Metabonomics and the Gut Microbiome Associated With Primary Response to Anti-TNF Therapy in Crohn's Disease.

BACKGROUND AND AIMS: Anti-tumour necrosis factor [anti-TNF] therapy is indicated for treatment of moderate to severe inflammatory bowel disease [IBD], but has a primary non-response rate of around 30%. We aim to use metabonomic and metataxonomic profiling to identify predictive biomarkers of anti-TNF response in Crohn's disease. METHODS: Patients with luminal Crohn's disease, commencing anti-TNF therapy, were recruited with urine, faeces, and serum samples being collected at baseline and 3-monthly. Primary response was defined according to a combination of clinical and objective markers of inflammation. Samples were measured using three UPLC-MS assays: lipid, bile acid, and Hydrophillic Interaction Liquid Chromatography [HILIC] profiling with 16S rRNA gene sequencing of faeces. RESULTS: Samples were collected from 76 Crohn's disease patients who were anti-TNF naïve and from 13 healthy controls. There were 11 responders, 37 non-responders, and 28 partial responders in anti-TNF-treated Crohn's patients. Histidine and cysteine were identified as biomarkers of response from polar metabolite profiling [HILIC] of serum and urine. Lipid profiling of serum and faeces found phosphocholines, ceramides, sphingomyelins, and triglycerides, and bile acid profiling identified primary bile acids to be associated with non-response to anti-TNF therapy, with higher levels of phase 2 conjugates in non-responders. Receiver operating curves for treatment response demonstrated 0.94 +/ -0.10 [faecal lipid], 0.81 +/- 0.17 [faecal bile acid], and 0.74 +/- 0.15 [serum bile acid] predictive ability for anti-TNF response in Crohn's disease. CONCLUSIONS: This prospective, longitudinal cohort study of metabonomic and 16S rRNA gene sequencing analysis demonstrates that a range of metabolic biomarkers involving lipid, bile acid, and amino acid pathways may contribute to prediction of response to anti-TNF therapy in Crohn's disease. PODCAST: This article has an associated podcast which can be accessed at https://academic.oup.com/ecco-jcc/pages/podcast.