Randomized clinical trial to evaluate the effect of fecal microbiota transplant for initial Clostridium difficile infection in intestinal microbiome

Abstract

Objective: The aim of this study was to evaluate the impact of fecal donor-unrelated donor mix (FMT-FURM) transplantation as first-line therapy for C. difficile infection (CDI) in intestinal microbiome.

Methods: We designed an open, two-arm pilot study with oral vancomycin (250mg every 6 h for 10-14 days) or FMT-FURM as treatments for the first CDI episode in hospitalized adult patients in Hospital Universitario "Dr. Jose Eleuterio Gonzalez". Patients were randomized by a closed envelope method in a 1: 1 ratio to either oral vancomycin or FMT-FURM. CDI resolution was considered when there was a reduction on the Bristol scale of at least 2 points, a reduction of at least 50% in the number of bowel movements, absence of fever, and resolution of abdominal pain (at least two criteria). From each patient, a fecal sample was obtained at days 0, 3, and 7 after treatment. Specimens were cultured to isolate C. difficile, and isolates were characterized by PCR. Susceptibility testing of isolates was performed using the agar dilution method. Fecal samples and FMT-FURM were analyzed by 16S rRNA sequencing.

Results: We included 19 patients; 10 in the vancomycin arm and 9 in the FMT-FURM arm. However, one of the patients in the vancomycin arm and two patients in the FMT-FURM arm were eliminated. Symptoms resolved in 8/9 patients (88.9%) in the vancomycin group, while symptoms resolved in 4/7 patients (57.1%) after the first FMT-FURM dose (P = 0.26) and in 5/7 patients (71.4%) after the second dose (P = 0.55). During the study, no adverse effects attributable to FMT-FURM were observed in patients. Twelve isolates were recovered, most isolates carried tcdB, tcdA, cdtA, and cdtB, with an 18-bp deletion in tcdC. All isolates were resistant to ciprofloxacin and moxifloxacin but susceptible to metronidazole, linezolid, fidaxomicin, and tetracycline. In the FMT-FURM group, the bacterial composition was dominated by Firmicutes, Bacteroidetes, and Proteobacteria at all-time points and the microbiota were remarkably stable over time. The vancomycin group showed a very different pattern of the microbial composition when comparing to the FMT-FURM group over time.

Conclusion: The results of this preliminary study showed that FMT-FURM for initial CDI is associated with specific bacterial communities that do not resemble the donors' sample.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1. Flowchart of the study design.

Fig 2. Resolution rates (%) of initial C. difficile infection in patients treated with vancomycin and fecal microbiota transplantation (FMT).

Fig 3. Bar charts showing the relative abundance of OTUs at the phylum level.

Distribution of bacterial phyla in the fecal microbiota of transplantation (FMT) group and the vancomycin group (Vanco). The label of each sample (e.g., FMT-2-0) denotes Treatment (FMT or Vancomycin), the assigned number of each patient, and days on treatment (0 = baseline, 3 = 3 days on treatment, 7 = 7 days on treatment). To facilitate comparison and visualization, the distribution of bacterial phyla of the donors’ pool is presented for the FMT group (Donors’ pool). Please note that the border of the three most abundant groups was highlighted for better visualization and the bars were organized based on the abundance of the highest abundant group (i.e., Firmicutes).

Fig 4. Average bacterial composition at the phylum level.

FMT: fecal microbiota transplantation.

Fig 5. Bar charts showing the relative abundance of OTUs at the family level.

Distribution of bacterial families in the fecal microbiota of transplantation (FMT) group and the vancomycin group (Vanco). The label of each sample (e.g., FMT-5-0) denotes Treatment (FMT or Vancomycin), the assigned number of each patient, and days on treatment (0 = baseline, 3 = 3 days on treatment, 7 = 7 days on treatment). To facilitate comparison and visualization, the distribution of bacterial families of the donors’ pool is presented for the FMT group (Donors’ pool). Please note that the border of the three most abundant groups was highlighted for better visualization and the bars were organized based on the abundance of the highest abundant group (i.e., Enterobacteriaceae).

Fig 6. Number of observed species.

The figure shows the number of observed species (i.e., operational taxonomic units at 97% similarity, y-axis) and the number of sequences per sample (x-axis) for the FMT group and Vancomycin group across collection days of fecal samples. Error bars represent standard deviations.

Fig 7. PCoA plots of weighted UniFrac distances for the FMT group and the vancomycin group.

PCoA plots of weighted UniFrac distances for both groups across the days of the collection of fecal samples. There was no significant clustering of microbial communities according to time point (please note that the analysis of unweighted UniFrac distances did not reveal any significant clustering of microbial communities either).