Faecal microbiota transplantation in Ulcerative Colitis: A randomised controlled trial

Background An aberrant microbiota has been implicated in the pathophysiology of ulcerative colitis (UC) and several case series reported favorable effects of faecal microbiota transplantation (FMT). We aimed to assess the efficacy FMT in active UC patients in a randomized parallel group study. Methods Patients with mild-moderate UC (Simple Clinical Colitis Activity Index [SCCAI] score of 4-11 and a Mayo endoscopic score of at least 1) were randomly assigned (1:1) to FMT derived from faeces from a healthy donor: FMT-D or their own faeces used as a placebo: FMT-P. FMT was administered via a naso-duodenal tube at week 0 and 3. Patients, physicians and endoscopists were blinded, with exception of the nurse who performed randomisation and prepared the faeces filtered with normal saline. The composite primary endpoint was clinical remission (SCCAI score 2 or lower) and endoscopic response (at least 1 point decrement on the Mayo endoscopic score) at week 12. Main secondary endpoints were microbiota composition in faecal samples and safety. Results Of 50 patients who were screened, 48 were randomised (23 to FMT-D and 25 to FMT-P. In the ITT analysis, seven out of 23 patients (30.4%) in the donor arm versus five out of 25 patients in the placebo arm (20.0%) achieved the primary endpoint, P= .51. In the per protocol analysis 37 patients completed endoscopic follow-up; seven out of 17 patients in the donor arm (41.2%) versus five out of 20 (25.0%) in the placebo arm achieved the primary endpoint, P= .29. The majority of patients experienced mild adverse events with spontaneous recovery. Serious adverse events occurred in four patients and were considered not related to FMT. The trial was terminated after 50 inclusions due to futility. Redundancy analysis showed that at 12 weeks faecal microbiota of responders in the FMT-D group overlapped with healthy donors, which was associated with occurrence of Clostridium cluster XIVa. This was not seen in responders from the FMT-P group. Conclusion FMT with faeces of a healthy donor did not result in statistically higher clinical and endoscopic remission rates as compared to placebo in moderately active unselected UC patients. However, both FMT(donor) and FMT(placebo) are associated with signature changes in responders. Future studies should focus on mode of administration, matching of for selected patients, as well as the observed shifts in microbiota composition in responders. ClinicalTrials.gov number NCT01650038.