Metabolic predictors of disease outcomes in multiple sclerosis

Project Summary Multiple sclerosis (MS) is a common inflammatory and neurodegenerative disorder. In MS progression of disability is irreversible, and prognosis is highly variable; some individuals rapidly progress to a disabled state whereas others experience only mild symptoms. However, mechanisms contributing to the observed heterogeneity in disease evolution are poorly understood. Discovery of novel biomarkers associated with risk of disease progression will not only allow for more accurate individualized prognosis but also facilitate the discovery of new therapeutic targets that may be relevant for targeting the progressive aspects of MS. Metabolomics is an ideal technology for biomarker discovery; an individual’s metabolic phenotype incorporates multiple levels of biologic interaction (e.g., endogenous metabolism, the exposome, and activity of the gut microbiota). We previously found robust metabolic alterations in people with MS when compared to healthy people in a study including nearly 1000 metabolomic profiles. Results suggest a marked disruption of multiple amino acid pathways, with notable reductions in metabolites related to aromatic amino acid metabolism (phenylalanine, tryptophan, and tyrosine). Lower levels of these and other metabolites were also strongly correlated with disability levels at a single time point. The overall goal of the proposed studies is to build upon these initial results by evaluating whether certain metabolic changes predict MS prognosis and explore potential contributing mechanisms using a data-driven approach. We will evaluate, in a prospective design, whether (1) certain metabolic changes predict subsequent MS prognosis in Aim 1; (2) characterize potential contributing mechanisms by considering the mediation by MS disease modifying therapies in Aim 2; and (3) assess the added predictive value of metabolomic markers when combined with traditional measures of disease severity in Aim 3. Our central hypothesis is that metabolic changes, both in in AAA metabolism, as well as other novel pathways, strongly predict subsequent clinical and radiological MS outcomes (i.e., MS prognosis). To evaluate this hypothesis, we will use data and samples collected from nearly 1500 PwMS participating in three randomized studies. These datasets offer an abundance of advantages in evaluating metabolic predictors of MS outcomes. For example, they are large cohorts in which standardized collection of biospecimens and rigorously assessed outcomes are collected at pre-specified longitudinal intervals. These valuable resources will be combined with validation in clinical, real-world NIH-funded observational cohorts of 400 PwMS. Lastly, our study will apply an innovative analytic strategy applying advanced epidemiological modeling tools rooted in causal inference. The collective results of this study stand to (1) provide novel insight into underlying mechanisms contributing to disability accumulation in PwMS; and (2) identify novel therapeutic targets that are relevant to progressive aspects of MS for which there are few treatments.