Search in the Abstract Database

Abstracts Search 2018

DOP089 Individualised variations in the gut microbiota during remission precede Crohn’s disease flares

T. Braun1, A. Di Segni1, S. Neuman1, N. Levhar1, M. Bubis1, O. Picard1, M. BenShoshan1, G. Efroni1, S. Farage Barhom1, E. Glick Saar1, A. Lahad1, D. Shouval1, B. Weiss1, A. Lahat1, R. Eliakim1, S. Ben-Horin1, U. Kopylov1, The Leona M. and Harry B. Helmsley Charitable Trust supported Israeli IBD Research Nucleus (IIRN), Y. Haberman1,2*

1Sheba Medical Center, Tel Hashomer, Israel, 2Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA


Altered microbial composition (dysbiosis) occurs in Crohn’s disease (CD), but whether this is an epiphenomenon or a primary etiopathogenic process inciting disease flares remains undefined. Murine studies showed that gut dysbiosis from CD patients augments pro-inflammatory responses in the host, suggesting contribution to pathogenesis.


We characterised prospective longitudinal fecal samples by 16S rRNA amplicon sequencing of meticulously characterised CD patients cohort in remission from the Leona M. and Harry B. Helmsley Charitable Trust supported Israeli IBD Research Network (IIRN, 42 CD cases and 220 samples). As a reference for comparison, we also included samples obtained from active inflamed CD (n = 17), and healthy controls samples (n = 22). Follow-up of the patients in remission included PGA, CDAI, and biomarkers (fecal calprotectin and CRP) every 3 months, and flares were recorded during follow-up.


CD patients in clinical and biomarkers remission showed significantly reduced alpha diversity (richness within sample) and increased dysbiosis index in comparison to healthy controls (p < 0.001), suggesting that dysbiosis is not just secondary to inflammation. Dysbiosis index negatively correlated with alpha diversity (spearman r = -0.49, p < 0.0001) and positively correlated with fecal calprotectin levels (spearman r = 0.42, p < 0.0001). Significant reduced abundance of Ruminococcaceae and Lachnospiraceae, and increased abundance of Fusobacteriaceae taxa is noted in CD patients in remission. Patients in remission had longitudinal fecal samples. We next characterised the within-subject variations in fecal samples preceding the flare in comparison to variations in subjects that where in remission throughout. Remarkably, we detected higher within-subject microbial alteration in samples preceding the CD flare (p = 0.02, ROC AUC=0.75).


CD patients in remission still show altered microbial communities and within-subject microbial alteration precedes CD flares, thereby supporting a role for Dysbiosis in disease pathogenesis, rather than being merely an epiphenomenon of it. Our model can be used to guide preemptive therapy intensification strategies during remission. These findings indicate the need to explore if manipulating or perhaps strictly avoiding further microbial alteration during remission may change the course of CD.