Plasma Nuclear Magnetic Resonance metabolomics discriminates between high and low endoscopic activity and predicts progression in a prospective cohort of patients with ulcerative colitis.
Probert F., Walsh A., Jagielowicz M., Yeo T., Claridge TDW., Simmons A., Travis S., Anthony DC.
Endoscopic assessment of ulcerative colitis (UC) is one of the most accurate measures of disease activity, but frequent endoscopic investigations are disliked by patients and expensive for the health care system. A minimally-invasive test that provides a surrogate measure of endoscopic activity is required.Plasma Nuclear Magnetic Resonance (NMR) spectra from 40 patients with UC followed prospectively over 6 months were analysed with multivariate statistics. NMR metabolite profiles were compared with endoscopic (Ulcerative Colitis Endoscopic Index of Severity: UCEIS), histological (Nancy Index), and clinical (Simple Clinical Colitis Activity Index: SCCAI) severity indices, along with routine blood measurements.A blinded principal component analysis spontaneously separated metabolite profiles of patients with low (≤3) and high (>3) UCEIS. Orthogonal partial least squares discrimination analysis identified low and high UCEIS metabolite profiles with an accuracy of 77±5%. Plasma metabolites driving discrimination included decreases in lipoproteins and increases in isoleucine, valine, glucose, and myo-inositol in high compared to low UCEIS. This same metabolite profile distinguished between low (Nancy 0-1) and high histological activity (Nancy 3-4) with a modest though significant accuracy (65±6%) but was independent of SCCAI and all blood parameters measured. A different metabolite profile, dominated by changes in lysine, histidine, phenylalanine, and tyrosine, distinguished between improvement in UCEIS (decrease >1) and worsening (increase >1) over 6 months with an accuracy of 74±4%.Plasma NMR metabolite analysis has the potential to provide a low-cost, minimally invasive technique that may be a surrogate for endoscopic assessment, with predictive capacity.