Fecal microbiota transplantation to maintain remission in Crohn's disease: a pilot randomized controlled study

Abstract

Background: The role of the gut microbiota in Crohn's disease (CD) is established and fecal microbiota transplantation (FMT) is an attractive therapeutic strategy. No randomized controlled clinical trial results are available. We performed a randomized, single-blind, sham-controlled pilot trial of FMT in adults with colonic or ileo-colonic CD.

Method: Patients enrolled while in flare received oral corticosteroid. Once in clinical remission, patients were randomized to receive either FMT or sham transplantation during a colonoscopy. Corticosteroids were tapered and a second colonoscopy was performed at week 6. The primary endpoint was the implantation of the donor microbiota at week 6 (Sorensen index > 0.6).

Results: Eight patients received FMT and nine sham transplantation. None of the patients reached the primary endpoint. The steroid-free clinical remission rate at 10 and 24 weeks was 44.4% (4/9) and 33.3% (3/9) in the sham transplantation group and 87.5% (7/8) and 50.0% (4/8; one patient loss of follow-up while in remission at week 12 and considered in flare at week 24) in the FMT group. Crohn's Disease Endoscopic Index of Severity decreased 6 weeks after FMT (p = 0.03) but not after sham transplantation (p = 0.8). Conversely, the CRP level increased 6 weeks after sham transplantation (p = 0.008) but not after FMT (p = 0.5). Absence of donor microbiota engraftment was associated with flare. No safety signal was identified.

Conclusion: The primary endpoint was not reached for any patient. In this pilot study, higher colonization by donor microbiota was associated with maintenance of remission. These results must be confirmed in larger studies (NCT02097797). Video abstract.

Keywords: Crohn’s disease; Fecal microbiota transplantation; Randomized controlled trial.

Conflict of interest statement

HS received unrestricted study grants from Danone, Biocodex, and Enterome; board membership, consultancy, or lecture fees from Carenity, Abbvie, Astellas, Danone, Ferring, Mayoly Spindler, MSD, Novartis, Roche, Tillots, Enterome, Maat, BiomX, Biose, Novartis, and Takeda; and a co-funder of Exeliom bioscience. TS received unrestricted study grants from Amgen, AstraZeneca, Bayer, BMS, Boehringer-Ingelheim, Daiichi-Sankyo, Eli-Lilly, MSD, Pfizer, and Sanofi; and board membership, consultancy, or lecture fees from AstraZeneca, Astellas, BMS, MSD, Novartis, Sanofi, and Pfizer. The other authors declare that they have no competing interests.

Figures

Fig. 1

Flow chart of patients and primary endpoint. a Flow chart of patients included and excluded from analysis, according to CONSORT, Consolidated Standards of Reporting Trials. b Sorensen similarity index between donor and recipient fecal microbiota 6 weeks after FMT or sham. For b, the Wilcoxon rank-sum test was used

Fig. 2

Clinical efficacy of FMT in CD patients who achieved clinical remission with steroids. a Flare-free survival of patients in the FMT and the sham groups compared using log-rank test. b Steroid-free remission at week 10 after FMT or sham transplantation compared with Fisher’s exact test. Change in CDEIS (c) and CRP (d) between day 0 and week 6 for FMT and sham treatment groups, evaluated with the paired Wilcoxon test. One patient in each group was not evaluable for CDEIS because of a bowel-cleansing problem at week 6. One sample was not available for CRP in the FMT group at week 6

Fig. 3

Effect of FMT on fecal microbiota composition in recipients. a Change in Shannon and Chao1 alpha diversity indices compared to day 0. For donors, the change was calculated using the mean of the FMT group. b Principal coordinate analysis of Bray-Curtis distance. PC1, PC2, and PC3 represent the top three principal coordinates that captured most of the variance. Arrows connect samples from the same patient before (day 0) and 6 weeks after FMT or sham. Two different view angles of the same PCoA plot are shown for clarity. Groups were compared using the ANOSIM method (9999 permutations). c Sorensen similarity index between donor and recipient fecal microbiota 6 weeks after FMT or sham. Red dots are considered as “FMT failure.” For the sham group, the mean of the Sorensen with each donor was indicated. d Evolution of Sorensen similarity index between donor and recipient fecal microbiota. e, f Proportion of different OTUs absent in samples before FMT or sham (W2 and D0) and present after FMT or sham. For a, d, and f, two-way Anova with q < 0.1, according to a Benjamini-Hochberg FDR, was used. For c, the Wilcoxon rank-sum test was used. *p < 0.05; **p < 0.01; ***p < 0.001

Fig. 4

Evolution of the fecal microbiota beta-diversity in patients with successful or unsuccessful colonization by the donor microbiota. Principal coordinate analysis of Bray-Curtis distance in one patient with FMT success (a) and one patient with FMT failure (b). PC1, PC2, and PC3 represent the top three principal coordinates that captured most of the diversity. The fraction of diversity captured by the coordinate is given as a percentage

Fig. 5

Evolution of the fecal microbiota bacterial taxa with successful or unsuccessful colonization by the donor microbiota. Abundance of bacterial taxa (at the genus/species level) during the follow-up period in one patient with FMT success (a) and one patient with FMT failure (b). Only taxa representing > 0.1% of the microbiota were taken into account in the analysis

Fig. 6

Microbial predictors of clinical outcome. a Bacterial taxa differentially enriched at baseline (day 0) in patients with FMT success and FMT failure (generated using LeFSE analysis). b Bacterial taxa at week 6 in the whole study population associated with flare vs no flare before the end of follow-up. Results were generated using LeFSE analysis and only statistically significant (linear differential analysis scores > 2, p < 0.05) taxa are shown