P075 SER-287, AN INVESTIGATIONAL MICROBIOME THERAPEUTIC, INDUCES WIDESPREAD METABOLOMIC AND HOST TRANSCRIPTIONAL CHANGES RELATED TO CLINICAL REMISSION IN PATIENTS WITH ACTIVE MILD-TO-MODERATE ULCERATIVE COLITIS

Abstract

Background: Firmicute bacteria and the metabolites they produce are perturbed in ulcerative colitis (UC) patients. SER‐287, an oral formulation of live Firmicute spores, was found to be safe and effective in achieving clinical remission in a Phase 1b clinical trial in mild‐to‐moderate UC. We assessed changes in the stool metabolome and host transcriptome among remitters and non‐remitters as a measure of SER‐287 pharmacodynamics. Methods: 58 UC subjects (modified Mayo score 4‐10) were randomly assigned to one of four 8‐week induction treatment arms preceded by a 6‐day antibiotic preconditioning phase as follows: a) placebo/placebo b) placebo/SER‐287 weekly c) vancomycin/SER‐287 weekly or d) vancomycin/SER‐287 daily. Clinical remission was defined as a Total Modified Mayo Score ≤2 plus an endoscopic subscore ≤1. Stool samples were evaluated for microbiome composition (whole metagenomic sequencing), targeted and global metabolomics (liquid chromatography‐mass spectrometry) and transcriptomics (RNA‐Seq with mucosal biopsy) at baseline and at week 8. Results: Vancomycin/SER‐287 daily was associated with higher rates of clinical remission compared to placebo/SER‐287 weekly, vancomycin/SER‐287 weekly or placebo/ placebo (40%, 13.3% 17.6% and 0%, respectively; p=0.024 for the comparison of daily SER‐287 versus placebo/placebo). Significant changes in microbial‐associated metabolites (bile acids, indole, taurine, and arachidonate) were associated with clinical outcome across all arms (p<0.05, Mann‐Whitney; Figure A). Similar associations with treatment were also observed (p<0.05, Mann‐Whitney). Host transcriptomic changes associated with clinical outcome across arms included decreased expression of genes associated with signaling of TNF, NF‐kappaB and IL‐17, as well as increased expression of genes associated with the metabolism of the short chain fatty acid propanoate (adjusted p‐value<0.05, gene set enrichment analysis, GSEA; Figure B). Similar associations with treatment were also observed (adjusted p‐value<0.05, GSEA). Conclusions: In this Phase 1b study of SER‐287 the highest rates of remission were associated with daily dosing of SER‐287, preceded by vancomycin preconditioning. Across treatment arms, clinical remission was associated with a wide range of metabolomic and transcriptomic changes relevant to UC pathogenesis. Changes in metabolomics and transcriptomics are associated with immune cell recruitment and activation, epithelial integrity and mucosal barrier function, giving insights into the mechanism of action of this microbiome therapeutic. Metabolite abundances in clinical remitters and non‐remitters at baseline and Week 8 for select metabolites. Metabolites shown have p‐value<0.05 (Mann‐ Whitney U) comparing remitters and non‐remitters at Week 8. Points outside of the box‐and‐whiskers indicate outliers. KEGG pathways found to be significantly differentially expressed between clinical remitters and non‐remitters. Directionality of expression is given in the third column. All p‐values are calculated using gene‐set enrichment analysis (GSEA) and adjusted for multiple testing.