Bile Acid Profiles are Not Altered by Fecal Microbiota Transplantation for the Treatment of Primary Sclerosing Cholangitis: Category Award (Liver): Presidential Poster Award

Abstract

Introduction: Primary sclerosing cholangitis (PSC) is a progressive cholestatic liver disease. Bacterial profiles associated with abnormal bile acid metabolism have been described in PSC. Fecal microbiota transplantation (FMT) alters the bacterial composition in PSC, but the impact upon bile acid profiles is unknown. Accordingly, this study assessed bile acid profiles in PSC patients pre- and post-FMT. Methods: Open-label study of PSC patients (n=10) with concurrent inflammatory bowel disease with colonic involvement and at least one of AST, ALT or ALP > 1.5X the upper limit of normal. Participants underwent an FMT by colonoscopy. Liver enzyme profiles and stool metabolomic analysis was conducted at baseline and week 1, 4, 8, 12 and 26 post-FMT. The primary efficacy outcome was a decrease in ALP ≥50% from baseline by week 26 post-FMT. Fecal samples were lyophilized and bile acids were extracted. Multivariate analysis of ultra-performance liquid chromatography-massspectrometry (UPLCMS) bile acid profiling data was performed. Results: 10 patients underwent FMT. Mean age was 41.1 years (SD 15.7) and were primarily male (8/10). 9 patients had concurrent UC and 1 with Crohns. The mean baseline ALP was 489. Overall, 30% of patients experienced a ≥50% decrease ALP and 70% experienced a 30% decrease in at least one of their liver biomarkers post-FMT. Multi-variate analysis of UPLC-MS bile acid data demonstrated clustering of serial samples by study participant (Fig 1A), but there was no clear effect of FMT upon fecal bile acid profile clustering (Fig 1B). Bile acid profiles pre-FMT were similar to that of the donor and FMT had no significant effect on bile acid profiles (Fig 2). Spearman's rank analysis demonstrated negative correlation between lfts and ursodeoxycholic acid levels (r=-0.423 for ALP, r=-0.53 for AST, r=-0.61 for ALT) and percentage of tertiary bile acids of total bile acids (r=-0.40 for ALP, r =-0.56 for AST, r =-0.61 for ALT). A positive correlation was found between lfts and percent of aminoconjugate bile acids of total bile acids (r=0.39 for ALP, r=0.41 for AST, r=0.43 for ALT). Linear regression demonstrated a negative correlation between ALP and % tertiary bile acids (p=0.0028) (Fig 3A), but a positive correlation between ALP and % tauro conjugates (p=0.01) (Fig 3B) and % glyco conjugates (p=0.0004) (Fig 3C).1017_A Figure 1. Principal component analysis (PCA) of bile acid profile data. A) PCA colored by study participant; B) PCA colored by timepoint within study.1017_B Figure 2. Relative abundance plot of bile acid profiles from the donor and all serial patient samples.1017_C Figure 3. Linear regression analysis of bile acid profiling data (see text for p values). A) ALP vs % tertiary bile acids for all serial patient samples; B) ALP vs % tauroconjugated bile acids for all serial patient samples; C) ALP vs % glyco-conjugated bile acids for all serial patient samples.Conclusion: In this study ALP correlated closely with certain bile acid groups, although bile acid profiles did not change post-FMT.