26 THE INTESTINAL MICROBIOME IN INFLAMMATORY BOWEL DISEASE ACROSS THE PEDIATRIC AGE RANGE. (7th February 2019)

• Record Type: Journal Article
• Authors: Conrad, Maire A; Bittinger, Kyle; Ren, Yue; Breton, Jessica; Devas, Nina; Devoto, Marcella; Bushman, Frederic; Baldassano, Robert; Kelsen, Judith
• Is Part Of: Inflammatory bowel diseases. Volume 25(2019)Supplement 1
• Page Start: S70
• Page End: S70
• Publication Date: 2019-02-07
• Subjects: Inflammatory bowel diseases -- Periodicals; Colitis, Ulcerative -- Periodicals; Crohn Disease -- Periodicals; Inflammatory Bowel Diseases -- Periodicals; 616.344
• DOI: 10.1093/ibd/izy393.176
• Languages: English
• ISSNs: 1078-0998

Abstract

Abstract: : The influence of genomics, immune response, and environment on inflammatory bowel disease (IBD) varies among patients. We aimed to characterize the differences and determine age and disease effects on the gut microbiome across the pediatric age range, measured by the maturity of the microbiota. Methods: Eight week observational study of patients with very early onset IBD (VEO-IBD), diagnosed ≤6yo, pediatric IBD (>8yo) and healthy pediatric controls (HC) was performed at the Children's Hospital of Philadelphia, including IBD patients on maintenance therapy. Antibiotic use was not an exclusion criteria. Subjects on/or recent antibiotic use were analyzed separately. Metadata including disease activity, therapies, and diet was collected. Total DNA from stool was sequenced using the Illumina HiSeq paired-end method. MetaPhlAn2 assessed taxonomic composition, identifying clade-specific marker genes from bacterial, fungal, archaeal and viral sources. A Microbiota Maturity Index (MMI) was generated from HC (Subramanian et al, Nature 2014). We trained a random forest model to predict age based on 13 bacterial taxa that changed in abundance with age. Subsequently, we built a random forest model to apply to the IBD cohorts, determining age predicted by the taxa abundance. The predicted age is the MMI, which is plotted against the chronological age. Results: 170 subjects were enrolled, 73 HC (4mo-17yo), 10 HC on antibiotics (16mo-6yo), 21 pediatric IBD, and 66 VEO-IBD. CD was diagnosed in 75% of all IBD subjects. Disease activity measured by fecal calprotectin was significantly higher in VEO-IBD than pediatric IBD. Current therapies included anti-TNF and vedolizumab in 28% of the VEO subjects and 90% of pediatric IBD subjects, and 23 VEO-IBD subjects were on antibiotics at baseline. The relative abundance of 6 species were significantly different between VEO-IBD and pediatric IBD, including increased Veillonella dispar, Haemophilus parainfluenzae, Coprobacillus sp in VEO-IBD. The generated MMI was most influenced by the abundance of Firmicutes and Bacteroidetes. Several of the 13 taxa in the MMI significantly differed between younger and older HC, yet no difference was seen in the IBD cohorts. IBD cohorts were predicted to have a less mature microbiome than expected for their chronological age, however, the relative MMI index was significantly lower in pediatric IBD than VEO-IBD with/without antibiotic use. Conclusion: Both IBD cohorts demonstrated a lower microbiota maturity as compared to HC, indicative of disease effect on the microbial community. The MMI was lowest in pediatric IBD, suggesting that duration of disease may play a role in the maturity of the intestinal microbiome. Interestingly, there were several differences in the microbial community within IBD, suggesting an age effect and different environmental drivers in VEO-IBD and pediatric IBD.