Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

DOP075 Persistent dysregulated colonic mucosal gene expression in ulcerative colitis patients with endoscopic healing after infliximab or vedolizumab therapy

I. Arijs*1, G. De Hertogh2, L. Van Lommel3, K. Machiels1, J. Van der Goten1, M. Ferrante1, F. Schuit3, G. Van Assche1, P. Rutgeerts1, S. Vermeire1

1KU Leuven, Clinical and Experimental Medicine, Leuven, Belgium, 2KU Leuven, Imaging & Pathology, Leuven, Belgium, 3KU Leuven, Cellular and Molecular Medicine, Leuven, Belgium

Background

Mucosal healing on endoscopy is considered the treatment goal in inflammatory bowel diseases, including ulcerative colitis (UC). However, endoscopic healing is no cure and relapses are still observed in the majority of patients when treatment is discontinued in these patients. As the triggers for recurrence of inflammation are unknown, we studied if mucosal gene expression profiling could identify pathways important in relapse. We, therefore, compared colonic mucosal gene expression profiles of UC patients achieving endoscopic mucosal healing induced by infliximab (IFX) or vedolizumab (VDZ) therapy to these of healthy controls.

Methods

Colonic biopsies from 23 UC patients before and W4-6 after first IFX, from 44 UC patients before and W6, W12 and W52 after VDZ (GEMINI I and LTS), and from 12 non-IBD healthy controls were studied. Endoscopic mucosal healing was assessed at the same time points as the biopsies. Total RNA from biopsies was analyzed for whole genome gene expression via Affymetrix Human Gene 1.0 ST arrays (false discovery rate< 5% and >2-fold).

Results

In VDZ healers, there were no significant gene probe sets different at W6 and only 5 (down: IDO1, REG3A, KLK6, SAA2 and up: PCK1) at W12 when compared to W0. As many as 593 (462 down/131 up) gene probe sets were significantly different in VDZ healers at W52 vs. W0. The majority of the observed changes at W52 by VDZ encoded genes that were involved in immune cell trafficking, cellular movement and inflammatory response, and overlapped (63%) with the probe sets identified in IFX healers at W4-6 vs. W0 [481 (388 down/93 up) significant probe sets] probe sets. After therapy at each of the studied time points, many gene probe sets remained significantly dysregulated in the IFX and VDZ healers when compared with controls, and again a great overlap was seen between IFX and VDZ of these persistent dysregulated genes (eg. up: IL1B, TIMP1, CCL20, DEFA5/A6, PI3, AREG, PTGS2, C2, SERPINB5, FAM5C and down: AQP8, MT1H/M).

Conclusion

This study demonstrates that VDZ and IFX restore, although incompletely, the colonic expression of many immune-related genes in UC patients achieving endoscopic healing with VDZ at W52 and with IFX at W4-6. Persistent abnormalities in gene expression remain after therapy in healers and may explain why mucosal lesions recur if patients do not receive maintenance therapy. Furthermore, the significant overlap in persistent dysregulated genes between VDZ and IFX healers suggests that unidentified triggers of inflammation are incompletely blocked by these biologic agents.