Hardly a week goes by without news of the most recent reported link between an element of the human microbiome and risk of disease. Most of these studies begin as simple correlations of microbes, microbial signatures, or microbial diversity with health outcomes of interest. Increasingly, however, formal clinical trials aimed at modulating and manipulating the human microbiome for better health are being proposed and carried out1,2,3,4,5,6. Prior to the initiation and implementation of therapeutic manipulation of the microbiome, a thoughtful consideration of the need for and role of regulatory challenges ought to be undertaken. These considerations apply to both transfer of naturally occurring microbial communities and cultured microbial consortia.

Viewing therapeutic manipulations of the microbiome through a regulatory lens

In the United States, companies that develop products and therapeutics for the manipulation of the microbiome follow one of two pathways: (i) the low-cost, low-rigor pathway of over-the-counter probiotics with ‘wellness’ and ‘structure and function’ claims or (ii) the high-cost, high-rigor pathway of US Food and Drug Administration (FDA)-approved therapeutics for specific medical treatment indications4,5,6. These therapies include live or quiescent microbes that span the spectrum from single cultured probiotic organisms, to consortia of microbes (isolated from humans, or cultured in vitro and combined), to ‘complete’ microbial ecologies collected from human volunteers (for example, fecal microbiota transplants (FMT)).

The Center for Biologics Evaluation and Research (CBER) is the section of the FDA that oversees biological products, including blood and blood products, vaccines, allergenics, gene therapies, xenotransplantation, and human cells and tissue products (HCT/Ps), that are intended for implantation, transplantation, infusion, or transfer into a human recipient. Human-derived microbial products and other live organisms administered with therapeutic or preventative intent are also overseen by this section of the agency. For example, using FMT to treat recurrent Clostridium difficile colitis, inflammatory bowel disease, or autism would clearly involve a claim of therapeutic intent and would be regulated by CBER. The FDA has fast-tracked approval of microbiome therapeutics for C. difficile infection and has issued guidance to physicians indicating that FMT for C. difficile may be undertaken for treatment of individual patients without interaction with the agency if criteria that are described elsewhere are met7,8. Separate procedures exist for approval of nutritional supplements or food products for which treatment of a specific medical condition by the supplement is not claimed. Thus, probiotics may be regulated as dietary supplements, foods, or drugs depending on the intended use of the product.

The Center for Food Safety and Nutrition is another office within the FDA that is responsible for clear and truthful product labeling and the safety of these products8. A point of significant interest and discussion at a recent meeting at the FDA (Science and Regulation of Live Microbiome-Based Products Used to Prevent, Treat, or Cure Diseases in Humans, September 2018) was what studies involving live bacterial products (LBPs) could be undertaken without submission of an investigational new drug (IND) application, and what constitutes a food-like entity or a health and wellness claim. For example, it would be possible to study a live bacterial product mixture for ‘improvement of GI health’ without filing an IND application and without formal supervision of the FDA, but use of the same product to ameliorate the symptoms of Crohn’s disease would require submission as an IND and FDA review. There is a spectrum of intent perhaps not obvious to all. For example, one could evaluate a topical microbe spray to alternatively promote ‘glowing skin’, prevent mild acne, or treat severe acne. Under US FDA regulations, each of these studies and their claims (e.g., promotion, prevention, or treatement) would be considered quite differently.

Risks and limitations

There are quantifiable and readily appreciated risks to donor-derived microbial therapeutics. Known risks of transferring donor-derived microbes include those that can be screened for, such as incidental transfer of viruses and fungi, which can be detected in the blood or stool (this includes HIV, hepatitis viruses, cytomegalovirus, and more1,2,3,4,7,8,9). However, there is also the risk of transferring pathogenic viruses and bacteria that pose a substantial risk to human health, but that do not reside in the blood (such as HPV in vaginal fluid or picornavirus in nasal washes). To what extent do we screen for these rare pathogens that are often difficult to detect, and at what cost? The detection of platelet transfusion products contaminated with bacteria in recent years is a cautionary tale worth noting9.

Given the risk of human-to-human pathogen transfer, there is currently extensive effort to define precise bacterial strain concoctions (such as microbes cultivated in the laboratory) that confer health benefits with the goal of eventually generating ‘clean’ versions of the microbial cocktails for clinical use. However, live microbial products are complicated by the challenges of strain specificity, purity, potency, mutability, viability (or lack thereof), the potential for microbial genetic recombination, and dose verification and quantification. Most orally administered single-strain probiotics (LBPs) do not durably alter or shift intestinal microbiomes10 and persist only with repeated administration, a characteristic that is perhaps beneficial from both a safety and commercial sales perspective. Moreover, it is not presently clear what resident microbial communities would be necessary to allow administered microbial consortia, or complete microbiomes, to durably shift complex microbiomes like that of the gut, or simpler communities like that of the vagina, to ‘better states’.

Future risks that are yet unknown

There have been reports detailing the role of the microbiome in modulating risk of disease that may take years or decades to develop. These include everything from metabolic disorders, such as obesity and insulin resistance, to cancer and chemotherapeutic sensitivity, to infertility, to mood disorders and depression11,12. By transferring a relatively young person’s microbiome, there is an inherent risk of future disease that has already been ‘programmed’ into the microbiome but is presently asymptomatic. For example, what if the 19 year old donating her stool develops major depressive disorder at age 27, obesity at age 30, or inherently chemo-resistant pancreatic cancer at age 40? Similarly, what if the 22-year-old donor of a vaginal microbiome is destined to deliver an infant preterm or suffer from infertility and recurrent pregnancy loss? Each of these conditions has been associated with modulations in the microbiome, but predictive associations are lacking10,11,12. In light of loosely held microbiome–disease associations, even in the absence of causality, it seems reasonable to mandate formal review of the potential future risks by the FDA. While suggested standard screening tests for donors have been cataloged7, there is no clear consensus on how to assess for such future risks. The possibility that drug metabolism, drug responsiveness, weight, mood, social functioning, and other health outcomes may be linked to the microbiome give one pause when considering the utility of FMT, especially in children.

The role of institutional review boards and ethics committees

We suggest that institutional review boards that oversee the approval of research on human subjects complement, but not replace, a review by regulatory entities like the FDA. Thoughtful and imaginative scientists need to ponder the outcomes and data points worth studying after microbiome manipulation and consider qualitative approaches (such as patient focus group research) to formally assess what both their clinical colleagues and potential subjects might deem acceptable and unacceptable risk.

In our experience, it is possible for institutional review boards to either go ‘overboard’ and be so precautionary that reasonable risk is not allowed, or fail to fully consider substantial future risk. Tracking possible transmission of infections seems like a relatively discreet, concrete task when compared to tracking longer term health outcomes related to durable microbiome manipulation. Nevertheless, points that need to be considered include: what duration of follow-up is necessary? What degree of microbiome change is rationally attributable to a live microbial therapeutic? How might microbial changes be altered by future drugs, diseases, or aging? At the completion of a study in which participating subjects are given live microbial therapeutics, should these subjects receive antibiotics for eradication of the study microbe (regardless of study findings)? If so, what collateral damage might such antimicrobial therapy subsequently unleash? Is there a possibility of incidentally identifying human subjects when the microbial therapeutic used is rare in contemporaneous populations?

Recommendations

It is our opinion that it would be wise to exercise significant caution in the transfer of microbiomes and microbial consortia, especially in younger and reproductive-aged individuals. In addition, standards for systematically studying these risks should be established. Potential risks of incidental transfer of common, chronic conditions, such as obesity and mood disorders, ought to be discussed with human subjects in frank terms while acknowledging risks that are currently poorly quantifiable. Similarly, the possibility of transfer of other viral and fungal pathogens with known immediate and long-term risk should be delineated. A systematic approach and standard language related to possible unknown future risks could be of value. At present, the field of metagenomic medicine remains in its infancy, and what we know with confidence is dwarfed by uncertainties. It is precisely these uncertainties that demand a high degree of regulatory oversight, scrutiny, and honest conversation with patients and the public at-large.