Results of the First Pilot Randomized Controlled Trial of Fecal Microbiota Transplant In Pediatric Ulcerative Colitis: Lessons, Limitations, and Future Prospects

The incidence of childhood-onset inflammatory bowel disease (IBD) has increased significantly over the past decade.1Carroll M.W. Kuenzig M.E. Mack D.R. et al.The impact of inflammatory bowel disease in Canada 2018: children and adolescents with IBD.J Can Assoc Gastroenterol. 2019; 2: S49-S67Crossref PubMed Google Scholar Despite advances in therapy families remain concerned about long-term harms of immunosuppressive therapy and their associated costs.2Ledder O. Assa A. Levine A. et al.Vedolizumab in paediatric inflammatory bowel disease: a retrospective multi-centre experience from the Paediatric IBD Porto Group of ESPGHAN.J Crohns Colitis. 2017; 11: 1230-1237Crossref PubMed Scopus (73) Google Scholar The role of the enteric microbiome in IBD is well described and has been implicated in the pathogenesis of ulcerative colitis (UC).3Torres J. Danese S. Colombel J.-F. New therapeutic avenues in ulcerative colitis: thinking out of the box.Gut. 2013; 62: 1642-1652Crossref PubMed Scopus (64) Google Scholar While drug development continues to focus on modifying dysregulated intestinal immune pathways, targeting enteric microbiota has become increasingly attractive. Four randomized controlled trials (RCTs) of fecal microbiota transplant (FMT) in UC have been conducted since 2015. Three demonstrated clinical and endoscopic remission in adult patients with active UC, and a 2017 systematic review reported a pooled rate of clinical and endoscopic remission of 27.9%, with a number needed to treat of 5 (95% confidence interval [CI], 4–10).4Narula N. Kassam Z. Yuan Y. et al.Systematic review and meta-analysis: fecal microbiota transplantation for treatment of active ulcerative colitis.Inflamm Bowel Dis. 2017; 23: 1702-1709Crossref PubMed Scopus (149) Google Scholar Compelling data from adult UC has increased interest for pediatric UC, but no RCTs have been conducted in children. Three uncontrolled case series and case reports have yielded conflicting results with limited follow-up duration. Kunde et al5Kunde S. Pham A. Bonczyk S. et al.Safety, tolerability, and clinical response after fecal transplantation in children and young adults with ulcerative colitis.J Pediatr Gastroenterol Nutr. 2013; 56: 597-601Crossref PubMed Scopus (257) Google Scholar administered serial enemas containing healthy donor stool for 5 days to 9 patients with UC, ages 7 to 21, and 6 patients showed clinical response at the 1-month follow-up. Suskind et al6Suskind D.L. Brittnacher M.J. Wahbeh G. et al.Fecal microbial transplant effect on clinical outcomes and fecal microbiome in active Crohn's disease.Inflamm Bowel Dis. 2015; 21: 556-563Crossref PubMed Scopus (194) Google Scholar,7Kronman M.P. Nielson H.J. Adler A.L. et al.Fecal microbiota transplantation via nasogastric tube for recurrent Clostridium difficile infection in pediatric patients.J Pediatr Gastroenterol Nutr. 2015; 60: 23-26Crossref PubMed Scopus (68) Google Scholar reported 2 case series of FMT for Crohn's disease and UC in which a single FMT was administered via nasogastric tube to 4 patients with UC, with no clinical response seen. Kellermayer et al8Kellermayer R. Nagy-Szakal D. Harris R.A. et al.Serial fecal microbiota transplantation alters mucosal gene expression in pediatric ulcerative colitis.Am J Gastroenterol. 2015; 110: 604-606Crossref PubMed Scopus (61) Google Scholar reported 3 pediatric patients with UC who received serial FMT enemas and colonoscopic infusions, and all achieved clinical remission at weeks 2 and 4. Three further case reports reported an 18-month-old, 3-year-old, and 11-year-old with severe, treatment refractory colitis.9Vandenplas Y. Veereman G. van der Werff Ten Bosch J. et al.Fecal microbial transplantation in early-onset colitis: caution advised.J Pediatr Gastroenterol Nutr. 2015; 61: e12-e14Crossref PubMed Scopus (28) Google Scholar, 10Shimizu H. Arai K. Abe J. et al.Repeated fecal microbiota transplantation in a child with ulcerative colitis.Pediatr Int. 2016; 58: 781-785Crossref PubMed Scopus (23) Google Scholar, 11Kumagai H. Yokoyama K. Imagawa T. et al.Failure of fecal microbiota transplantation in a three-year-old child with severe refractory ulcerative colitis.Pediatr Gastroenterol Hepatol Nutr. 2016; 19: 214Crossref PubMed Scopus (15) Google Scholar All received serial FMT infusions over varying frequencies and routes of administration, and 2 showed response.12Wang A.Y. Popov J. Pai N. Fecal microbial transplant for the treatment of pediatric inflammatory bowel disease.World J Gastroenterol. 2016; 22: 10304-10315Crossref PubMed Scopus (28) Google Scholar Current evidence for FMT in UC offers reasons for both optimism and further study. Adult RCTs demonstrate that FMT leads to significant rates of clinical and endoscopic healing. One study suggests FMT may have a particular role in patients with new-onset disease duration. Colonic FMT appears more effective for the treatment of UC than upper-tract delivery, and serial, multidose treatments correlate with greatest efficacy in both adult and pediatric recipients.4Narula N. Kassam Z. Yuan Y. et al.Systematic review and meta-analysis: fecal microbiota transplantation for treatment of active ulcerative colitis.Inflamm Bowel Dis. 2017; 23: 1702-1709Crossref PubMed Scopus (149) Google Scholar,12Wang A.Y. Popov J. Pai N. Fecal microbial transplant for the treatment of pediatric inflammatory bowel disease.World J Gastroenterol. 2016; 22: 10304-10315Crossref PubMed Scopus (28) Google Scholar An RCT of FMT in pediatric patients with UC has never been reported, and there are multiple feasibility questions associated with running a placebo-controlled pediatric FMT trial. In 2015, after publication of the first RCT by adult gastroenterology colleagues from our center and strong interest from families,13Moayyedi P. Surette M.G. Kim P.T. et al.Fecal microbiota transplantation induces remission in patients with active ulcerative colitis in a randomized controlled trial.Gastroenterology. 2015; 149: 102-109.e6Abstract Full Text Full Text PDF PubMed Scopus (1108) Google Scholar our team launched the first pilot trial of FMT in pediatric UC. Our pilot feasibility protocol has been reported previously.14Pai N. Popov J. Protocol for a randomised, placebo-controlled pilot study for assessing feasibility and efficacy of faecal microbiota transplantation in a paediatric ulcerative colitis population: PediFETCh trial.BMJ Open. 2017; 7e016698Crossref PubMed Scopus (22) Google Scholar We discuss our findings and experience conducting the Pediatric FEcal microbiota Transplant in ulcerative Colitis (PediFETCh) Trial and important lessons that may support future investigators. During a 36-month study period, 48 patients, aged 4 to 17 years, with active UC were referred by the patients' primary pediatric gastroenterologists for screening across 3 pediatric IBD centers in Canada (McMaster Children's Hospital, Hamilton, Ontario; Children's Hospital at London Health Sciences Centre, London, Ontario; and Centre Hospitalier Universitaire Sainte-Justine, Montreal, Quebec). Of these, 23 patients were excluded: 30.4% (n = 7) had medication changes made by their primary gastroenterologist within 4 weeks of trial entry, 34.8% (n = 8) were unable to fulfill trial requirements of in-hospital FMT treatments, 21.7% (n = 5) withdrew interest for unspecified reasons, and 13.0% (n = 3) were in remission by trial commencement. Ultimately, 52.1% (n = 25) of referred patients entered the trial. Patients were randomized to FMT (n = 13) or placebo (n = 12) arms; 46% of patients in the trial were diagnosed with UC for less than 1 year. At diagnosis, 77% (10 of 13) in the FMT arm had pancolitis vs 58% (7 of 12) in the placebo arm. During the trial, 54% (7 of 13) in the FMT arm received antitumor necrosis factor (TNF) (n = 3) or immunomodulator (IM) (n = 4) therapies vs 17% (2 of 12; 1 anti-TNF, 1 IM) in the placebo arm. Seven patients randomized to the placebo arm crossed over to the open-label arm (Figure 1). We did not reach our primary feasibility outcome of achieving recruitment targets (50 patients over 2 years). We reached our composite clinical end point (improvement in pediatric UC activity index, C-reactive protein, or fecal calprotectin) in 92% (11 of 12) assigned to FMT vs 50% (6 of 12) assigned to placebo at week 6 (risk ratio, 1.8; 95% CI, 1.1–3.7). At 12 months, 75% (9 of 12) had maintained clinical response (Table 1). β-Diversity trended higher from baseline to week 6 in FMT vs placebo arms (Supplementary Figure 1). Several bacterial taxa were associated with achieving the composite clinical outcome after multiple test correction, including Alistipes spp and Escherichia spp (Supplementary Table 1).Table 1Secondary Outcome Measures Comparing Fecal Microbiota Transplant vs Placebo ArmsVariableFMT (n = 12)Placebo (n = 12)RR95% CIComposite clinical outcomeaComposite clinical outcome defined as any improvement from baseline of at least 1 of PUCAI, FC, or CRP; improvements in individual outcomes further described in rows below.11 (91.7)6 (50.0)1.8b9 stool samples from FMT group, 7 stool samples from placebo group.1.1–3.7Clinical response, week 6cClinical response defined as improvement from baseline PUCAI.7 (58.3)4 (33.3)1.80.7–4.6FC response week 6dFC response defined as improvement from baseline FC.8 (88.9)b9 stool samples from FMT group, 7 stool samples from placebo group.3 (42.9)b9 stool samples from FMT group, 7 stool samples from placebo group.2.11.0–5.7CRP response week 6eCRP response defined as improvement from baseline CRP.8 (66.7)2 (18.2)4.01.3–14.8Clinical response week 30cClinical response defined as improvement from baseline PUCAI.6 (50.0)3 (25.0)2.00.7–6.3FC response week 30dFC response defined as improvement from baseline FC.6 (85.7)f7 stool samples from FMT group, 8 stool samples from placebo group.2 (25.0)f7 stool samples from FMT group, 8 stool samples from placebo group.3.41.2–12.2CRP response week 30eCRP response defined as improvement from baseline CRP.5 (41.7)3 (25.0)1.50.5–5.0Clinical remission week 6gClinical remission defined as PUCAI <15.4 (33.3)4 (33.3)1.00.3–3.0Clinical remission week 30gClinical remission defined as PUCAI <15.5 (41.7)4 (33.3)1.30.5–3.5FC <250 μg/g week 305 (71.4)f7 stool samples from FMT group, 8 stool samples from placebo group.4 (50.0)f7 stool samples from FMT group, 8 stool samples from placebo group.1.40.6–3.6FC change week 6hReported as change (+/−) from baseline.−881.1b9 stool samples from FMT group, 7 stool samples from placebo group.−390.4b9 stool samples from FMT group, 7 stool samples from placebo group.……CRP change week 6hReported as change (+/−) from baseline.+0.1+5.7……FC change week 30hReported as change (+/−) from baseline.−1282.9f7 stool samples from FMT group, 8 stool samples from placebo group.−583.2f7 stool samples from FMT group, 8 stool samples from placebo group.……CRP change week 30hReported as change (+/−) from baseline.−1.3+7.1……Patients with serious adverse events5i3 patients developed worsening colitis requiring hospitalization for intravenous methylprednisolone; 2 patients with a past history of Clostridioides difficile colitis were diagnosed with Clostridioides difficile colitis within 2 weeks of discontinuing therapy. (41.7)1j1 patient developed worsening colitis requiring hospitalization for intravenous methylprednisolone. (8.3)5.01.0–30.2NOTE: Categorical data are shown as n (%) and continuous data as mean values. Continuous data were measured using the independent sample t test, and categorical data were measured using Fisher's exact test. Missing values were excluded. All analyses are modified intention to treat.CI, confidence interval; CRP, C-reactive protein; FC, fecal calprotectin; FMT, fecal microbiota transplant; PUCAI, Pediatric Ulcerative Colitis Activity Index; RR, risk ratio.a Composite clinical outcome defined as any improvement from baseline of at least 1 of PUCAI, FC, or CRP; improvements in individual outcomes further described in rows below.b 9 stool samples from FMT group, 7 stool samples from placebo group.c Clinical response defined as improvement from baseline PUCAI.d FC response defined as improvement from baseline FC.e CRP response defined as improvement from baseline CRP.f 7 stool samples from FMT group, 8 stool samples from placebo group.g Clinical remission defined as PUCAI <15.h Reported as change (+/−) from baseline.i 3 patients developed worsening colitis requiring hospitalization for intravenous methylprednisolone; 2 patients with a past history of Clostridioides difficile colitis were diagnosed with Clostridioides difficile colitis within 2 weeks of discontinuing therapy.j 1 patient developed worsening colitis requiring hospitalization for intravenous methylprednisolone. Open table in a new tab NOTE: Categorical data are shown as n (%) and continuous data as mean values. Continuous data were measured using the independent sample t test, and categorical data were measured using Fisher's exact test. Missing values were excluded. All analyses are modified intention to treat. CI, confidence interval; CRP, C-reactive protein; FC, fecal calprotectin; FMT, fecal microbiota transplant; PUCAI, Pediatric Ulcerative Colitis Activity Index; RR, risk ratio. There was a numeric increase in at least 1 adverse event occurring in the FMT (83.3% [10 of 12]) vs placebo group (41.7% [5 of 12]) during the 6-week intervention period (P = 0.68; Table 1). Worsening colitis developed in 4 patients (3 FMT, 1 placebo) and required hospitalization for intravenous methylprednisolone, and 2 FMT patients with a prior history of Clostridioides difficile (C difficile) colitis were diagnosed with C difficile colitis within 2 weeks of trial withdrawal due to lack of improvement. Aliquots of FMT donor stools were retested in these patients and confirmed the absence of C difficile in donor samples. Low sample sizes prevented subgroup analyses of potential risk factors for adverse events, such as concurrent medications. Secondary outcomes compared all recipients of FMT (randomized plus open-label placebo crossover arms) vs placebo (Supplementary Table 2). In this cohort, the composite clinical end point was reached in 84% (16 of 19) assigned to FMT vs 50% (6 of 12) assigned to placebo at week 6 (risk ratio, 1.7; 95% CI, 1.0–3.4). Among eligible patients, 34.8% (n = 8) declined participation due to requirements to attend 12 biweekly in-hospital visits for rectal enema treatments. A further 21.7% (n = 5) declined enrolment for unspecified reasons after learning more about the trial protocol from study personnel. Our study attempted to maximize potential benefit through an intensive administration protocol, but this also compromised recruitment. We conducted qualitative follow-up interviews with participants, and intensive administration protocols were reported as a barrier.15Popov J, Hartung E, Hill L, et al. Pediatric patient and parent perceptions of fecal microbiota transplantation for the treatment of ulcerative colitis. J Pediatr Gastroenterol Nutr. Published online November 20, 2020. https://doi.org/10.1097/MPG.0000000000002995Google Scholar Future studies could reduce the number of rectal enema infusions or use home-based interventions, such as offering self-administered FMT enemas or lyophilized, oral FMT capsules. Our recruitment target was based on 1 of 2 RCTs reported at the time of our study's conception. Our results may support more precise power calculations to establish sample size in future trials. Current data suggest that FMT trials should be powered assuming a 20% response in the FMT arm and 5% response in the control arm, provided a stringent definition of remission is used as a primary end point. Our eligibility criteria also limited recruitment. These criteria required patients to have active disease but also be clinically stable, defined as <4 weeks between medication changes. In several instances where patients and families were permissive of ongoing, active symptoms before trial entry, primary providers encouraged the initiation of evidence-based treatments such as biologic therapies. This had several impacts: patients enrolled in our study either had mild or severe disease that had failed multiple biologics. Patients were also frequently withdrawn from the trial by the primary clinician before meeting formal study end points, if they did not show immediate signs of improvement. Placebo-controlled clinical trials in pediatrics can be ethically challenging for investigators and primary providers.15Popov J, Hartung E, Hill L, et al. Pediatric patient and parent perceptions of fecal microbiota transplantation for the treatment of ulcerative colitis. J Pediatr Gastroenterol Nutr. Published online November 20, 2020. https://doi.org/10.1097/MPG.0000000000002995Google Scholar When our study began in 2015, limited data were available to share with prospective families. As further safety and efficacy data were published, interest significantly increased. Early optimism from studies like ours may encourage greater confidence from primary providers and patients of pediatric FMT trials in the future. Our study was also impacted by high withdrawal rates from patients in our control group, limiting opportunities for better placebo comparisons. Clearer expectations of participants in RCTs and stringent withdrawal criteria may improve analyses in future trials.16Kahn S.A. Rubin D.T. When subjects violate the research covenant: lessons learned from a failed clinical trial of fecal microbiota transplantation.Am J Gastroenterol. 2016; 111: 1508-1510Crossref PubMed Scopus (8) Google Scholar An important consideration for performing FMT is the selection of appropriate donors. The microbiome of children and adults demonstrate considerable variation in composition and function. The pediatric intestinal microbiome is age-dependent and characterized by decreased diversity, increased representation of Bacteroides and glycan synthesis, and predominance of catabolic pathways. The adult microbiota has higher abundance of Blautia, greater carbohydrate degradation, and a shift to anabolic pathways. These differences may support cell division and growth during childhood.17Radjabzadeh D. Boer C.G. Beth S.A. et al.Diversity, compositional and functional differences between gut microbiota of children and adults.Sci Rep. 2020; 10: 1040Crossref PubMed Scopus (74) Google Scholar While evidence for the impact of inoculating children with adult stool is lacking, the long-term effects of such alterations to commensal microbiota on growth and metabolism will be an important outcome in future FMT trials. Our analysis of donor effect was affected by our use of multiple donors during the intervention phase. While samples from multiple donors were never pooled for any single treatment, samples from different donors were given to the same patient at different time points in the study. This approach was influenced by the desire to minimize potential "donor effect," whereby samples from 1 donor may confer disproportionate benefits. We collaborated with an experienced third-party microbiome company to optimize safety and logistics.18Rebiotix Inc. Microbiota Restoration Therapy.http://www.rebiotix.com/Google Scholar Donor samples were prescreened, approved by federal regulatory bodies, and banked for retesting. This was useful as testing requirements for multidrug-resistant organisms evolved throughout our trial.19U.S. Food and Drug AdministrationImportant Safety Alert Regarding Use of Fecal Microbiota for Transplantation and Risk of Serious Adverse Reactions Due to Transmission of Multi-Drug Resistant Organisms. June 13, 2019.https://www.fda.gov/vaccines-blood-biologics/safety-availability-biologics/important-safety-alert-regarding-use-fecal-microbiota-transplantation-and-risk-serious-adverseGoogle Scholar However, there are also theoretical safety concerns with this approach. If 1 donor harbors a yet unidentified pathogen, there is potential to cause wider harm than if a single donor were used for a single recipient. There are also questions of which control to use in FMT trials. Some trials have used autologous stool as a placebo so that the person administering the product can be appropriately masked.4Narula N. Kassam Z. Yuan Y. et al.Systematic review and meta-analysis: fecal microbiota transplantation for treatment of active ulcerative colitis.Inflamm Bowel Dis. 2017; 23: 1702-1709Crossref PubMed Scopus (149) Google Scholar However, microbial shifts occur when autologous stool is administered, and this could have an unknown impact on UC (therapeutic, neutral, or harmful), making results difficult to interpret. We used normal saline as a placebo in our trial. Appropriate double can be with this approach an product as a The sample or bacterial is across FMT UC as is of et in and future for the use of faecal microbiota transplant in the treatment of inflammatory bowel Gastroenterol. 2019; 12 PubMed Scopus Google Scholar We administered a of (50 over 12 infusions to all This approach was to treatments and influenced by a adult trial at our P. Surette M.G. Kim P.T. et al.Fecal microbiota transplantation induces remission in patients with active ulcerative colitis in a randomized controlled trial.Gastroenterology. 2015; 149: 102-109.e6Abstract Full Text Full Text PDF PubMed Scopus (1108) Google Scholar There is and to We administered FMT but in this may have been trials administered FMT less efficacy to a trial that administered FMT 5 N. Kassam Z. Yuan Y. et al.Systematic review and meta-analysis: fecal microbiota transplantation for treatment of active ulcerative colitis.Inflamm Bowel Dis. 2017; 23: 1702-1709Crossref PubMed Scopus (149) Google Scholar therapy is and trials have end points at 8 In our trial, to use fecal testing as a of mucosal healing. However, our lack of endoscopic outcomes was a and was impacted by and to for research Adult RCTs assessing the efficacy of have used endoscopic pediatric biologic trials have described J. et and therapy with for children with to severe ulcerative Gastroenterol 10: Full Text Full Text PDF PubMed Scopus Google R. et as and therapy for ulcerative J 2019; PubMed Scopus Google Scholar efficacy of pediatric FMT for UC future trials will to endoscopic outcomes Pediatric UC FMT trials adolescents may with for pediatric Full Text Full Text PDF PubMed Scopus Google Scholar FMT for pediatric C difficile treatment has shown rates of adverse S. et al.Fecal microbiota transplantation for recurrent Clostridium difficile infection and in Pediatr Gastroenterol Nutr. 2019; PubMed Scopus Google Scholar The safety of FMT in pediatric IBD or in pediatric C difficile with IBD is less S. R. et al.Fecal microbiota transplant for recurrent Clostridium difficile infection in pediatric inflammatory bowel Pediatr Gastroenterol Nutr. 2019; PubMed Scopus Google Scholar In our a higher of serious adverse in patients active (FMT) treatment, which has not been demonstrated across N. Kassam Z. Yuan Y. et al.Systematic review and meta-analysis: fecal microbiota transplantation for treatment of active ulcerative colitis.Inflamm Bowel Dis. 2017; 23: 1702-1709Crossref PubMed Scopus (149) Google Scholar Our rates may have been impacted by our case hospitalization was as a including if patients were to intravenous treatments, if this was as treatment of active colitis. of C difficile was also as a unable to a recurrent from an inflammatory UC C difficile rates have been described in children with S. R. et al.Fecal microbiota transplant for recurrent Clostridium difficile infection in pediatric inflammatory bowel Pediatr Gastroenterol Nutr. 2019; PubMed Scopus Google Scholar Our 2 patients who developed C difficile had a prior history of C difficile Patients in the FMT arm also showed greater baseline disease 77% vs or 54% vs than patients in the placebo arm. We were unable to for disease history of C difficile or medication use in our trial, but baseline differences between study arms impacted our As more pediatric centers are performing FMT for the treatment of C difficile share our experience of FMT in pediatric IBD to increase confidence patients and primary providers to support future trials. This may be further by with adult gastroenterology centers to share the of FMT and and federal regulatory trials with pediatric study arms could of these This first pilot RCT suggests that recruitment the significant A number of centers are required to but pediatric FMT studies are also affected by the of the intervention and regulatory needed to these trials. in and safety trials also for the of adverse Despite the this our trial offers the first pilot RCT evidence that FMT may have an important role in symptoms and inflammatory in pediatric UC. data suggest response may be for to weeks after the FMT and microbial changes may improvements in clinical response and mucosal associated with We did not reach our primary feasibility outcome of achieving recruitment These findings are and data on efficacy and adverse are learned from this first placebo-controlled trial will support future pediatric in this approach to pediatric IBD We future trials assessing FMT for pediatric UC should endoscopic to better mucosal microbiome functional and studies to further of the role of FMT in UC Pediatric Fecal Microbiota Transplant and from the Health of of and Hamilton, of and of Montreal, Health of Hamilton, of Clinical of and Hamilton, Health of Hamilton, of Health and Hamilton, and of Hamilton, and to study and and were for recruitment and screening of all patients, with support provided by and was for recruitment and screening of patients at Centre Fecal microbiota were administered by and provided support and study the of the with from and provided review and of the of the The no This was an study through received from the Health Sciences New Health Sciences and of Health and and A for a which provided the therapeutic and placebo to this and our had no role in study data data data or of the Pai had to all the data in the study and had for the to the