Fecal microbiota transplantation (FMT) from healthy donors was no more effective than placebo for moderate-to-severe irritable bowel syndrome (IBS) in a phase 3 randomized trial that its authors described as the largest of its kind, published in The Lancet. At 90 days, 40% of FMT recipients and 38% of placebo recipients met the trial’s threshold for symptom improvement.
“I think these findings tell us that it is not enough to alter the gut microbiota towards healthy people to relieve symptoms in patients with irritable bowel syndrome,” explained Dr. Frederik Emil Juul, MD, PhD, the study’s co-author and postdoctoral fellow at the Clinical Effectiveness Research Group, Oslo University Hospital, Norway. “Still, the gut microbiota in these patients is altered compared to healthy individuals. The challenge is that we do not understand the mechanisms of the alterations.”
The double-blind REFIT2 trial enrolled 450 adults aged 18 to 65 years with moderate-to-severe IBS, defined by Rome IV criteria and an IBS-Severity Scoring System (IBS-SSS) score of 175 or higher on a 500-point scale, at five hospitals in Norway between May 2021 and July 2022. Participants had a median age of 36, and 65% were women. They were randomly assigned to receive stool from one of three screened healthy donors or placebo consisting of their own stool, for an overall 2:1 ratio; the authors described it as the first IBS trial to randomize participants by donor. After bowel cleansing, treatment was given once by rectal enema. The primary endpoint was the proportion of participants with a reduction of 75 points or more on the IBS-SSS at 90 days, and follow-up continued for one year.
Among the 448 participants included in the analysis, 40% of those receiving donor FMT met the primary endpoint compared with 38% in the placebo group, a difference that was not statistically significant. The absence of benefit persisted at one year. At 365 days, 44% of participants in the intervention group and 45% in the placebo group achieved at least a 75-point improvement in IBS-SSS.
No significant improvements were seen in other clinical outcomes, including adequate relief, defined by both global symptom improvement and at least a 30% reduction in abdominal pain (20% vs. 18%); stool consistency among patients with diarrhea-predominant IBS; or complete spontaneous bowel movements among those with constipation-predominant disease.
Donor FMT did not increase the proportion of participants who gained 14 points or more on an IBS quality-of-life measure at 90 days or one year. A repeated-measures analysis did show a higher average quality-of-life score and fewer participants with mood disorder in the FMT group, though the authors did not discuss these results and concluded that the trial showed no clinical benefit.
Over one year, there were 24 adverse events in the FMT group, including six serious events, and eight in the placebo group, none serious. Events judged possibly or probably related to treatment occurred in four FMT recipients and three placebo recipients; none were serious, and no participants died.
In exploratory analyses, recipients in all three donor groups showed significant shifts in gut microbial composition toward their assigned donor’s profile at 90 days, while the placebo group showed none. However, differences in engraftment between donors did not translate into differences in symptom response.
The absence of clinical effect despite engraftment suggests that altering overall gut microbiota composition through FMT is not sufficient to relieve IBS symptoms, the authors wrote. The results also did not support the hypothesis that certain “super donors” produce better outcomes: 40%, 41%, and 39% of recipients of the three donors met the primary endpoint.
“If the gut microbiota in patients with IBS truly is a target that can lead to symptom relief, we need to be much more targeted” than today’s donor selection allows, Dr. Juul said. He added that the treatment can still be “highly effective in patients with [Clostridioides difficile] infection.”
Thirty-eight percent of placebo recipients met the primary endpoint at 90 days, in line with the roughly 40% placebo response in previous IBS trials, the authors noted, even though they used a stricter responder threshold than the commonly used 50 points. Response rates varied markedly across study sites, they wrote, though site sample sizes were small.
“The placebo response was indeed substantial, which is known from earlier IBS trials,” confirmed Dr. Juul. “We do not have a good explanation for the driver of this effect, but do think it is essential to conduct double blinded trials when investigating new treatments in IBS patients ... to see if the new treatment is indeed better than placebo.”
The study concluded that nonspecific microbiota replacement with FMT is unlikely to provide consistent therapeutic benefit for IBS, although the gut microbiota may still play a role in the disorder’s pathophysiology. A 2024 AGA clinical practice guideline suggests against conventional FMT as a treatment for IBS except in clinical trials.
“We will continue our research on the pathogenesis of IBS by analyses of blood, stool and urine samples, as well as dietary records, from all participants in the REFIT2 trial [and] hopefully this will give us a better understanding of the relationship between gut microbiota and IBS symptoms,” said Dr. Juul.
The authors noted that outcomes relied on patient-reported measures collected as weekly recall rather than daily diaries, which could introduce recall bias, though any such bias would have affected both groups equally. Participants volunteered through an online screening questionnaire, and only 19% of applicants were enrolled, raising the possibility of self-selection bias. The trial also lacked functional microbial analysis and collected no dietary information beyond whether participants followed a low-FODMAP diet at baseline. The authors added that delivery through the upper gastrointestinal tract, which the trial did not test, might affect immune activation differently than enema delivery.
The trial was funded by KLINBEFORSK. Dr. Juul reported giving a pro bono talk at a Tillotts Pharma-organized meeting; co-authors reported lecture fees from Takeda, congress funding from Pharmacosmos, data monitoring committee roles, and travel reimbursement from the Norwegian Society for Clinical Nutrition. The other authors reported no competing interests.