By Rachel Sachs
Yesterday, the FDA announced a new draft guidance regarding its exercise of its enforcement discretion around the investigational new drug (IND) requirements as they apply to fecal microbiota transplantation, or FMT. For almost three years now, the FDA has exercised its enforcement discretion for FMT under a rather permissive set of guidelines, enabling patients to access FMT while companies shepherd a set of products through the clinical trial process. I recently co-authored an article about the FDA’s regulation of FMT, and I’m concerned about this new guidance, in terms of safety, access, and regulatory clarity. This is one of the wonkiest posts I’ve written in some time (and that’s saying something), so I’ll endeavor to be as clear as possible.
What is FMT and why is it important? To make a long story short, it’s a poop transplant. Filtered stool from a healthy donor is transplanted into the gastrointestinal tract of a sick patient. Although scientists are continuing to explore the use of FMT for a range of indications, we already know that FMT is a startlingly effective treatment for recurrent C. difficile infection. C. difficile infections have become among the most common hospital-acquired infections in the United States, with more than 450,000 total incident infections annually. Unfortunately, many of these infections are resistant to antibiotics: with those 450,000 infections came 80,000 recurrences and 29,000 deaths. But FMT may provide a way forward. A recent randomized trial of patients with recurrent C. difficile infections was stopped early, when 94% of patients in the FMT group were cured, as compared to roughly 30% of those getting only antibiotics.
How is the FDA involved? In May 2013, the FDA announced that fecal microbiota would be regulated as a drug. All uses of FMT would therefore need to be part of an IND application, and patients who wanted to be treated with FMT for recurrent C. difficile would need to participate in a clinical trial to do so. Physicians and scientists responded with concern, arguing that available evidence supporting FMT’s effectiveness as a therapy for recurrent C. difficile infection was too compelling for regulators to restrict its availability to clinical trials. As such, in July 2013, the FDA announced that it would exercise enforcement discretion when FMT was used to treat patients “with C. difficile infection not responding to standard therapies,” so long as the treating physician obtained informed consent.
The FDA’s decision not to enforce the IND requirement for recurrent C. difficile infection has significantly expanded access to the therapy, in no small part by creating a window in which companies like OpenBiome, a nonprofit stool bank spun off from MIT that screens donors, processes samples, and ships them to hospitals around the country, could operate. [Disclosure: my co-author on the article about the FDA’s regulation of FMT was Carolyn Edelstein, the Director of Global Partnerships at OpenBiome. However, I have no financial conflicts of interest in the company.]
What would the new draft guidance do? The new draft guidance, if finalized, would impose two additional conditions on the situations under which the FDA will exercise its enforcement discretion regarding the use of FMT. First, “the FMT product is not obtained from a stool bank,” and second, “the stool donor and stool are qualified by screening and testing performed under the direction of the licensed health care provider for the purpose of providing the FMT product for the treatment of the patient.” At present, stool banks like OpenBiome perform screening and testing procedures on the FMT product they then distribute to health care providers. This new draft guidance would prevent them from operating as they do now. Instead, the draft guidance contemplates that hospitals are permitted to develop independently the capacity to screen and test FMT donors, to serve the patients under the care of the physicians working at the hospital. In other words, the current centralized system of stool processing and distribution would be rapidly decentralized to hospitals around the country.
What are three reasons why we should care? First, if the FDA chooses to finalize this guidance before a company receives FDA approval for their FMT product (maybe a big if, as I explore in the next section), we might have concerns about access to the treatment. The FDA contemplates that hospitals may develop their own facilities for the collection, preparation, and storage of FMT for their patients’ use. However, some number of hospitals (perhaps a large fraction) will decline to develop the infrastructure and personnel needed to maintain such a bank, depriving their patients of access to FMT through their institution. Put another way, stool banks can achieve economies of scale in a way that many hospitals cannot. These are ultimately empirical questions, and I can only hypothesize at their magnitude, but they are unlikely to be trivial.
Second and relatedly, if the FDA chooses to finalize this guidance soon, we might have concerns about safety. It’s important to remember that FMT has an additional feature which may make any observations in terms of access to care particularly problematic in the context of FMT relative to other procedures. Specifically, with FMT, there is real do-it-yourself potential. The reality is that patients without access through licensed health care providers are likely instead to watch a YouTube video and take matters into their own hands, not that they would be unable to undergo FMT at all. For safety reasons, we would prefer to have these patients treated in licensed facilities, with regulated products.
Third, although the FDA has declared its intention to regulate FMT as a drug, this draft guidance is in many ways evocative of a different regulatory paradigm: tissue. In our article, Carolyn and I argued that the FDA’s system for regulating human tissue was a better fit for FMT than was the FDA’s system for regulating drugs, although both had benefits and drawbacks. No one company has a monopoly on producing and selling tissues like blood – the FDA nonexclusively licenses facilities to store, package, and distribute blood. The draft guidance would decentralize the oversight of FMT from a small number of stool banks to a much larger number of hospital facilities, effectively moving from a system resembling the drug paradigm (one or a small number of highly regulated producers) to one resembling the tissue paradigm (a wide array of facilities regulated for storage and distribution).
What’s really going on here? The FDA is certainly aware of these concerns about safety and access. Although I don’t have special insight into the FDA’s thought process, I suspect their thinking is as follows. Two companies (Seres and Rebiotix) are quickly approaching FDA approval for their products to treat recurrent C. difficile infections. Once one of those companies gets approval, the FDA will be reticent to permit entities like OpenBiome to remain in operation, especially outside of the traditional drug regulatory approval pathway. As such, the FDA may envision this guidance taking effect at some time in the near future, once the FDA approves another company’s product.
Yet why not simply revoke the existing enforcement discretion guidance at that time, preventing OpenBiome or any other unlicensed entity from operating? Why create an entirely new decentralized structure getting rid of OpenBiome but creating capacity for hospitals to engage in a more artisanal form of stool banking? It’s here that I think the FDA is concerned about costs and access (although they certainly couldn’t say so). OpenBiome currently charges $385 or $535 per FMT treatment, depending on the route of delivery. The products coming to market are likely to cost several thousand dollars, and understandably so, given the expenditures involved in the clinical trial process. To the extent that providers and payers are concerned about costs or patients may not be able to access more highly-priced therapies, they might prefer the safety valve of allowing hospitals to develop such capacity. This is akin to the compounding pharmacy safety valve the FDA has permitted in the context of Makena, another drug facing a similar regulatory history.
So maybe the FDA’s draft guidance is not as odd as it may seem at first glance. The FDA is aware that finalizing the guidance immediately would create significant safety and access concerns, and it may therefore have no plans to do so. But even on that assumption, its intention to permit hospitals to dispense FMT product is still puzzling. If (as the FDA has said) FMT is a drug, I can come up with no example in which the FDA has permitted small-scale manufacturing outside the context of the drug regulatory pathway. (Compounding pharmacies are different, in my view, for both historical and regulatory reasons.) If FMT is a tissue, the FDA’s regulatory paradigms do not permit it to grant seven years of market exclusivity for a company to produce and sell poop. With this draft guidance, the FDA is attempting to straddle the line between the two paradigms, and whether it achieves salutary results remains to be seen.
These are just a few of the issues raised by the draft guidance. Others (about its effects on innovation, the role of standard-setting bodies, the elimination of the “known donor” provision, etc.) could be the subject of entirely separate posts. I will be very interested to see the public comments submitted to the FDA on this issue (submit yours here!), and I’ll keep the two of you who made it to the end of this post updated as we move forward.