Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

P062 Metabolic phenotyping of healthy and IBD children: Perspectives for individual metabolic monitoring

F.-P. Martin*1, O. Cominetti1, L. Da Silva1, J. Ezri2, S. Collino1, I. Montoliu1, J.-P. Godin3, A. Nydegger2

1Nestle Institute of Health Sciences, Molecular Biomarkers, Lausanne, Switzerland, 2University of Lausanne, Pediatric Gastroenterology, Lausanne, Switzerland, 3Nestle Research Center, Analytical Sciences, Lausanne, Switzerland

Background

Whilst the prevalence of Inflammatory Bowel Disease (IBD) has increased considerably over the last decades, its clinical feature does not allow accurate prediction of disease progression or response to specific therapy. In this context, we used omics technologies to generate a systemic view of IBD pathogenesis and generate further working hypothesis for multiple pathway-integrated therapies.

Methods

A total of 21 paediatric patients with IBD (mean age 14.8 years (range 12.7 - 16.7), 8 males) have been enrolled from the pediatric gastroenterology outpatient clinic over two years. The protocol was approved by the local Swiss Ethical Committee. Clinical and anthropometric data were collected at baseline, 6 and 12 months. A control group of 29 healthy children (mean age 13.1 years (range 10.1 - 16.7), 17 males) has been assessed at baseline. Anthropometric parameters, total and resting energy expenditure have been performed in all individuals, inflammatory and growth markers only in IBD patients. Morning spot urine samples were collected at each visit and subjected to 1H Nuclear Magnetic Resonance (NMR) spectroscopy.

Results

Metabolic profiling of urine samples based on 1H NMR spectroscopy combined with multivariate analysis described two phenotypes corresponding to IBD and healthy children. Orthogonal Projection to Latent Structure Discriminant Analysis (OPLS-DA, explained variance in metabolic data: R2X=0.11, indicator of model robustness: Q2Y=0.28 using 1 predictive and 1 orthogonal components), described distinct central energy metabolic status and gut microbial metabolic signatures. Based on clinical and anthropometric data, sub-populations of IBD patients could be stratified in active disease states, or in remission with either body weight catch-up or body weight and height catch-up. OPLS-DA model generated with 2 predictive and 2 orthogonal components showed statistically significant differences between groups (R2X=0.23, Q2Y=0.24). Children with active disease had higher urinary excretion in purines and pyrimidines, which may reflect immune cell and drug metabolism. Children with body weight catch-up had a differential excretion of gut microbial metabolites (aromatic and short-chain organic acids), whilst children with body weight and height catch-up showed increased IFG-1, IGF-3, and urinary excretion of intermediates in urea metabolism.

Conclusion

The use of metabonomics in chronically ill pediatric patients might be useful in defining the metabolic requirements at the different stages of the disease. Integration of metabolic trajectories as a next step may provide additional insights on metabolic readouts, which might be relevant for individual status monitoring.