Abstract Background Rigorous donor preselection, strict anaerobic processing, and repeated faecal microbiota transplantation (FMT) administration did not improve outcomes for induction of clinical remission in the RESTORE-UC trial1. We studied the luminal, and mucosal microbiota composition as well as gene expression throughout the RESTORE-UC study, to further understand the results and instruct future FMT trial design. Methods The RESTORE-UC trial was a multi-centric double-blind, sham-controlled randomized trial. Patients with moderate to severe UC (total Mayo 4-10) were randomly allocated to receive 4 anaerobic-prepared allogenic or autologous donor FMT. Mucosal biopsies were collected for RNA- and 16S sequencing at week 0 and 8. Quantitative microbiota profiling was performed on donor samples, and in patients at weeks 0, 1, 2, 3, 4, 8, 12, 26, and 52 (n=756). Clinical success (steroid free clinical remission) and -response (≥3 points or ≥50% reduction in combined Mayo subscores for rectal bleeding and stool frequency) were evaluated at week 8. Metagenomic success (week 8) was defined as restoration of eubiosis (Bacteroides 2 (B2) to non-B2 enterotype) and failure was considered with subtypes being no transition, non-B2 to other non-B2 transition, lead dysbiosis (non-B2 to B2), and maintenance of dysbiosis (B2 to B2). Results Baseline beta diversity (metric for overall microbial similarity) was significantly associated with future clinical and metagenomic success (resp. p=0.035; p=0.003). Similar trends were observed at week 8 (resp. p=0.056; p=0.001). Restoration of eubiosis was characterized by a rapid increase in bacterial richness after the first FMT administration, which was maintained up to 12 months after intervention (Fig A). A greater difference (Fig B) in donor-recipient richness determined the metagenomic success (p=0.06) and clinical response (p=0.01), while this was not observed for clinical success (p=0.16). Metagenomic success was accompanied by increased engraftment after FMT (Fig C, p<0.01). Independently from clinical success, patients with dysbiotic faecal communities at baseline had distinctive mucosal communities in their intestinal biopsies (p<0.01), which were significantly changed in samples where eubiosis was restored (p=0.012), but not in those where clinical success was observed (p=0.34). However, this could be explained by the low overall response rate. No specific transcriptomic profiles in mucosal biopsies at baseline were reminiscent of metagenomic (p=0.24), nor clinical success (p=0.22). Conclusion Patient and donors should be rigorously selected in future FMT trials, and selection should be based on microbial composition and by obtaining a great difference in richness between receptor and donor. References 1 Caenepeel*, Deleu*, Vazquez* et al., 2024
Read full abstract