NEW YORK (GenomeWeb) – The US Food and Drug Administration is considering changing how it regulates fecal microbiota transplants for patients with recurrent Clostridium difficile infections.
The agency released a draft guidance last month that would exclude stool banks from the enforcement discretion it is currently exercising over fecal microbiota transplants (FMTs) for treating C. difficile. If enacted, the guidance would require stool banks like OpenBiome and AdvancingBio to have an investigational new drug (IND) application in place before they can distribute samples.
"The 2016 draft guidance proposes a revised policy with regard to patient access to FMT," FDA told GenomeWeb in an email.
While the stool banks said that they can contend with such paperwork, they are unsure how these requirements would affect the physicians to whom they provide samples and their patients. The draft guidance shouldn't, though, affect research into fecal microbiota transplants and might even encourage it, if enacted, researchers and clinicians in the field said.
FMT involves delivering stool from a healthy person to someone who is sick — usually through a colonoscopy, enema, nasal-duodenal tube, or sometimes a pill — in a bid to alter their gut microbiome and make the microbial community there healthy again.
It's typically an effective treatment. When a University of Amsterdam-led team of researchers sought to examine the effect of FMT on recurrent C. difficile infections, they had such striking results that they stopped the study early. Initially, the team randomly assigned 43 patients to receive donor feces, vancomycin, or vancomycin and bowel lavage treatment, as they reported in the New England Journal of Medicine.
Of the 16 patients who underwent FMT, 13 were cured after their first round of treatment and two more were cured after an additional round. By contrast, four of the 13 patients in the vancomycin-only group were cured and three of the 13 in the vancomycin-and-lavage group were cured.
The approach, though, isn't without its risks. "The product intrinsically has risks; we're talking about poop here," Alexander Khoruts, an associate professor of medicine at the University of Minnesota, told GenomeWeb.
There have been sporadic reports of adverse effects and even death following FMT. Clinicians in the UK reported last year in Clinical Infectious Diseases on the first death due to FMT complications at their clinic out of 14 total patients treated. That patient accidentally aspirated the liquid, as it was being administered to the duodenum through an enteroscope, and later died of pneumonia.
Because of its risks, FDA has said it will regulate FMT. In a guidance document issued in July 2013, FDA noted that while it had oversight over fecal microbiota transplants to treat disease, it was creating a carve-out for FMT as a treatment for recurrent C. difficile infections. Clinicians using FMT to treat such infections wouldn't have to have an IND.
A follow-up draft guidance issued in 2014 included a provision that would have required the person donating the fecal matter to be known to either the patient or the doctor performing the treatment.
"When FDA wrote these documents, our intention was to accommodate the immediate needs of very sick patients with C. difficile [who were] not responsive to standard therapies while we further consider the matter," the agency told GenomeWeb in an emailed statement. "There are often few or no other treatment options for these patients, and the illness can be life‐threatening,"
But, FDA said, the way this 2014 provision was written opened it up to "difficulties in interpretation."
The draft guidance issued this year offers a tweak: it says that the enforcement discretion is not applicable if the material for the FMT is obtained from a stool bank. But that carve-out still holds for doctors performing the procedure who have obtained informed consent, as long as the stool donor and stool are both screened and tested for the procedure.
This, FDA said, "more accurately reflects our intent to mitigate risk, based on the number of patients exposed to a particular donor or manufacturing practice rather than the risk inherent from any one donor."
It noted, though, that stool banks would be able to request a waiver of certain IND regulations applicable to investigators and sub-investigators.
The comment period on the draft guidance is open through the end of May.
These changes would likely affect stool banks like those operated by OpenBiome or AdvancingBio, which provide screened and tested materials for FMTs to physicians for their patients.
"If it were enacted just like it looks now, it definitely would impact us," Carolyn Edelstein, the director of policy and global partnerships at OpenBiome, told GenomeWeb.
Any doctor wanting to use material from a stool bank would have to do so under an IND, she added. That would mean that those clinicians would have to be named as sub-investigators.
A section of the draft guidance, though, also suggests clinicians might be spared some of the regulatory rigmarole. It says that sponsors could request a waiver of certain regulations, but it's not yet clear which regulations could be waived.
"The guidance does talk about the need for an IND, but it also indicates the potential for variances," said Nicole Anderson, the executive vice president of AdvancingBio.
Both OpenBiome and AdvancingBio largely welcome regulation, though they noted that it would have to be weighed against patients' access to treatment.
"Regulation is a positive thing. It is going to ensure a safe product," Anderson said. "Availability of products is a big a concern, and so balancing both of those things is extremely important."
OpenBiome's Edelstein noted that her organization has filed a biologics master file with FDA that discusses their donor and stool screening process, a process on which they have received agency feedback.
OpenBiome screens donors and the samples they provide for a number of conditions and diseases. After collecting samples, they wait a period of time and re-test donors to determine whether they've gotten sick in the interim. They also ask the physicians they work with to submit de-identified patient outcome data to monitor safety and efficacy.
"We actually would say that by having donors used repeatedly on a lot of different patients, that you're actually re-disking them," she said.
AdvancingBio, which is related to blood bank BloodSource, also screens its donors and samples, borrowing from its blood banking experience.
Getting an IND can be a complex, time-consuming process. Khoruts' center at Minnesota, which has evaluated some 2,000 patients and treated about 400 patients via FMT, has an IND. He said that it took him and his colleagues about a year to prepare the file.
He argued that if someone was going to open a bank that deals with a material like stool, which is intrinsically risky, that person has to be willing and able to deal with protocols aimed at minimizing that risk. "If an organization can't put it together, they shouldn't be in this sort of business," Khoruts said.
"We could pull together an IND if we need to," OpenBiome's Edelstein added. She noted that they actually have one for a trial they are conducting.
The sticking point, she said, is whether or not it would become difficult for patients to get treated, as their treating physicians, to use material from a stool bank, would have to be sub-investigators under the IND.
"The burden is sub-investigator sites, the people who actually administer [the treatment]," Khoruts noted.
Still Khoruts argued that the guidance, if enacted, won't have too much of an effect on clinical care. Some smaller or low-volume practices might be put off by the paperwork, he noted, and choose not to continue the procedure.
But to Khoruts, that's not necessarily bad. "There is value at this time to really concentrate these treatments in centers that do a lot of these patients," he said. He noted that he's seen some 2,000 recurrent C. difficile patients and that he still learns something new from each new one. "I just don't think that it is optimal at this stage where we are still on a fairly steep learning curve for just anybody to do this kind of stuff," he added.
He acknowledged, though, that it might be harder for patients to make it to a tertiary care center.
Both Khoruts and Florian Fricke, an adjunct assistant professor at the University of Maryland and professor at the University of Hohenheim in Germany, noted that the draft guidance likely wouldn't have a large impact on research.
Fricke said that the work he's done in the US wouldn't be affected as it was done as part of a small clinical study, and the clinician he collaborated with knew both the donors of the stool samples and the patients.
"In the clinician there is the link the FDA is asking for from donor to patient, so we are not affected by that," Fricke said.
Similarly, Khoruts noted that at his center, he provides this link between donors and recipients as he runs both the donor and treatment programs.
Regulation, Khoruts added, would also mean that more data would be collected on patients who receive FMT treatments and their outcomes, including whether certain patient populations have different responses to treatment. He lamented that much of the data on patients treated using stool bank samples has gone unpublished.
"I know that there are sub-groups within this population that respond differently and have different challenges, so that is essentially lost right now," he said. "What are the different nuances? How do you treat a paraplegic?"
Patients who develop recurrent C. difficile are typically older and have other health issues.
His group recently published a study based on the first 272 patients his group treated between 2008 and 2015. As they reported in Clinical Gastroenterology and Hepatology online in February, they found that FMT was a bit less effective in treating recurrent C. difficile patients who also had inflammatory bowel disease — nearly 75 percent effective versus 92 percent effective. In addition, they found that more than 25 percent of patients who also had IBD had a disease flare after FMT treatment.
Having this data not only means that tweaks could be made to the product's formulation, but Khoruts said that it also means he can paint a better picture of what's to come for patients.
"Now that I have done so [many FMT procedures], when I consent patients, if somebody has underlying inflammatory bowel disease, I have a different risk-benefit presentation for them than for somebody who doesn't," he said. "Different efficacy, different risks.
An open question for many is how FDA will implement the draft guidance.
Both OpenBiome and AdvancingBio said that some details on how FDA would apply the draft guidance would be welcome. The groups are trying to determine, for instance, how clinicians who use their services would be affected as well as how this fits into FDA's longer-term view of FMT and, in particular, what safety testing and procedures the agency would like to have included in the testing panel.
"From a public health standpoint, [regulation is] the right move, and we just hope to get some clarity around what this particular proposal would mean both for stool banks and for physicians and, by extension, the patients they are trying to serve," OpenBiome's Edelstein said.
In addition, Khoruts noted that some lead-in time would be helpful. When the 2013 guidance on FMT was released, it went into effect immediately and that, he said, led to some consternation.
"They came out and said you have to do all these things now, starting tomorrow," he said. "They have to be more measured and orderly in implementation. This kind of paperwork just takes a long time."
OpenBiome's leaders have also suggested a different approach for regulating stool banks. Rather than treating stool as a drug, OpenBiome Co-founder Mark Smith and board member Eric Alm have argued in a Nature comment that it could instead be treated like tissue or even blood, which has its own regulatory category. This type of regulation, they said, could allow patients access to safe products as well as enable research.
While Edelstein said OpenBiome would prefer that type of regulation, as it would allow for flexibility in how stool would be screened and prepared, she noted that it's still so early in the field that they are still figuring out what the best course would be.
Hohenheim's Fricke also noted that the procedure itself rather than the stool sample might be the riskiest part of the procedure. The recent report of an FMT-linked fatality due to aspiration of fecal matter was actually due to how it was administered, he noted.
Still, he said regulation is a necessary next step and is a sign of a maturing field. "We read all these papers about the microbiota and how important it is. But this is the only clinical example where we actually treat based on microbiota manipulation," Fricke said. "So, that's obviously new and requires new regulations, new research, everything. We're in the middle of that process right now."