FMT (Fecal Microbiota Transplantation) Super-donors and Abandoning the “One Stool Fits All” Approach

Not all stools were created equal

There was a comment recently left on this blog posing the question of what makes a good donor for FMT (Fecal Microbiota Transplantation), or a “poop transplant” in plain English.

FMT is actually an approved therapy for Clostridioides difficile infection (CDI). Research has shown  FMT to be more effective than the antibiotic vancomycin. To quote from the research, “The infusion of donor feces was significantly more effective for the treatment of recurrent C. difficile infection than the use of vancomycin”.

FMT might not be for discussion at the dinner table, but it is highly effective in some instances.

FMT is actually far more widely used than you might imagine.  In one of today’s papers from China they had treated 1,387 people using 20 donors, for a wide variety of conditions.

In the US, autism researchers at Arizona State University showed a benefit that was maintained after a period of two years.

Autism symptoms reduced nearly 50 percent two years after fecal transplant

At two years post-treatment, most of the initial improvements in gut symptoms remained. In addition, parents reported a slow steady reduction of ASD symptoms during treatment and over the next two years. A professional evaluator found a 45% reduction in core ASD symptoms (language, social interaction and behavior) at two years post-treatment compared to before treatment began.

An earlier study with only vancomycin (an antibiotic) had found major temporary improvements in GI and autism symptoms, but the benefits were lost a few weeks after treatment stopped despite use of over-the-counter probiotics.

The obvious question to ask is whether FMT has a potential benefit to people with autism who do not have GI dysfunction.  I think this question is far from being answered.

We have seen in earlier posts that modifying the microbiome has great potential to fine-tune the function of the brain.  Researchers at UCLA showed that the high fat ketogenic diet controls epileptic seizures not through the action of ketones in the brain, but via the high fat intake changing the mix of bacteria in the gut.

The Gut Microbiota Mediates the Anti-Seizure Effects of the Ketogenic Diet

FMT is just one way to modify the microbiome.  The UCLA researchers are developing a medical food to produce similar effects on the microbiome as the ketogenic diet.

Very likely a personalized bacteria transfer, customized to the symptoms of the person, might effectively treat many more conditions than just GI problems.  

It does look likely that for some conditions there may be super-donors, people whose microbiome is particularly effective, when transferred to others.

But the research cautions against what is called the “One Stool Fits All” Approach.  The donor and recipient need to be “compatible”.

The Super-Donor Phenomenon in Fecal Microbiota Transplantation

The microbial diversity of the donor is a good predictor of FMT success in the recipient. However, donor-recipient compatibility also plays an influential role in determining FMT success. Donor-recipient compatibility can stem from genetic factors such as differences in innate immune responses, or environmental factors including diet, xenobiotic exposure, and microbial interactions.

FMT for Inflammatory Bowel Disease (IBD): The Emergence of the FMT Super-Donor

IBD encompasses both Crohn's disease and ulcerative colitis; two debilitating disorders characterized by chronic relapsing inflammation of the intestinal. In contrast to CDI, there is no evidence that IBD results from an overgrowth of one specific pathogen. Rather, the disease is likely brought on by complex interactions involving the host's genetics, immune system, and gut microbiota. Both Crohn's disease and ulcerative colitis are broadly characterized by a reduced diversity of the gut microbiota with lower relative abundances of the Bacteroidetes and Firmicutes phyla and higher proportions of Proteobacteria. A specific reduction in the abundance of butyrate-producing bacterial species, particularly Faecalibacterium prausnitzii, has been observed for both Crohn's disease and ulcerative colitis. Meanwhile, for Crohn's disease, an increase in a pro-inflammatory form of Escherichia coli has also been reported.

The first successful case report of an FMT for the treatment of IBD was published in 1989 when a male with refractory ulcerative colitis achieved clinical remission for 6 months following a retention enema with healthy donor stool. Subsequently, a large number of FMT studies have been conducted on IBD patients with variable clinical outcomes, remission rates, and longevity of effect. Recently, Paramsothy et al. performed a systematic review and meta-analysis of 53 studies (four RCT, 30 cohort, 19 case studies) of FMT in IBD patients. Avoiding publication bias, their analysis of cohort studies revealed FMT was more effective at inducing remission in Crohn's disease patients when compared to patients with ulcerative colitis (52 vs. 33%, respectively). With regard to ulcerative colitis, a larger number of FMT infusions and a lower gastrointestinal tract administration were associated with improved rates of remission.

In contrast to studies of CDI, FMT studies conducted on IBD patients have frequently identified differential recipient responses that have been associated with variability in the donor stool. Currently, the stool used for FMT is not standardized in terms of donor selection (related vs. unrelated), preparation (fresh vs. frozen, aerobic vs. anaerobic), or the dose that is administered (single vs. multiple doses). While inconsistencies in FMT protocols make it difficult to compare different studies, there is a large degree of variability in clinical responses to FMT between recipients who have been subjected to the same study design. It is unfortunate that information on a recipient's genetic background or dietary intake is not yet routinely assessed, particularly given that some instances of IBD have an underlying genetic component. Due to the lack of genetic information, investigators have instead focused on the donor-dependent effect and proposed the existence of so called super-donors to explain the variation in recipient responses.

The first study to record the super-donor effect was a randomized control trial that was investigating the efficacy of FMT for inducing clinical remission in patients with ulcerative colitis. Moayyedi et al. assigned 75 patients with active disease to weekly enemas containing either fecal material or water (placebo) for a period of 6 weeks. FMT was shown to be superior to the placebo, resulting in significantly higher rates of endoscopic and clinical remission, albeit of modest effect (24 vs. 5%, respectively), after 7 weeks. Of the nine patients who entered remission, seven had received FMT from the same donor. Thus, it was argued that FMT success was donor-dependent.

Currently, it is not possible to predict the clinical efficacy of a donor before FMT in IBD patients. It has been suggested that remission rates could be improved by pooling donor's stool together, limiting the chances a patient will receive only ineffective stool. This stool pooling approach was recently investigated on an Australian cohort of 85 mild to moderate ulcerative colitis patients, in the largest randomized control trial of FMT for IBD to date. Rather than receiving FMT from just one donor, patients in the treatment arm were administered a stool mixture that contained contributions from up to seven different donors with the hope that donor-dependent effects could be homogenized. In addition to this, a far more intensive dosing program was adopted with an initial FMT delivered by colonoscopy that was followed by fecal enemas, five times a week for 8 weeks. Despite the multi-donor and intensive dosing approach, Paramsothy et al. achieved post-FMT remission rates (FMT, 27% vs. placebo, 8%, p = 0.02) that were similar to those reported previously. Notably, however, both clinical and endoscopic remission were required for primary outcome achievement in this study, whereas previous studies have mostly focused on either endoscopic or clinical remission rates alone. The pooled stool mixture was demonstrated to have higher microbial diversity than individual stool alone based on OTU count and phylogenetic diversity measures. Subsequent analysis of the different stool batches discovered that one donor appeared to exhibit a super-donor effect. Specifically, patients that received FMT batches that contained stool from this one donor exhibited a higher remission rate than those whose FMT batches did not include the super-donor (37 vs. 18%, respectively).

FMT for Other Disorders: Is There Also a Super-Donor Effect?

Evidence of FMT super-donors in other disorders outside of IBD is currently lacking. Case series and reports limit the capacity to identify super-donor effects because of limited sample sizes. However, despite the lack of large cohort studies, several studies have hinted at the possibility of a donor-dependent effect on FMT outcome. For example, in a short-term FMT pilot trial on 18 middle-aged men with metabolic syndrome, FMTs from lean donors (allogenic FMT) were found to correspond with a 75% increase in insulin sensitivity and a greater diversity of intestinal bacteria in the recipient compared to autologous FMTs (recipient-derived). It was later noted that the patients who experienced a more robust improvement of insulin sensitivity post-FMT had all been in receipt of the same donor. In a subsequent study on 38 Caucasian men with metabolic syndrome, lean donor FMT also resulted in a significant improvement in peripheral insulin sensitivity at 6 weeks. However, this effect was lost by the 18 week follow up. For the allogenic FMT, 11 lean donors were used, seven of which were used for more than one recipient. Whilst donor-dependent effects were not reported, the authors noted that the “multiple fecal donors might explain the transient and variable effects seen in the allogenic group.” As FMT research in this field progresses from small-scale case series to larger-scale randomized placebo controlled clinical trials, it remains to be seen whether the super-donor phenomenon generalizes to other conditions outside of IBD.

Abandoning the “One Stool Fits All” Approach

Microbial dysbiosis is a blanket term for an unhealthy or imbalanced gut community. As such, the population structure that is considered to represent microbial dysbiosis is variable between different disorders. Moreover, the microbiome deficit of one individual may not necessarily mirror that of another individual and therefore it is not surprising that patients respond differently to FMT. As more FMT-related clinical and microbial data are generated, it is becoming clear that “one stool does not fit all” in the context of treating chronic diseases with microbial dysbiosis. Equally so, the selection of donors based solely on clinical screening guidelines provides no guarantee of FMT success. It appears a patient's response to FMT predominantly depends on the capability of the donor's microbiota to restore the specific metabolic disturbances associated with their particular disease phenotype. If this is true, a donor-recipient matching approach, where a patient is screened to identify the functional perturbations specific to their microbiome, may be the best way forward. The patient could then be matched to a specific FMT donor known to be enriched in taxa associated with the metabolic pathway that needs to be restored. Immune tolerance screening would also be beneficial for reducing the impact of donor-recipient incompatibilities stemming from underlying differences in innate immune responses.

Framework for rational donor selection in fecal microbiota transplant clinical trials

Early clinical successes are driving enthusiasm for fecal microbiota transplantation (FMT), the transfer of healthy gut bacteria through whole stool, as emerging research is linking the microbiome to many different diseases. However, preliminary trials have yielded mixed results and suggest that heterogeneity in donor stool may play a role in patient response. Thus, clinical trials may fail because an ineffective donor was chosen rather than because FMT is not appropriate for the indication. Here, we describe a conceptual framework to guide rational donor selection to increase the likelihood that FMT clinical trials will succeed. We argue that the mechanism by which the microbiome is hypothesized to be associated with a given indication should inform how healthy donors are selected for FMT trials, categorizing these mechanisms into four disease models and presenting associated donor selection strategies. We next walk through examples based on previously published FMT trials and ongoing investigations to illustrate how donor selection might occur in practice. Finally, we show that typical FMT trials are not powered to discover individual taxa mediating patient responses, suggesting that clinicians should develop targeted hypotheses for retrospective analyses and design their clinical trials accordingly. Moving forward, developing and applying novel clinical trial design methodologies like rational donor selection will be necessary to ensure that FMT successfully translates into clinical impact.

Association Between the Clinical Efficacy of Fecal Microbiota Transplantation in Recipients and the Choice of Donor

Objective: To examine the association between the clinical efficacy of fecal microbiota transplantation (FMT) in recipients and the choice of donor, and to observe the characteristics of intestinal flora and metabolites among different donors. 

Methods: A retrospective case-control study was conducted. Donor whose feces was administrated for more than 30 recipients was enrolled. Data of 20 FMT donors and corresponding recipients at Intestinal Microecology Diagnosis and Treatment Center of the Tenth People's Hospital from October 2018 to December 2019 were collected retrospectively.

During follow-up, the efficacy of each recipient 8-week after FMT treatment was recorded and analyzed. Based on the efficacy of each donor, the donors were divided into three groups.Association of the efficacy of each donor group with the morbidity of complications, and association of efficacy of recipients with donors were analyzed. The evaluation indicators of FMT efficacy included objective clinical effectiveness and/or subjective effectiveness. Objective effectiveness indicated clinical cure plus clinical improvement, and subjective effectiveness indicated marked effectiveness plus medium effectiveness through questionnaire during follow-up. 

Results: A total of 1387 recipients were treated by 20 donors, including 749 cases of chronic constipation, 141 cases of chronic diarrhea, 107 cases of inflammatory bowel disease (IBD), 121 cases of irritable bowel syndrome (IBS), 83 cases of autism, and 186 cases of other diseases, such as radiation bowel injury, intestinal pseudo-obstruction, paralytic intestinal obstruction, functional bloating and allergic diseases. There were 829 cases, 403 cases, and 155 cases in high efficacy group, moderate efficacy group and low efficacy group respectively. Baseline data among 3 groups were not significantly different (all P> 0.05).

In comparison of bacterial abundance (operational taxonomic unit, OTU) among different effective donor groups, the high efficacy group was the highest (330.68±57.28), the moderate efficacy group was the second (237.79±41.89), and the low efficacy group was the lowest (160.60±49.61), whose difference was statistically significant. 
In comparison of butyric acid content among three groups, the high efficacy group had the highest [(59.20±9.00) μmol/g], followed by middle efficacy group [(46.92±9.48) μmol/g], and the low efficacy group had the lowest [(37.23±5.03) μmol/g], whose difference was statistically significant (F=10.383, P=0.001). The differences of acetic acid and propionic acid among three groups were not statistically significant (all P>0.05). A total of 418 cases developed complications (30.1%). Morbidity of complication in low efficacy group, moderate efficacy group and high efficacy group was 40.6% (63/155), 30.0% (121/403) and 28.2% (243/829) respectively, and the difference was statistically significant (χ(2)=9.568, P=0.008). The incidence of diarrhea in low efficacy group, moderate efficacy group and high efficacy group was 7.1% (11/155), 4.0% (16/403) and 2.8% (23/829) respectively, and the difference was statistically significant (χ(2)=7.239, P=0.027). Comparing the incidences of other types of complications, no statistically significant differences were found (all P>0.05). Follow up began 8 weeks after the FMT treatment. The total follow-up rate was 83.6% (1160/1387). The overall effective rate 58.3% (676/1160). Effective rates of various diseases were as follows: chronic constipation 54.3% (328/604), chronic diarrhea 88.5% (115/130), IBD 56.1% (55/98), IBS 55.1% (59/107), autism 61.6% (45/73), and other diseases 50.0% (74/148). Comparing the effective rate of three groups of donors for different diseases, there was no statistically significant difference in chronic diarrhea (P>0.05); there was a positive correlation trend in IBD, IBS and autism, but the differences were not statistically significant (all P>0.05). For chronic constipation and other diseases, high efficacy group had the highest effective rate [65.0% (243/374) and 63.2% (55/87)], followed by moderate efficacy group [49.4% (86/174) and 38.1% (16/42)], and low efficacy group had the lowest [16.1% (9/56) and 15.8% (3/19)], whose differences were significant (all P

Conclusions: Different donors have different efficacy in different diseases. Chronic constipation, radiation bowel injury, etc. need to choose donors with high efficacy. IBD, IBS and autism may also be related to the effectiveness of donors, while chronic diarrhea is not associated to the donor. The efficiency of the donor is negatively correlated to the morbidity of complications. The abundance and diversity of intestinal flora and the content of butyric acid may affect the efficacy of the donor.

Conclusion

FMT in practice today does look rather primitive, but seems to be beneficial more than half of the time, even in autism in the Chinese study.

As expected, different donors have different efficacy in different diseases.  As FMT becomes more popular you would expect that more super-donors will be stumbled upon and then clinicians will have a better chance to match the donor to the recipient.

For certain GI conditions that do not respond well to current drug therapy, FMT does look a good option to investigate.  The level of success is likely to vary depending on the availability and selection of the donor.

It does seem that orally ingested bacteria in the form of probiotics often do not colonize the gut as hoped for, and just past straight through, with only a limited and transient effect.  The fact that FMT can have a very long-lasting effect is remarkable and likely due to the fact that these bacteria are direct from another human.

Modifying the microbiome is only now emerging as a treatment idea and it will take many decades to fully develop it.

Ingesting a mix of another human’s bacteria is not without risk.  

Massachusetts General Hospital oversaw trial that led to the first death from a fecal transplant, a new paper shows

This spring, a 73-year-old man with a rare blood condition became the first person to die from drug-resistant bacteria found in a fecal transplant. New details about that unprecedented incident emerged on Wednesday.

The man was a participant in a clinical trial run at Massachusetts General Hospital and received fecal transplant capsules made in November with fecal material from one stool donor, according to a paper published Wednesday in the New England Journal of Medicine. Tests after the man’s death revealed that material contained a rare type of E. coli bacteria.

FMT seems to be becoming fashionable, with all kinds of people offering it.  The American Journal of Gastroenterology even published a study on Do-it-Yourself FMT. "Almost all indicated that they would perform DIY FMT again, though many would have preferred to have FMT in a clinical setting."  I would vote for the clinical setting and a carefully selected/screened donor.