This transcript has been edited for clarity.
Peter Higgins, MD, PhD: Hello. I’m Dr Peter Higgins, professor of the Division of Gastroenterology, director of the Inflammatory Bowel Disease Program, chair of the Gastroenterology and Hepatology Research Advisory Committee, and director of the Michigan Clinical Trial Support Unit at the Department of Internal Medicine at the University of Michigan, Ann Arbor. Welcome to Medscape’s InDiscussion series on Crohn’s disease.
Today we’ll be discussing early risk markers, intervention, and possibly prevention of inflammatory bowel disease (IBD). First, let me introduce my guest, Dr Jean-Fred Colombel, professor of medicine, director of the Susan and Leonard Feinstein IBD Clinical Center, and the Leona M. and Harry B. Helmsley Charitable Trust Inflammatory Bowel Disease Center at the Department of Gastrointestinal Medicine, Icahn School of Medicine at Mount Sinai in New York. Welcome to InDiscussion. Thank you for joining me, Jean-Fred.
Jean-Frederic Colombel, MD: Thank you, Peter. It’s a pleasure, as always.
Higgins: We always like to start with folks who are in the IBD field and ask, what is it about IBD that first drew you to this field?
Colombel: When I was still in Lille, in France, I was seeing all those patients coming to the hospital. Because, Peter, I was there in the 1970s, when we were starting to see the increasing incidence of IBD in northern France. We were late compared with what you have seen in northern America or Scandinavia. And then suddenly I had my attention attracted to those patients. They were coming to the clinic. I wanted to understand why.
Higgins: For folks less familiar with French geography, Lille is in the north.
Colombel: North, on the Belgian border.
Higgins: There’s a striking gradient across France from the north, where IBD is much more common, to the south or even the islands off the coast of France, where it’s much less common. That was really one of our first epidemiologic clues that environment is important in IBD. Many of my patients who are parents ask me, is there anything I can do to prevent my child from developing IBD? And I generally answer that there is a small genetic component more in Crohn’s disease than ulcerative colitis (UC), but most of the risk appears to be environmental, and our environmental data are quite limited. It appears from epidemiologic studies that growing up with a garden, avoiding antibiotics and nonsteroidal medications, and avoiding smoking and intestinal infections are associated with lower rates of IBD. But for parents to achieve that is pretty difficult. Based on recent research, some of which has been done by Dr Colombel, we may be able to identify children and young adults, family members who are at higher risk for future IBD and potentially have the ability to intervene and possibly prevent IBD. So Jean-Fred, can we start with how can we identify patients at risk for future IBD?
Colombel: What is fascinating is that we are now convinced that there is life before symptoms. There is a preclinical phase of IBD. So basically, you start with someone who is not yet a patient who is predisposed and has a gene and then is in contact with the environmental factors that you are discussing, which is still a kind of black box. And then progressively, there is this initiation phase and amplification, which is still silent. There are no symptoms. Then, suddenly, you have symptoms. So the goal is to identify those different phases, right? And then to find predictive tools that can accurately determine the risk throughout these different preclinical stages, and then that can serve for prevention studies. So, it’s a long way, but I’m certain now that it’s feasible. There is a growing interest, basically, across all immune-mediated diseases about that.
Higgins: And it’s very interesting. You’ve done a lot of work on serology or antibodies that are circulating in the serum of the blood that can increase over time and identify people at risk. Whereas another group, several other groups, have looked at what’s called leaky gut and lactulose-mannitol ratios. Can you tell us more about the serology and what these antibodies are against?
Colombel: The point is that we have been mostly working on sera, taking advantage of the PREDICTS score from the US Army Corps of the DoD (the Department of Defense), which is amazing because they have serial samples. They have samples collected from recruits who eventually developed Crohn’s disease. You have samples at diagnosis, 2 years before diagnosis, 4 years up to 10 years. So you can really look at their lifespan and the progression of markers.
We have looked not only at microbial antibodies but also many different markers that you can find in sera. For instance, talking about environmental factors, we are now looking at the exposome; you can look at the metabolome, you can look at the glycome — doing this kind of multi-omics approach that we are now starting.
As far as these microbial antibodies, you are absolutely right. And it’s fascinating. Up to 10 years before diagnosis, those people who eventually develop Crohn’s disease have antibodies against microbial antigens: ASCA, CBir antigen, FlaX antigens. And even more striking, in those who will present with complicated disease at diagnosis, the titers of antibodies are higher, as though they are predestined to develop complicated Crohn’s disease. There are multiple groups across the world, especially this group from Canada: the Crohn’s and Colitis Canada Genetic, Environmental, Microbial (GEM) Project led by Ken Croitoru, who has done an amazing job locating 5000 first-degree relatives of patients with Crohn’s disease, following them prospectively, and then eventually they have approximately 100 new cases of Crohn’s disease. Then, of course, they can see what the predictive factors are for developing Crohn’s disease in a case-control study. As you mentioned, what they recently published is that there is an increased permeability, as measured by the lactulose mannitol test. So in other words, there is a defect in the mucosal barrier that is associated with the development of Crohn’s disease. But interestingly, they didn’t find any link between microbial antibodies and the increase in leaky gut. We’ve been saying, you have bacteria and you have leaky gut, and because of the leaky gut, you’re making more antibodies. No, this is not how it’s working, actually. It’s more complex than that.
Higgins: It’s clearly not a one-to-one relationship, but it’s interesting that having a leaky gut in a way that we can measure with the lactulose mannitol test and developing these antibodies which increase over time — antibodies to contents of the gut, the yeast, the bacteria — are both associated with increased risk for Crohn’s disease. It suggests that there is a barrier function problem that may worsen over time. You wonder whether there is a direct way that we could treat that if we could identify people at high risk. I should quantify that a little bit. Jean-Fred, in your serology studies, when somebody is very high risk, how high is the risk, actually? Is that are they 90% likely to develop Crohn’s, or is it more like 15%?
Colombel: Is much less than that? Unfortunately, we are not where, for instance, people working on diabetes are, because when you combine having family risk, antibodies, and polygenetic risk in diabetes, your risk is approximately 100%, and we are we are not yet there. For instance, when you have those serologic markers, your risk of developing Crohn’s disease is multiplied by six. So it’s not at all at the same level. What we need to do, and you are absolutely right, is to combine different markers, serologic markers, maybe microbial markers, permeability, and, maybe exposome markers and the metabolome, so that we can basically model the risk and have a predictive tool that would be accurate enough so that we can propose a prevention study to people at risk.
Higgins: It would be an amazing to get there and to be able to intervene. At this point, if we’re focusing on the barrier function, there aren’t a lot of treatments directed toward improving barrier function, although some of our older treatments actually do improve things like zonulin and the tight-junction function — including steroids, oddly enough, one of our oldest therapies. So there may be not just anti-inflammatory benefits to steroids, but also more recently, JAK inhibitors have been shown to have some benefits for barrier function in improving tight junctions. And thinking about whether we could get there with multi-omics to a prediction where you could tell a parent, your 15-year-old has an 80% risk of developing Crohn’s disease in the next 10 years. Do you think we would be ready for an intervention trial with either short courses of possibly budesonide as a lower-risk steroid or a JAK inhibitor?
Colombel: We are doing a study to understand what parents would be ready to accept for their kids, because this is a very important question. For instance, let’s imagine I am a father of four kids, and I know that my son has a very high risk. I’m not sure I would be ready to accept an intervention with biologics or a small molecule and so on, but maybe something that’s safe, such as the diet, changing the microbiome. In diabetes, they have done this beautiful study using teplizumab, which is an anti-CD3 antibody. This is a strong intervention in kids who are at very high risk for Crohn’s disease. And they were able to show that they are not able to completely block the occurrence of diabetes, but they are significantly its delaying the occurrence. Same in rheumatoid arthritis (RA), where they did a study in patients with mild articular pain and citrulline antibodies. They had the guts, if I may, to give them one infusion of rituximab, and again they were able to delay the onset of RA. I don’t think we are there yet, and I hope we will find something with a very good risk-benefit ratio that parents and those kids would be okay to accept.
Higgins: That discussion of prevention and microbiome leads to a really interesting paper that you mentioned earlier. Very recently, it was shown that in a similar sort of serology study, development of Epstein-Barr (EBV) antibodies was associated 32-fold — a really strong association — with later development of multiple sclerosis, another autoimmune disease. In journalist common parlance, they’re starting to talk about multiple sclerosis as a sort of “long EBV,” kind of the parallel to long COVID. And we’re starting to think more about the long-term implications of infections. What does this association between EBV and multiple sclerosis mean for other autoimmune diseases?
Colombel: This is a very timely question. In a month, we will present at DDW some data that we collected with Scott Snapper’s group at Harvard showing it’s not the same magnitude. But we have evidence that EBV infection can also predispose to the development of Crohn’s disease. These are data from the PREDICTS study of the US Army Corps and also the GEM Project. We are still analyzing the data, but it looks like there is also something in Crohn’s disease, which is fascinating because, to your point, they are already some companies that are thinking about developing an EBV vaccine. This could be a perfect prevention tool. I’m pretty certain they are already thinking about that in MS, because you mentioned that in MS, the risk is very high.
Higgins: Remarkably, in a similar somewhat coincidence in Australia, a few years ago they started rolling out nationwide a vaccination against rotavirus, a diarrheal illness that can sometimes be severe and life-threatening to young children. They noticed — in part because the government pays for insulin in Australia — that the rate of type 1 diabetes in children who had been in those cohorts went down significantly by 15%, suggesting that if you find the right target and can immunize against those infections, even if the vaccine isn’t perfect, you can have a significant reduction in other autoimmune diseases.
We just talked about EBV as a possible future vaccine, but I wonder if the possibilities might be even stronger in ulcerative colitis and Crohn’s disease for bacterial antigens because many, many of the genes associated with IBD are actually related to bacterial handling; many of the serologies you’ve discovered are actually bacterial. Whether it’s OmpC, the flagellum on some bacteria, or other markers, including at least one with yeast, would it make sense, in terms of trying to do the most reduction, to have some vaccinations against gut infections — some with bacteria, some possibly with yeast? We’d need the evidence and serology evidence like you are talking about with EBV. Could we come up with serology evidence for a particular bacterial species?
Colombel: The point is that I’m not sure we’d be able to identify a unique infectious agent or predominant infectious agent that we could target for prevention. This would be ideal, of course, but it’s not 100% excluded.
Higgins: It’s striking that in Denmark and Sweden, they’ve had multiple studies associating prior infection, and usually it’s Escherichia coli, Shigella, Salmonella, Campylobacter. There are usually four or five major players that mean, within the next 5-10 years, those patients are at increased risk for IBD — suggesting that, much like EBV, a prior infection may shift either the microbiota or the host reaction to the microbiota in a way that predisposes patients to later IBD.
Colombel: Maybe there is already something, before and then suddenly you are coming with an infection. You go to Mexico or you have food poisoning, and then the disease starts. I’m saying that, Peter, because I am certain you will agree that in clinical practice — even in patients who have already Crohn’s disease and UC and are doing well — sometimes they have an infection that is triggering a flare. So maybe this infection is not causal, but it’s just there to start the process, which was so far completely silent.
Higgins: It certainly seems to amplify the immune [response]. It is very well documented with Clostridioides difficile. I’m a little suspicious with norovirus. Now that we have good PCR for norovirus, we see a norovirus wave seasonally. But I wonder whether there are a lot of other diarrheal infections that we don’t necessarily detect or measure or identify a pathogen for, particularly in young children, that may be triggering in a way that explains [disease]. We struggle to identify what it is about the environment or the exposome that’s really triggering this, but I wonder whether infections are a piece of the puzzle.
Colombel: When you think about it, the strongest risk factor for IBD is exposure to antibiotics during childhood. But it’s also true for elderly people. We will present that at DDW as well, showing that elderly people — those older than 60 years — who develop de novo IBD were also more often exposed to antibiotics before the development. So, it’s a risk factor not only in childhood but also in elderly persons. And of course, antibiotics could change the microbiome, but maybe it’s a surrogate for an infection that we have not recognized.
Higgins: It’s hard to know whether it’s treating an infection that we’re not great at measuring or whether it’s opening an ecological niche for a new pathobiont to occupy. But it raises the idea that reducing antibiotic use when not necessary is important, especially in children. Using our mRNA technology, we might be able to have multivalent vaccines fairly soon, and it might be an important application. Even if, as in Australia, you reduce diabetes by 15%, if we could reduce Crohn’s disease by 15%, that would be remarkable.
Colombel: I suspect that parents would be ready to accept that, because they would reduce the risk for something that may happen. But hypothetically, they will also reduce the risk for a chronic disease, which would be perfect.
Higgins: I don’t know if you have any early results from your questionnaires for parents, but are they willing to consider vaccines, or is it too early to say?
Colombel: We don’t have results yet. And, we need to do this study across continents, because I suspect that the parents from US and parents from Europe may not be ready to accept the same risk. There are huge cultural differences, and this is why also we need to extend these prediction prevention studies to other continents. We are now trying to develop some studies in India, Brazil, and Saudi Arabia, because maybe the environmental factors in those countries are completely different.
Higgins: I think about the RA studies and the diabetes studies, and it’s very interesting. They’re doing short term interventions to try to turn the clock back. I wonder whether 8 weeks of budesonide or 8 weeks of a JAK inhibitor might be acceptable. Clearly, parents would struggle with long-term immunosuppression.
Colombel: I fully agree. And even a long-term diet will not be easy for kids, because kids as they are don’t like diet. So it has to be either something that is strong enough to completely block the development of the disease, even though it’s given short-term, or something that is more acceptable that you can give long-term. And, Peter, you are absolutely right; maybe we will not be able to prevent the disease, but we may be able to attenuate the disease or delay disease onset, which is already be big, big progress.
Higgins: Delayed onset is clearly a win. But if we could take those really high-risk people with high serologies and even if we reduced their severity significantly, that would be a huge win.
Colombel: What we discovered recently — and you will like this story — is that in the PREDICTS study, we found anti–granulocyte-macrophage colony-stimulating factor (GM-CSF) antibodies that precede the development of Crohn’s disease up to 10 years. They are very specific to Crohn’s disease and are associated as well with complicated Crohn’s disease. This can lead to a therapeutic intervention because those antibodies are directed against a modified GM-CSF. And then we can basically build a new GM-CSF. That could be a fantastic intervention in those who have those antibodies.
This is where it becomes very complex, Peter, because maybe we have the phenotype of Crohn’s disease, which is driven by multiple different pathways, and maybe the prevention will not be the same in those people who have anti–GM-CSF, those people who have increased lactulose mannitol test, those with high antibodies. This is where it may be even more complex than diabetes and RA.
Higgins: It certainly sounds like we have multiple targets and a lot of potential here. I think the examples of diabetes and RA are very encouraging as we know more and learn more and how to predict this.
At this point, I think the key takeaways are that we’re learning more and more about infections leading to autoimmune diseases like MS and diabetes, and that there are possibilities for prevention in both diseases and potentially vaccines early on to prevent EBV infection, which clearly might help in MS but might also help in Crohn’s disease, as we’ll learn next month at DDW. It’s a topic for the future, but we have real possibilities for prevention of IBD.
Colombel: I agree. This is my dream. Thank you, Peter.
Higgins: Thank you, Jean-Fred.