 and for UC by a partial Mayo score of < 3. Normal CRP was defined as CRP < 5 mg/L in patients having a high CRP level at the start of treatment. Faecal calprotectin was defined as normal for a level below 250 µg/g of stools. The TLI cut-off was investigated using a ROC curve analysis to isolate a threshold associated with CR, and normal CRP and calprotectin levels.

Results

213 patients (131 with CD; mean age: 38 years; M:F sex ratio: 0.8; mean duration of IFX: 14.9 months) were included. 145 patients were in CR. Mean TLI were significantly higher during CR (2.6 vs. 1.2 µg/ml; p<0.01). The optimal cut-off associated with CR was 2.1µg/ml (sensitivity: 78%; specificity: 76%). Of the 140 patients who had high CRP levels at the start of treatment, 85 showed normal CRP and clinical remission. The mean TLI were significantly higher when normal CPR levels associated with CR observed (3.5 vs. 1.6 µg/ml; p<0.01). The optimal cut-off associated with normal CRP associated with CR was 2.9 µg/ml (sensitivity: 69%; specificity: 66%). 121 patients showed normal faecal calprotectin associated with clinical remission. The mean TLI were significantly higher in presence of normal faecal calprotectin levels associated with clinical remission (4.9 vs. 1.7 µg/ml; p<0.001). The optimal cut-off associated with normal faecal calprotectin was 3.9 µg/ml (sensitivity: 74%; specificity: 80%).

Conclusion

The TLI to target varies depending on the treatment objective chosen in IBD patients. The therapeutic level required for obtaining deep remission defined by clinical and biomarker remission, is higher than the level required to achieve clinical remission only.