Banking Previous Poos: Could a Transplant of Feces from Your Past Heal You?

The future of personalized medicine could lie in a person’s own past. Given the influence gut microbes have over a wide variety of health conditions, researchers from Brigham and Women’s Hospital and Harvard Medical School have an unusual proposal: people could bank their poop when they’re young and in good health so that it can be transplanted back into their system when their health is waning.

The idea, proposed today (June 30) as a review in Trends in Molecular Medicine, comes from existing evidence on autologous fecal microbiota transplantation (FMT)—a fecal transplant of a person’s own stool. The researchers suggest that such FMTs could rejuvenate the microbiome later in life, and in doing so, may help prevent conditions ranging from obesity to asthma, though they’re careful to avoid saying that there’s definitive proof such a procedure would help.

The Scientist spoke with coauthors Yang-Yu Liu and Scott Weiss to learn more about the proposal and how the unusual hypothesis might be validated down the road.

The Scientist: I don’t think I’ve ever come across this notion of banking stool samples to rejuvenate the microbiome before. Before we get into the details, can you talk about where the idea for the study originally came from?

Yang-Yu Liu: Back in July of last year, my colleague and long-term collaborator, professor Scott Weiss, forwarded a very interesting New York Times article entitled, You’re Missing Microbes, But Is ‘Rewilding’ the Way to Get Them Back? That’s why we decided to write this paper: because we think rewilding the microbiome using samples collected from nonindustrial societies like the Hadza people in Africa sounds ridiculous, and we don’t believe it will work. It might even be harmful to the people in the industrialized societies, because there’s a dramatic mismatch between the preindustrial microbiome and the industrialized lifestyle and diet patterns.

We thought about this, and we came up with an alternative solution. We call it rejuvenating the human microbiome. Basically, we want to collect samples when the host is in a healthier, younger age, and then use the samples for the treatment in the future.

TS: Can you explain why transplanting a preindustrial microbiome might be dangerous? 

Scott Weiss: Well, I think that there’s a mismatch between the diet and lifestyle of these preindustrial populations and the current diet and lifestyle, and that’s a major concern. The other point that we discussed was that there are ethical concerns here, too. It’s not necessarily ethical to be taking the microbiome of some Indigenous tribe in Africa and trying to commercialize that. . . . So for both of those reasons, I think it’s a bad idea.

And as Yang said, that was sort of the motivation that got us started on thinking about this idea. And then we laid out in the paper a whole series of questions and where we sort of thought the data were that supports this.

I think it’s fair to say that a lot more research needs to be done here. It’s not like, “Oh, let’s go do this, and it’s guaranteed to work and we’re on to the next.” I mean, we need a lot more information before you can fully implement something like this. But still, as a framework for future research, we think it’s valuable.

See “Fecal Microbiota Transplantation Is Poised for a Makeover”

Y-Y L: Our idea can be considered a generalization of [a] personal biome service. We want to consider a wide range of potential applications beyond treating a CD infection.

TS: Whether it’s rewilding or rejuvenating, can you explain the impetus or motivation to bank samples and alter the microbiome to return it to some earlier state?

SW: Conclusive evidence that the young, healthy adult microbiome is going to protect you from the development of disease is not there. I mean, we have this suggestion that, in areas like obesity and aging, the microbiome becomes abnormal in advance of the development of these medical conditions. And that doesn’t just come from human data, but it comes from model organism and animal data [too]. I don’t think that we would make the claim that you can just do autologous FMT and not do anything else in terms of a healthy lifestyle, and that would be protective. I mean, we view this as part and parcel of maintaining a plant-based, healthy diet, maintaining ideal body weight, appropriate exercise, stress reduction, all of these other things. The benefits of a healthy microbiome should be viewed in the context of all these other health-promoting behaviors, not in isolation from them.

I think we view this as a framework for future research. Not just “Oh, you know, we should go do this. And we know that it’ll work.”

TS: That’s actually one of the things I appreciated about the paper: that you list caveats, limitations, and unknowns throughout. Given all of those, how confident are you that this would be a beneficial thing for someone to do?

SW: I think it’s like everything else: You’ve got to do the studies and see. Doing research is always a question of how are you going to portion your time? There are lots of things you could work on. We obviously think that this is something that does merit additional work.

We’re clearly not the only people that are thinking like this. There’s a company that’s been formed in Singapore to do exactly what we propose in the article. And that came about after we wrote the paper, but before it was published, so there are clearly other people that are thinking this way. And we certainly have had discussions with venture capitalists and other people who are interested in aging and disease prevention, where there’s a lot of interest in this.

I don’t think we’re in a position to say to you, “Oh, you know, we think 60 percent this is going to work.” I don’t think it’s that; I think that there are some hints out there that this is worthwhile. And, you know, we think that what the paper provides is a framework for future research that would provide the data, yea or nay. And that the paper is a way of providing a way forward to try to figure this out.

Y-Y L: It’s not an original research article, but a perspective paper, right? It wants to promote the idea to ask more people to seriously consider the underlying caveats, the fundamental challenges. But overall, we think it’s a very promising direction, even though [much of the evidence comes] from preclinical models, very few from human studies, but we think for certain applications, this idea is very promising.

TS: Are you able to talk about how those meetings with potential investors went? Do you have plans to explore this in a commercial way?

SW: I think our strength is as research scientists. That’s who we are and that’s what we do. As I said, I think we view the article as a framework for a research program and a way of viewing the research on the microbiome.

I would make the analogy to [how] there’s a substantial body of evidence that suggests that caloric restriction and intermittent fasting is potentially helpful for longevity. But one of the things that no one has investigated, that no one has any idea about, is what’s the effect of that sort of dietary pattern on the gut microbiome? So again, here’s an example where there’s a strong link between other health behaviors that are potentially important: how you eat, how much you eat, what type of foods you eat, [all of those are] clearly going to have an impact on the gut microbiome and disease development. And if there’s evidence in animal models that mice and rats are living longer on this intermittent fasting, then studying the gut microbiome in that context could potentially be helpful to figuring out the mechanism of how this is actually working, because no one knows.

See “Fecal Transplants More Successful From ‘Super-Donors’”

One of the observations that Yang made relatively early on, looking at some data from model organisms, is that before these model organisms die, there’s a marked change in the gut microbiome. So again, suggestive evidence that changes in the gut microbiome may be associated with longevity. But we don’t know exactly how.

Even though there are a lot of questions about autologous FMT, we do know from the experience with [CD] that this is very safe. Most of the [CD] patients are very sick, have cancer or some other significant medical illness. And yet, this treatment has virtually no side effects when it comes to treating these patients. So you’re talking about a procedure that, although we need to study it more, seems to be very safe and relatively low-risk. So that also plays into the calculus about how you would do this and whether it might be feasible. I’m not talking about something that’s a CAR T cell therapy for cancer. This is really a pretty benign type of approach.

Is this going to happen tomorrow? Or is this ten years from now? And it’s definitely ten years from now. It’s not tomorrow.

Y-Y L: This idea is quite similar to the idea of cord blood banking. It’s been available for a while. So right now, we just want to preserve stool samples at very low temperature, like liquid nitrogen temperature. So the idea and the framework is quite similar.

TS: In terms of the possible next steps and the kinds of data that would need to be gathered, are there any specific roadblocks preventing researchers from figuring out whether autologous FMT would have health benefits?

Y-Y L: I think working off preclinical models is one thing, but to really conduct clinical trials, this might be very costly because we’re talking about collecting samples, and for the future use, [it would be] several decades.

SW: I think that the timeframe for onset of different diseases provides an opportunity for thinking about how to do this. For instance, a disease like preeclampsia occurs basically over a nine-month pregnancy. So, you could collect stool samples in the first trimester and relate those stool samples in the first trimester to disease in the third trimester.

Asthma [is] another disease and 90 percent of all asthma occurs by age six. . . . [And] we know, for instance, that cesarean section is associated with asthma risk [and] with an abnormal microbiome. . . . You could do a clinical trial of autologous or even maternal FMT for the baby and do it over a relative timeframe that’s not too long. In fact, that’s a great grant idea. We should be writing a grant about that, because that’s one we can do. You’re not going to take 100,000 people, look at their microbiome at age 30, and follow them until they’re 90. There’s no way you’re going to do that study. Because it’s way, way, way too expensive and too difficult.

But the more you know about different diseases and how they start, the more you can tailor the sorts of clinical studies you could do to attack this problem in a meaningful way at a reasonable cost. And the caloric restriction idea, just talking about this on the last couple of calls with reporters, I’m getting really excited about this idea. So I think that we’re going to try to pursue that if we can. I mean, for all we know, there may be microbiome data out there that hasn’t been looked at yet for that. And we could just do some quick data mining and get some ideas. The asthma idea, preeclampsia idea, those are doable right now. I think the key pieces here are knowing that autologous FMT is relatively safe and then leveraging that in the context of how we can learn as much as we can about disease prevention.

TS: Your paper has this figure of a hypothetical application in which a healthy person banks their stool samples, and then after getting sick, they consult with a doctor and have their health restored due to a transplant. In terms of logistics, when and how would one know that it’s time to use up their samples and perform the transplant?

SW: Again, let’s take the example of the cord blood, because I think it’s useful. So you bank your cord blood sample, and all of a sudden, at age six, a child develops acute leukemia and needs a bone marrow transplant. Well, that’s a really severe disease. But having that cord blood means that they can have an autologous bone marrow transplant, which is extremely low risk for [human leukocyte antigen] complications and is much safer and can be done right away to treat the disease.

I think that the timing for conditions like obesity, or aging, or autoimmune disease are much more subtle and we don’t know. I think that we do have some indications with the microbiome and asthma, where we would say that you want to do that autologous stool transplant very early in life, probably before the age of three. It’s going to depend on the disease, and we know less about obesity, aging, autoimmunity, [and] coronary disease than we do about some of these other diseases. But earlier, in general, for all disease and treatments, right? Earlier is better. The longer you wait, the more established the disease is, the more likely your treatment isn’t going to work.

There are clearly other aspects of this, though. We’re not saying—and I think this is very important—we’re not saying that you just do this transplant, and then you walk away. There are dietary interventions, there are other things that you would have to do in addition to the transplant.

TS: Your study says that “the creation of a global ‘microbial Noah’s ark’ is warranted to protect the long-term health of humanity.” Can you elaborate on what you meant by that?

Y-Y L: I want to clarify that the Noah’s ark, the so-called microbiome vault, this project was initiated by some scientists from Rutgers University. And they just want to protect the diversity of the human microbiome in case of some disaster. So our idea is kind of a personal Noah’s ark. We want this for your personal microbiome for your own future use. The reason why we wanted to mention the microbiome vault is because they also are relying on these so-called cryogenic preservation techniques. We’re using liquid nitrogen, for example, to store the samples. So that being said, I think that from the technical perspective, we are facing the same problem. We need to resolve the long-term safe storage problem. For that particular initiative, we[‘ve] already done a very interesting visibility study, we show some promising results that long-term storage is feasible. This gives us more confidence that we can do that. And then we can also store our own set of samples for a very long time. Of course, more research needs to be done. Because we not only want to store the samples for a very long time, but also we want to make sure those bugs are still alive.

Editor’s note: This interview has been edited for brevity.