Jones CMA¹, Connors J², Dunn KA³, Bielawski JP^3,4, Comeau AM⁵, Langille MGI^1,5, Van Limbergen J^2,6,7. Inflamm Bowel Dis. 2020 Jan 21. pii: izaa001. doi: 10.1093/ibd/izaa001. [Epub ahead of print]

Author information

1 Department of Pharmacology, Dalhousie University, Halifax, Canada.

2 Department of Pediatrics, Dalhousie University, Halifax, Canada.

3 Department of Biology, Dalhousie University, Halifax, Canada.

4 Department of Mathematics & Statistics, Dalhousie University, Halifax, Canada.

5 Integrated Microbiome Resource (IMR), Dalhousie University, Halifax, Canada.

6 Tytgat Institute for Liver and Intestinal Research, Amsterdam Gastroenterology and Metabolism, Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands.

7 Department of Pediatrics, Division of Pediatric Gastroenterology & Nutrition, Emma Children's Hospital, Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands.

Abstract

BACKGROUND: The gut microbiome is extensively involved in induction of remission in pediatric Crohn's disease (CD) patients by exclusive enteral nutrition (EEN). In this follow-up study of pediatric CD patients undergoing treatment with EEN, we employ machine learning models trained on baseline gut microbiome data to distinguish patients who achieved and sustained remission (SR) from those who did not achieve remission nor relapse (non-SR) by 24 weeks.

METHODS: A total of 139 fecal samples were obtained from 22 patients (8-15 years of age) for up to 96 weeks. Gut microbiome taxonomy was assessed by 16S rRNA gene sequencing, and functional capacity was assessed by metagenomic sequencing. We used standard metrics of diversity and taxonomy to quantify differences between SR and non-SR patients and to associate gut microbial shifts with fecal calprotectin (FCP), and disease severity as defined by weighted Pediatric Crohn's Disease Activity Index. We used microbial data sets in addition to clinical metadata in random forests (RFs) models to classify treatment response and predict FCP levels.

RESULTS: Microbial diversity did not change after EEN, but species richness was lower in low-FCP samples (<250 µg/g). An RF model using microbial abundances, species richness, and Paris disease classification was the best at classifying treatment response (area under the curve [AUC] = 0.9). KEGG Pathways also significantly classified treatment response with the addition of the same clinical data (AUC = 0.8). Top features of the RF model are consistent with previously identified IBD taxa, such as Ruminococcaceae and Ruminococcus gnavus.

CONCLUSIONS: Our machine learning approach is able to distinguish SR and non-SR samples using baseline microbiome and clinical data.