(P123) DONOR-DERIVED GUT MICROBIOTA INFLUENCE THE EFFICACY OF COMBINATION WITH FMT AND MULTIPLE ANTIBIOTICS THERAPY FOR ULCERATIVE COLITIS

Background and aims: We previously reported that faecal microbiota transplantation (FMT) via colonoscopy after two weeks of oral administration of triple antibiotics (amoxicillin, fosfomycin, and metronidazole) therapy (antibiotic-FMT: A-FMT) leads to favourable clinical outcomes in patients with ulcerative colitis (UC) which is associated with intestinal dysbiosis. However, whether the beneficial effects associated with FMT efficacy in patients and/or donors are due to the transfer and engraftment of donor-derived bacterial species into patients remains unknown. Thus, we explored the impact of the optimal timing for A-FMT treatment and appropriate donor and patient–donor matching.

Methods: Ninety-seven patients with active UC were enrolled between March 2014 and October 2019, and FMT was performed. "Responder" was defined by any of the following conditions: 1) Lichtiger's clinical activity index (CAI) was ≤ 9 and improved by 3 or more points or 2) the CAI improved by 5 or more points at 4 weeks after A-FMT. “Remission” was defined as a further improvement in the CAI to ≤ 3 points. “Long-term responder” was defined as no increase in the CAI or intensification within 12 months after A-FMT. In addition, 16S rRNA gene sequence analysis for microbiome was performed on 147 faecal samples (pre-A-FMT, post-A-FMT, and donor) from 49 patient-donor combinations that were assigned using the one patient-to-one donor strategy.

Results: Of the 97 patients, 61 patients (62.9%) achieved a clinical response, and 35 (36.1%) achieved clinical remission. Effective donors were rich in taxa such as Bacteroidota, which are lost in UC, and the abundances of "patient-origin" and "new-amplicon sequence variant (ASV)" taxa were significantly lower in responders (p < 0.05). Moreover, "donor−derived" ASV sequences, such as Alistipes, were significantly enriched in responders (p < 0.01). Patient characteristics, such as the Mayo score, endoscopic evaluation, and taxonomic composition, had a significant impact (p < 0.05) on the efficacy of A-FMT. The predictive models showed that the similarity of specific taxonomic clades in the patient-donor microbiota significantly contributed to long-term responder (p < 0.01).

Conclusion: Beneficial bacteria are depleted in patients with UC and A-FMT allows the colonization of healthy donor-derived microorganisms to compensate for their absence, contributing to their efficacy. This study provides important insights for the development of FMT-based therapy for UC and patient-tailored precision medicine.