P063 A COMPREHENSIVE, MULTIOMIC DIET INTERVENTION STUDY COMPARING A LOW FAT, HIGH FIBER DIET TO AN IDEALIZED STANDARD AMERICAN DIET IN ULCERATIVE COLITIS PATIENTS

Background and Aims There is a lack of evidence-based dietary interventions in ulcerative colitis (UC) management. A diet high in fat and animal meat has been linked to an increased risk of UC. The aim of our study was to use a multilayered, multi-omic approach to comprehensively characterize the effect of a low fat, high fiber diet or a high fat diet in UC patients. Methods We enrolled patients with UC who were in remission or had mild disease with a flare within the last 18 months. We used a cross-over design in which patients received two dietary interventions: a low fat diet (LFD), containing 10% total calories from fat with an omega 6 to 3 ratio of below 3:1, and an idealized standard American diet (SAD), containing 35–40% total calories from fat with an omega 6 to 3 ratio of 20–30:1. Each diet was four weeks long with a two-week wash-out in between. The diet was catered and delivered to patients’ homes. Clinical symptoms, quality of life, and biochemical data were collected. Stool was collected for microbiome and metabolomic analyses. The primary endpoint was to determine adherence to a specified diet using catered meals; the secondary endpoint was to determine the clinical and subclinical effects of a low fat, high fiber diet or high fat diet in UC. Results Baseline diets varied widely but were generally lower in fiber as well as fruits and vegetables and higher in saturated fat than either of the study diets. There was a high rate of adherence to catered meals (SAD=86.68%, LFD=84.8%) with a 96.8% and 94.33% adherence to fat for SAD and LFD respectively. Patients that started in remission remained in remission (partial Mayo and sIBDQ). Following a LFD, patients saw a 20% improvement in their quality of life as measured by sIBDQ compared to their baseline. The effect of diet intervention on microbial diversity was reflected in the beta diversity with a significant increase in Faecalibacterium prausnitzii after LFD. CRP, sIBDQ, IL-6, and IL1β had a significant effect on overall gut microbiota composition as measured by Bray Curtis beta diversity (PERMANOVA)(P<0.007, P<0.001, P<0.021, P<0.048 respectively). The top taxa that contributes the most to this microbial variation from these clinical parameters was Faecalibacterium prausnitzii. Patients following a SAD had an increase in lauric acid, myristic acid, and N-oleoyl-L-phenylalanine with an increase in omega-6 metabolism pathways. Patients following a LFD had higher glycine, alanine, and phenyllactic acid with omega 3 metabolism pathways increased after LFD. Conclusions A low fat, high fiber diet is well tolerated and did not increase biochemical markers of inflammation. Catered meals and collection of microbiome, metabolome and biochemical data may allow early stratification of diet responders.