Tryptophan as a biomarker for the course of IBD and anti-TNF alpha response

Background Disturbed tryptophan (TRP) metabolism is associated with chronic inflammation. The administration of TRP and selected metabolites showed anti-inflammatory effects on the mucosa in animal models of colitis.1 The aim of our study was to systematically evaluate serum levels of TRP in a cohort of patients with inflammatory bowel disease (IBD), as an estimate for the mucosal availability. Within this cohort we focused on the induction of anti-inflammatory therapy. Methods Serum tryptophan levels were determined in 555 outpatients with IBD and in 291 healthy blood donors, and correlated with activity scores and markers of inflammation. TRP levels were quantified using high-performance liquid chromatography. To further analyse TRP levels during the induction of an anti-inflammatory therapy with infliximab as anti-TNF α -inhibitor (n = 36) or vedolizumab as integrin-inhibitor (n = 30), we selected sub-groups of IBD patients with longitudinal assessment before (= baseline), 2 weeks, 6 weeks, 14 weeks, and 6 months after induction. We used the Spearman’s rank correlation coefficient, Wilcoxon-rank-sum tests and paired t-tests. Results In our first analysis of 555 IBD patients, serum TRP levels were significantly reduced compared with controls (p < 0.001). Disease activity and C-reactive protein showed an inverse (p < 0.001) correlation with TRP levels. In our second analysis of the responding patients to an anti-TNF α -therapy we observed 2 weeks vs baseline (p = 0.04) and 6 months vs baseline (p = 0.03) significantly up regulated TRP levels. This effect was neither found in other time points, nor during the induction of vedolizumab, 2 weeks vs baseline (p = 0.695) and 6 months vs baseline (p = 0.135). Conclusion TRP levels and disease activity showed a significant negative correlation. During the induction of an anti-TNF α therapy TRP levels were significantly upregulated after 2 weeks compared with baseline. This effect might be explained either by chances in the microbiome or by the down regulation of TNF α, as an important inductor of the TRP degrading enzyme Indolamin-2, 3-Dioxygenase. This effect was not observed after 6 weeks and 14 weeks, when the concentration of anti-TNF α in the metabolism is lower. After 6 months, we found clinically a mild disease or remission, which explains the significantly upregulated TRP levels. In contrast, during the induction of vedolizumab, which is characterised by a longer period until remission, no changes of TRP were measured. These findings indicate that TRP serves as a possible biomarker for the disease activity and response to anti-TNF α therapies.