Crohn's Disease Podcast

Breaking the Therapeutic Ceiling in Crohn's Disease

Peter Higgins, MD, PhD; Bram Verstockt, MD, PhD


August 23, 2022

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, and director of the Michigan Clinical Trials Support Unit at the Department of Internal Medicine at the University of Michigan, Ann Arbor, Michigan. Welcome to Medscape's InDiscussion series on Crohn's disease. Today we'll be discussing breaking the therapeutic ceiling.

First, let me introduce my guest, friend, and colleague. Dr Bram Verstockt. Bram is a consultant gastroenterologist at the inflammatory bowel disease (IBD) unit of the university hospitals in Leuven, Belgium, originally founded by Professor Paul Rutgeerts and currently led by Séverine Vermeire. After having spent 1 year at Cambridge University and 3 more years in the IBD lab in Leuven, he defended his PhD thesis on personalized medicine in IBD in 2019. Besides his clinical duties in the IBD care program in Leuven, Bram is developing intestinal ultrasound as a new part of their multidisciplinary IBD care program. Furthermore, he does a lot of clinical and translational research on disease monitoring and predictive medicine and is an ideal person to discuss targeting therapies to break the therapeutic ceiling in IBD. Welcome to InDiscussion.

So Bram, tell me, what is it about IBD that first drew you to this field?

Bram Verstockt, MD, PhD: When I was in medical school, I did my last rotation before graduating and was really inspired by Professor Paul Rutgeerts at the time. When I started my residency in internal medicine, within the first weeks already, I felt that gastroenterology was really the specialty that I would dedicate the rest of my career to. I felt also, bearing in mind what Professor Rutgeerts had told me, that IBD was so close to my heart. I quickly reached out to Séverine, asking her to start a PhD in her team. I'm so grateful about where I am today and that I can join their team as a consultant nowadays.

Higgins: Leuven has had, largely due to Dr Rutgeerts of course, an amazing amount of foresight in storing bio samples and developing a comprehensive program for IBD. And that investment upfront is really paying off in very effective research.

So, Dr Verstockt, I often have patients who are referred to me who have expressed frustration that they tried several prior therapies before they found one therapy that worked really well for them. They wish that they could have started on their current therapy far earlier and wonder if they could have accumulated less bowel damage, fewer penetrating complications of Crohn's disease, and fewer surgeries if they had been able to start on the best therapy for them shortly after diagnosis. It brings up the question of what impact could be possible if we were able to start every patient with Crohn's on a highly effective therapy within 6 months of diagnosis that produced deep biologic remission? What do you think the impact would be globally on these patients and across patients with Crohn's disease generally?

Verstockt: Indeed, one of the biggest unmet needs at the moment — the holy grail — is that if we can really unravel what is driving the disease in a patient, we can basically, at the start of their disease, envision how their disease course would look and we then can hit the disease very hard from the start. We can make a huge difference for many, many patients. If you look from an epidemiologic point of view, we see more and more in the biological era that there is a decrease in resections in Crohn's disease, in hospital admissions. So slowly, step by step, we are getting a little bit closer, but I entirely agree and feel the same frustration day by day.

Indeed, at this stage, unfortunately, it's still a wild guess in so many patients that if we hit hard from the first time and if we had tried, then we are so lucky. But unfortunately, in so many patients we have to cycle through so many lines of therapies. Although this is, for me, absolutely the way forward; we also have to realize that, unfortunately, the complexity of the biology of IBD is still extremely difficult. Even if we have a very potent drug in a given individual, which is really matched to the disease biology, still for various other reasons there can be a loss of response, immunogenicity, pharmacokinetic issues. So there is much more. But I absolutely agree, if we can hit right from the start, that would be an amazing step forward.

Higgins: You know, in theory, we could reduce a lot of the bowel damage that leads to surgery. Now, the concept of precision medicine seems to have largely come from oncology, where they can find specific causative genetic mutations. A classic example is the BCR-ABL translocation that you can identify that can be targeted by specific molecules. How often are we in that situation in Crohn's disease and how complicated is Crohn's disease compared to these targetable cancers?

Verstockt: To me, that's a comparison we make more and more now, and that's absolutely where we have to aim for. The oncology field is years [ahead of] us; if you look at cancer as compared to IBD, in many tumors — although obviously cancer is very complicated as well — you can really drill down into the single mutation that is driving the biology of that given tumor. Then you can, for instance, with immunotherapy, really target the biology of that tumor. However, in IBD, I think that the complexity of the biology driving the disease is so different in that you have a genetic background, which absolutely does play a role, but there is so much more in terms of environmental factors which we still don't understand at all. That, to me, is why it's so hard to drill down and to identify the markers many of us are looking for at the moment.

Higgins: That's great. Is there usually one best therapy for a particular patient with Crohn's disease, or is it possible that several mechanisms of action may be effective in a given patient?

Verstockt: That's an excellent question. Presumably in some patients, it will indeed be the case that there are few routes of biology and pathways driving their disease. The only thing that we can see at the moment with what we know is that there are some clinical phenotypes for which there are preferred therapies. Thinking of Crohn's disease, for perianal disease, everyone is going for infliximab or an anti–tumor necrosis factor (TNF), at least in the first-line. Luckily, with all of the drugs coming onto the market — more potent drugs coming onto the market — our armamentarium is growing for so many patients. But there are so many unanswered questions at this time before we can really say, "Well, this is exactly the phenotype which matches that drug for that patient."

We also feel that for the clinical phenotypes that we are seeing, to me and to many others, the way we classify them clinically is definitely not sufficient. It's such a complex entity. It's basically a continuum of various different kinds of Crohn's disease molecularly. That to me is one of the main reasons why we really have to invest in starting to unravel from the biology: What are the different subgroups of patients? Then we can go much more into targeted therapies as we, for instance, are doing in oncology at the moment.

Higgins: Another phenotype that is interesting — and at this point the data are a little thin — are the patients who come in with Crohn's disease plus autoimmune skin diseases, like eczema or psoriasis (especially psoriasis), that resulted from [using] an anti-TNF therapy who seem to be very responsive to anti–interleukin (IL)-23 therapy.

Verstockt: I entirely agree. Another phenotype you might think of are those patients who are suffering from Crohn's disease and some rheumatologic conditions; there is emerging evidence, for the JAK inhibitors as well, that we can make a huge difference there. We start to see some glimpses, and as you properly acknowledged, the data are still very thin, but still we are making steps into the right direction there as well.

Higgins: Assuming that a particular patient in your clinic with Crohn's disease would have a significantly better response to anti–IL-23 than anti-TNF, how much better would that response have to be — 5% or 20% delta? — for a precision medicine approach to be practical?

Verstockt: That's one of the most important questions when it comes to companion diagnostics. Of course there is, first of all, the efficacy. But then, at some stage, payers will have obviously their view on this when it comes to cost-effectiveness. The availability of biosimilars might also drive part of that answer: If you're looking at two very expensive drugs, then, potentially, the difference between the two might be smaller as compared to a very cheap biosimilar. We see in many countries that payers are basically forcing physicians to prescribe the biosimilars, although there is emerging evidence for the more expensive ones. So that plays a role.

But obviously on top of that, if you look at it from a true scientific point of view, then we have to rely on what we already know from literature when we look at the efficacy overall for these drugs. If you can clearly prove with a biomarker that you can significantly exceed the efficacy that we've seen in the trials, then you have a strong case, and it's no longer a wild guess. Then, it absolutely justifies using that biomarker and getting it reimbursed together with your drug, basically.

Higgins: Right. So companion biomarker diagnostics hopefully are going to come in the future, particularly with possibly things like serum IL-22 for the anti–IL-23s. Assuming that we can find patients in whom there is going to be a significant therapeutic difference, how accurate does the novel prognostic have to be to make this a reality?

Verstockt: It's a bit in the same lines. If you can prove that you can significantly exceed the efficacy that you see in an all-comers population, then of course prevalence plays a role there in establishing and in confirming your predictive values. But there are some clear data showing that indeed, if you can [prove that you can exceed the efficacy in all-comers] with your biomarker and can significantly increase your pretest likelihood to a very high positive predictive likelihood, then we are speaking about biomarkers which can truly make a difference in clinical practice. Obviously, then will come the question of what is the minimal threshold we will require? But still, I think if you can make a significant and clinically relevant difference for patients (and this is just a wild guess) — if, for example, you instead of having 1 out of 3 patients responding to a drug, you can increase that and your accuracy is up to 60% or 70% — then in clinic, that's making a huge difference for patients if you can say, "We can double the likelihood that you will respond as compared with an all-comers population." To patients, that's something that matters and that can significantly impact their quality of life in the long-term.

Higgins: That may be the tipping point for a payer to say that they will pay for the more expensive new drug compared to the cheaper biosimilar.

Verstockt: I think there is a large responsibility for the payers and also for the FDA and the EMA [European Medicines Agency] to really aim at getting these companion diagnostics that are being developed during a drug development program. IBD is a very complex entity, still during drug development, it should at least be an attempt to get these biomarkers discovered. If they fail, then unfortunately they fail, but at least it should be incorporated into trial designs more and more if we want to make a difference in the long-term for patients. Twenty years ago, this wasn't an issue because we were really grateful that we had the anti-TNFs and we had something which was more potent and more safe than steroids. With the market being so crowded and so complex with very potent drugs, all with more or less a similar efficacy, that's where we can truly make a difference. From a company perspective, yes, you might narrow your potential market, but still, the gain might be so much larger if you can come into the market with the drug and a biomarker which increases the likelihood of response in the long-term.

Higgins: That's going to be increasingly important as the market gets crowded. Upadacitinib, which is remarkably effective for ulcerative colitis, is still second-line after anti-TNFs per the FDA approval. It would be interesting if there were a companion biomarker that could identify patients who would be incredibly responsive. It might make sense to make it first-line for that group of patients.

Verstockt: That's something we see in oncology and absolutely have to aim for in IBD. Then of course, for the companion diagnostics, we have to make cost effectiveness studies and not just look at the short-term for endoscopic remission, for instance, but also in the long-term in terms of how many patients can we prevent to need surgery to be absent of work. All of these things can truly make a difference, especially if we look at it from a broader perspective than just short-term efficacy.

Higgins: That's hopefully more likely to be true looking at the long-view. In countries where you have a centralized national health service… Unfortunately, in the US, we're very obsessed with choice. The downside of that is that Americans tend to change their insurance about every 2-3 years, which means the time horizon for what the insurer cares about is 2-3 years. It's an unfortunate problem with the American healthcare system, but hopefully in places where you're committed to caring for someone long-term, it makes sense to invest upfront.

I know that you've thought about this a lot with the biomarkers. What technologies or approaches are likely candidates for discovery of these prognostic biomarkers, whether they're proteomics or transcriptomics? It seems genomics won't be enough, but what do you think is likely to work?

Verstockt: That's an excellent question. And still, the answer is honestly that we don't know. That's why so many people are collecting so many different biomaterials and that we are applying so many different technologies, hoping that at some stage, some will truly show it. From a patient perspective, obviously, if we can come up with a noninvasive marker like a blood test, that would be the ideal case. This is what is currently being investigated for the prognostic biomarkers with the Cambridge signal at the time, which has been converted into a whole-blood commercially available test. So that to me is the way forward. Will it be feasible for all therapies? We don't know. What I do experience more and more myself is that, presumably as compared to a few years ago, I no longer believe that a single marker or a single protein will do the job. The disease is way too complex to capture that with a single protein or a single gene transcript. Whether it will come from blood, from tissue, or even the stools, that's still an open question. The less invasive we can be the less invasive we definitely should be. At some stage, it presumably will be a balance between accuracy and feasibility also for patients, and presumably, the final answer lies somewhere in the middle.

Higgins: Is it possible that disease course and past therapies or even current therapies could potentially change the biology and change the best therapeutic mechanism of action for each patient over time? Will we have to retest people with this prognostic panel during their disease course to see if it's changed?

Verstockt: Absolutely. That's something I strongly believe in. Of course, it really depends on the type of biomarker that you're looking at. If you're looking at a genetic test, obviously that would change. But if you're looking at proteomic markers, if you're looking at transcriptomic markers that's significantly affected, for instance, by steroids — we know that steroids are a very strong potentiator of so many different mechanisms, so they will affect many of these markers. So absolutely this is something which we have to be very aware of, that the condition or the clinical scenario where you are applying that biomarker absolutely should be in line with the scenario where the biomarker was originally founded. Otherwise, it presumably wouldn't work.

In terms of disease biology, also, there is emerging evidence, for instance, that in patients in whom an anti-TNF therapy is failing, after a while you see a switch in their immune system toward IL-23 mechanisms coming up; IL-23 receptor expression on their immune cells might increase their likelihood to respond to an anti–IL-12 or -23. So yes, there is molecular and biological evidence coming out that there is a significant influence of previous therapies. That's also something we experience on a daily basis in our own practice.

If you look at all of the trials, we know that the bio-naive populations are those who respond the best. There might be a bias, obviously, by disease duration, and the way their disease is evolving. But presumably, or at least hypothetically, there might also be a biological interference there with the therapies they've been on in the past.

Higgins: It certainly seems… I think you're referring to Raja Atreya's work out of Erlangen, Germany.

Verstockt: Yes, indeed, absolutely. It's the Erlangen group who did brilliant work on that.

Higgins: But it suggests that failure of certain therapies, like anti-TNF, may actually prime a response to other therapies like IL-12, making it seem like sequencing therapies might actually turn out to be pretty important and monitoring how your phenotype changes over time on a biomarker level may be really important.

Verstockt: That will definitely influence the cycling. Now, we cycle from one to another without exactly knowing what we are doing from an immunologic point of view. That's the way the Erlangen group is doing it. It is absolutely brilliant to further unravel that process.

Higgins: Turning toward that idea of multiple mechanisms of action, the recent VEGA study uses the combination of guselkumab and golimumab, an anti–IL-23 and an anti-TNF, and it appears to be breaking the therapeutic ceiling and getting a higher response in the neighborhood of 85%. What does this mean going forward if combination therapy is an option and we'll be able to identify responders to specific combinations of dual-targeted therapies?

Verstockt: I think this is one of the ways forward. Absolutely. It's presumably not the only one, but it might be one of the options to break that ceiling. Of course, we have to be cautious because for VEGA, we only have the week 12 data and we have to see how this will be translated into the maintenance program. At Digestive Disease Week (DDW) there will be the EXPLORER data being presented as well. So a lot is moving there.

Presumably that's one of the ways that we can try to dampen various inflammatory cascades, from various angles at the same time. Especially if we know in a given patient what is driving their disease, then there is absolutely a rationale for doing so. But again, it's a bit the same to me as with monotherapy that we have to understand in a given patient, what is the driver of their disease? If it's a strong TNF-driven disease, presumably a monotherapy might be sufficient. Whereas if we clearly see that multiple pathways are being involved, then combinations should absolutely be the way forward, at least to a certain extent. Whether or not this is for a long time, that's another discussion, obviously.

Higgins: Getting back to the oncology analogy, we know that in oncology, using a single chemotherapy and blocking one pathway generally leads to resistance. I think about the microbiome of the gut stimulating the gut immune system. Do you think monotherapy and resistance is a concept that applies to Crohn's disease and may necessitate dual-targeted therapy?

Verstockt: It presumably will in a subset of patients; there are absolutely patients who, over time, lose response and we really don't know why [because] they have a good exposure to the drug. There is no immunogenicity problem. Still, after doing very well for years, all of a sudden, they lose response. Presumably, there are these mechanisms, yes. Whereas in others — we have patients for 20 years on anti-TNF who are doing really brilliantly well. The question then obviously comes up: Is this a drug or is this just a natural cause of the disease? That's another option, obviously. But the resistance and, from an immunological point of view, yes, that could absolutely play a role and that justifies, again, why we should further explore combination therapy. To me, combination is not only combining two biologics or two anti-inflammatories. I think there are other ways to combine therapies considering, as you alluded to already, the microbiome and dietary interventions might support a combination in a way as well. Other drugs that are reinforcing epithelial barrier, etc., might also be potential future options of reconsidering and redefining the concept of combination therapy.

Higgins: Absolutely. You had mentioned dual-targeted therapy for a limited period. Do you think, especially early on when you were talking about hitting patients [with treatment] early, that there is a role for more aggressive [therapy]: either potent JAK inhibitors or dual-targeted therapy for induction? But then step down to a less intensive therapy for maintenance? It seems like at this point, a planned stepdown is relatively uncommon in clinical practice. We tend to stick with what worked. Can we predict which patients can step down safely?

Verstockt: That's an excellent question and so far, we don't have the answer. I do agree conceptually, but this is all hypotheses obviously, that if we are able with very potent drugs or even combination of drugs, if we can bring patients very quickly into remission, then we then can continue with a monotherapy. Obviously, we will need the trials, etc. Then again, the biomarkers will play a role. Edouard Louis, for instance, has done a lot of work on discontinuing anti-TNF therapy, not just from a clinical perspective but also trying to identify biomarkers. There seems to be some promising proteins there which might predict whether or not a patient will respond upon anti-TNF withdrawal, for instance. But the final word again from this team is not yet out unfortunately.

Higgins: So we've talked about the importance of differential effectiveness between therapies and phenotypes that identify potential responses to anti-TNF or anti–IL-23 or even JAK therapies. Precision medicine in IBD is not here yet, but we're vigorously looking for biomarkers that can predict who needs which therapy, who might need combination therapy, and who might be able to step down in therapy to less intensive maintenance therapy.

Bram, thank you so much for joining us. This has been very informative and great to have your expertise on this topic.

Verstockt: Thank you for the very nice discussion, Peter.


Crohn's Disease

A blood-based prognostic biomarker in IBD

Role of the IL23/IL17 Pathway in Crohn's Disease

A Study of Efficacy and Safety of Combination Therapy With Guselkumab and Golimumab in Participants With Moderately to Severely Active Ulcerative Colitis (VEGA)

Results of Novel Clinical Study Show Adults with Moderately to Severely Active Ulcerative Colitis Achieved Higher Rates of Clinical Response, Clinical Remission, and Endoscopic Improvement at 12 Weeks with Guselkumab and Golimumab Combination Therapy Versus Either Monotherapy Alone

Digestive Disease Week 2022

Triple Combination Therapy in High Risk Crohn's Disease (CD)

Anti-TNF and Crohn's Disease: When Should We Stop?

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.