Search in the Abstract Database

Search Abstracts 2016

* = Presenting author

P405 Infliximab trough level measured during treatment induction may be predictive of the loss of response to Infliximab during treatment maintenance in inflammatory bowel disease patients: a longitudinal observational retrospective study

C. Liefferinckx*1, C. Minsart2, A. Cremer1, L. Amininejad1, J.-F. Toubeau3, E. Quertinmont2, J. Deviere1, A. Van Gossum1, D. Franchimont1

1Erasme Hospital, Gastro-enterology, Brussels, Belgium, 2Université libre de Bruxelles, Laboratory of Experimental Gastroenterology, Brussels, Belgium, 3Université de Mons, Department of Electrical Engineering, Mons, Belgium

Background

Infliximab (IFX), is indicated for the treatment of inflammatory bowel disease (IBD). Nevertheless, loss of response (LOR) to IFX is reported in up to 20%–30% over 12 months of treatment. Importantly, proactive monitoring of IFX pharmacokinetics may help prevent LOR during treatment maintenance. Our objective is to analyse in a cohort of 263 patients the pharmacokinetics early on to predict LOR during maintenance treatment.

Methods

In total, 263 IBD patients have been treated with IFX on follow-up (median + range), (EC approved). Further, 2 331 samples were prospectively collected and measured retrospectively by ELISA in parallel with clinical data. Statistical analysis used Mann–Whitney test.

Results

During the maintenance, the median IFX trough level (TL) was statistically higher in patients on combo-therapy (IFX combined with immunomodulator, n = 107) compared with patients on monotherapy (n = 81) (1.32 µg/mL vs 2.14 µg/mL, p < 0.000001). Median IFX TL was higher under combo when patients were sequentially treated first with combo then mono (4.17 µg/mL vs 3.22 µg/mL, p < 0.01, n = 55). On the contrary, there was no statistical difference between mono- and combo-therapy when patients were sequentially treated first with mono- then combo-therapy (2.02 µg/mL vs 2.49 µg/mL, p < 0.17, n = 49). During maintenance, 19% of the patients (n = 49) experienced LOR, defined as secondary non-responders (NRII), requiring treatment optimisation by either shortening of the dosing interval and/or by increasing the dose. Moreover, 57% of these patients (n = 28), defined as secondary non-responders end (NRIIend), did not respond to any optimisation strategy and were switched to another treatment, whereas 43% of patients (n = 21), defined as secondary responders to optimisation (RIIopt), responded to optimisation. Looking at the TLs during induction, median IFX TL at induction (week 2 and 6) was not statistically different in NRII (n = 49) compared with patients with sustained response (n = 170) (6.66 µg/mL vs 11.91 µg/mL, p < 0.08).

Figure 1. The box plot represents the median IFX TL in induction phase (week 2 and 6) for patients with sustained response (n=170) and NRII patients (n=49). The interquartile range (25-75%) is represented by rectangle. The whiskers below and above the box represent the limits (2.5% and 97.5%) of the distribution. Median IFX TL is represented by the thick red line. Median IFX TL for NRII patients was 6.66 μg/ml [0.23-19.93μg/ml] and median IFX TL for patients with sustained response was 11.91 μg/ml [0.11-19.93 μg/ml]. There is no statistical difference (p <0.08).

However, median IFX TL at induction was statistically lower in NRIIend (n = 23) compared with the patients with sustained response (1.69 VS 11.91 µg/ml, p <0.05)

Conclusion

This study suggests that patients who do not respond to any optimisation strategy (NRIIend) seem to have lower IFX TLs at induction. IFX TLs measured at induction may predict clinical response to IFX during maintenance.