P1209 Baseline gut microbiota composition and function reflect response to 5-ASA treatment in Ulcerative Colitis

Abstract

Abstract Background Mesalazine (5-ASA) is a key initial treatment of ulcerative colitis (UC); recent studies have highlighted the role of the gut microbiota in 5-ASA metabolism. There is no established means of predicting treatment outcome. We sought to investigate changes in the gut microbiota and metabolome with 5-ASA treatment, and whether such changes may reflect clinical outcomes. Methods Longitudinal faeces, serum and urine samples were collected from patients with newly diagnosed UC before and after 5-ASA treatment. We documented longitudinal microbiota and metabolome changes with 5-ASA treatment and compared baseline gut microbiota and metabolome in responders (simple clinical colitis activity index (SCCAI) of <4 and faecal calprotectin (FC) of <100 µg/g) and non-responders to 5-ASA. Stool bacterial profiling was performed via 16S rRNA gene sequencing of the V1/V2 hypervariable regions. Host and microbial metabolites in stool, serum and urine were detected and semi-quantitated using liquid chromatography mass spectroscopy (LC-MS) and nuclear magnetic resonance spectroscopy (1H-NMR). Kruskal-Wallis tests and paired analyses using a Wilcoxon rank test were used and p values were adjusted using Benjamini-Hochberg false discovery rate (FDR) correction. Results 37 patients newly diagnosed with UC were recruited. Samples were collected at baseline and a median of 83 days (IQR 22.5 days) later; there were 27 responders and 10 non-responders. Comparing pre- and post- 5-ASA treatment samples demonstrated a reduction in gut microbial α diversity (P=0.01) (Fig 1). Faecal methionine increased with 5-ASA (P=0.04) in all patients, and isovalerate reduced with 5-ASA (P<0.01). Similar reductions after 5-ASA were noted in serum glycoprotein B (GlycB) (P=0.02). At baseline, responders tended to have increased α diversity compared to non-responders, but not statistically significantly (P=0.09). Similarly, there were β diversity differences, although these did not reach significance (P=0.06). At baseline non-responders had a higher relative stool abundance of Bacteroidetes and Firmicutes (P<0.01) compared to responders. Agathobacter, Coprococcus and Blautia genera were enriched in non-responders (P=0.05). At baseline, urinary hippurate was higher in patients going on to respond to 5-ASA compared to non-responders (P=0.04); faecal lysine (P<0.01) and putrescine (P=0.01) levels were also higher in non-responders at baseline. 5-ASA responders had reductions in faecal 2-methylbutyrate and isobutyrate (P=0.02) longitudinally. Conclusion We have shown that baseline relative abundance of certain taxa (such as Blautia), urinary Hippurate and faecal lysine reflect response to 5-ASA. Alterations in the gut microbiome and metabolome are associated with 5-ASA treatment.