A [Glyco]biomarker that Predicts Failure to Standard Therapy in Ulcerative Colitis Patients. (20th September 2018)

• Record Type: Journal Article
• Authors: Pereira, Márcia S; Maia, Luís; Azevedo, Luís F; Campos, Sara; Carvalho, Sandra; Dias, Ana M; Albergaria, André; Lima, Jorge; Marcos-Pinto, Ricardo; Lago, Paula; Pinho, Salomé S
• Is Part Of: Journal of Crohn's and colitis. Volume 13:Number 1(2019)
• Page Start: 39
• Page End: 49
• Publication Date: 2018-09-20
• Subjects: Ulcerative colitis -- biomarker -- glycosylation; Inflammatory bowel diseases -- Periodicals; 616.344005
• DOI: 10.1093/ecco-jcc/jjy139
• Languages: English
• ISSNs: 1873-9946

Abstract

Abstract: Background and Aims: There is a clinical need to identify biomarkers able to select patients who are most likely to develop aggressive/complicated disease, for early selection for appropriate therapy. Changes in the glycosylation profile of intestinal lymphocytic infiltrate were previously demonstrated to regulate T cell activity, being associated with disease severity in ulcerative colitis [UC] patients. We interrogated whether this heterogeneous expression of branched N-glycans in intestinal inflammatory infiltrate predicts therapy response early in disease course. Methods: The expression levels of the branched N-glycans in colonic biopsies collected around time of diagnosis from a well-characterised cohort of 131 UC patients were correlated with response to standard therapy. Receiver operating characteristic analysis and specificity/sensitivity were determined. Results: Branched N-glycans levels around time of diagnosis predict non-response to conventional therapy with 75% specificity. Moreover, high levels of branched N-glycans predict 78% of UC patients who will display a favourable disease course [exclusively under 5-aminosalicylate therapy for more than 5 years of disease]. The best predictive performance was observed in severe UC patients with Mayo endoscopic subscore 3 and in those that were naïve to therapy. Multivariable analysis revealed that low levels of branched N-glycans and high levels of C-reactive protein [CRP] around time of diagnosis act as independent predictors of non-response to standard therapy. A powerful effect of the combined use of the branched N-glycans and CRP was observed. Conclusions: Our results reveal a potential [glyco]biomarker that predicts, early in the disease course, patients who will fail to respond to standard therapy, benefiting thereby from other therapeutic strategies such as biologics.