Crohn's Disease Podcast

Anti–interleukin-23 Medications and Durable Remission for Patients With Crohn's Disease

Peter Higgins, MD, PhD; Raja Atreya, MD


July 21, 2022

This transcript has been edited for clarity.

Peter Higgins, MD: I am Dr Peter Higgins, professor of gastroenterology and director of the Inflammatory Bowel Disease (IBD) Program at the University of Michigan and the chair-elect of the Crohn's and Colitis Foundation, National Scientific Advisory Committee. Welcome to Medscape's InDiscussion series on Crohn's disease.

Today we'll be discussing anti–interleukin (IL)-23 medications, which have demonstrated durable remission in a proportion of patients after the drug is stopped in the maintenance phase of clinical trials. First, let me introduce my guest. Dr Raja Atreya is senior physician and head of the Inflammatory Bowel Disease Unit, Outpatient Clinic and Clinical Study Center at the University Hospital Erlangen in Germany. He is also Heisenberg Professor of Translational Immunology and IBD and Erlangen University. Dr Atreya, welcome to InDiscussion.

Raja Atreya, MD: Thank you very much, Peter, for this kind invitation. I'm really looking forward to discussing this very interesting topic from a scientific and clinical point.

Higgins: I always like to ask each of our guests, what is it about IBD that first drew you to this field?

Atreya: Very good question, Peter. Thank you. If I may think back about 20 years ago, I did my medical thesis examining signaling pathways that are involved in driving inflammatory bowel diseases. And that's how I became interested in this lifelong immune-mediated disease, because you could already sense at those times that growing insights into the immunopathogenesis of inflammatory bowel diseases led to the advent of new targeted therapies that improved outcomes and quality of life for our patients. So I really liked and still like this translational combination, this translational approach toward disease, because I think the more findings we have regarding the immunopathogenesis, the better therapies we can offer our patients.

Higgins: It's really been a fascinating development over the past 20 years, as our increasing understanding of the disease pathogenesis has changed therapy in a way that really benefits patients.

So, Dr Atreya, I've had several patients in anti–IL-23 clinical trials; these are the p19-specific monoclonal antibodies, including trials of mirikizumab, risankizumab, and previously ustekinumab. These patients generally respond to the drug in the induction phase over 8-16 weeks and then are re-randomized to drug or placebo in the maintenance phase, which usually lasts for about a year. And usually in most IBD trials, in the maintenance phase it becomes fairly clear who is re-randomized to placebo as they begin to flare 8-16 weeks after stopping the drug from induction.

But there's an interesting pattern in the anti–IL-23 trials. It appears that 20%-30% of these patients stay in remission without any further drug doses for the entire maintenance period of 1 year. This has led some observers to criticize network meta-analysis of maintenance trials as unfair to this drug class because they appear to have a surprisingly high placebo remission rate. But it raises the question: Is this really a placebo remission, or is this a long-standing remission induced particularly by this drug class?

I think you're the perfect person to discuss this, given your interest in cytokines and pathways and switching. Do you think that we have seen enough data from trials of different anti–IL-23 medications, and anti–IL-23 in the case of Stelara, to say that this seems to be a pattern in this mechanism of action?

Atreya: Thank you, Peter, for pointing out this very interesting effect that, from my point of opinion, we can see across all the trials with selective p19 inhibitors. I think we all know from clinical practice, of course, that the patients who respond to induction therapy are those patients who have the highest chance of having long-term remission. And these clinical trials were really designed such that the responders to the induction therapy were then switched — half of them to placebo therapy, for instance — and you could still see, I would say, a sort of a carryover effect. We have to point out that these were patients with really severe disease; they had a high clinical and endoscopic disease course. They had been exposed to previous biologic therapies, sometimes even multiple biologic therapies. And we all know from clinical practice that these are the patients who are really very hard to treat effectively, and that's really a very high challenge for the substance that we apply it to.

So I think these are very convincing rates that we see in patients that are maintained with the study medication, but it is also still a carryover effect of patients who are treated with placebo. It shows us that effective therapy with the selective p19 inhibitors during induction is somehow able to change, I would say, the immunophenotype of these patients for a long-lasting result, because these patients benefit from 12 weeks of therapy, for instance. And after a year of therapy, you still see remission rates of about 40%-50%, which is really remarkable in this highly selected group. I think this shows that selective inhibition of IL-23 is able to somehow change the immunophenotype in a way that has a long lasting remission. And that's the exciting effect that we are witnessing here.

Higgins: To play devil's advocate, the design of the responder randomization studies for maintenance assumes that these drugs will wash out and you'll see a clear difference. But is there any reason to think that these anti–p19-selective biologics might not be washing out — that they're somehow retained in the body and might still be acting 52 weeks after the last dose?

Atreya: I don't think so. Partly this might be an explanation, but if we look at the half-life, for instance, which is always a good measure, I think there is nothing unusual where you would say, the half-life is so long that has retained the patient. I think we have seen this in most of the other drug classes, at least biologic therapies. You always see this waning effect in the maintenance trial, and this is not on a comparable level visible in the selective p19 inhibition trial. So this shows me somehow that IL-23 is a central cytokine that drives the inflammation in inflammatory bowel diseases. And that inhibition of this, I would say, pivotal cytokine is able to induce a long-lasting effect also on the immune levels. I would be very interested to see, because of course there were tissue probes and biopsies also collected in these trials. And I would be very curious to find out how far significant changes in the immunophenotype can be correlated to effectiveness in these patients.

Higgins: I agree that this is structurally a monoclonal like any other monoclonal, and that this seems to be a durable drug-free remission. And it seems that the anti–p19-selective biologics are changing the immunophenotype. Do you think this is going to be durable even beyond a year? And do you think it might be an example of Th17 cell plasticity?

Atreya: That's a very interesting question from a scientific point of view. This would introduce a totally new concept into biologic therapies. We are all very hesitant to stop any efficacious therapy, because we always have the risk of, does the patient respond as well to this therapy if we start it again? Of course, immunogenicity is always a problem, and I think this is not, from all the data that we have seen so far, a really critical issue in selective IL-23/p19 inhibitors. So I think, if we are speaking about something like a drug holiday, would it be possible with this drug class? I think long-lasting trials will have to show this, but it would be very interesting to see, is this an effect that goes beyond this 1-year time frame, which you already mentioned.

Then, of course, there's also what happens in the patients on the mucosal level and immunophenotype if the patient then flares again. Is there perhaps even an early predictor on immunologic level for a flare? And if there is a predictor, could we use this to initiate the therapy again?

These are all very interesting questions, but it's very reassuring to know that this is a long-lasting effect. And if you then, as we do in clinical practice, continue an efficacious therapy, this gives us very good hope that this is then a therapy that could last a very long time at a very high effective level in these patients.

Higgins: It's interesting, the immunologic switch that we're proposing that these p19s accomplish — do you think that might also reduce their risk for immunogenicity by perhaps, as a secondary effect, turning down the immune response and maybe protecting the drug from antidrug antibodies?

Atreya: I think the specialty, at least from the data that we have of these inhibitors, is that they work on two different levels. On the one level, I think that is clear. They inhibit the pro-inflammatory function of IL-23, and we all know that IL-23 is an indispensable factor for the generation of Th17 or, let's say, pathogenic Th17 cells, and of course correspondingly also of the pro-inflammatory cytokines that they produce. But I think that is not the only effect that can be attributed to selective p19 inhibitors, because by inhibiting IL-23, they're also inhibiting the blockade of regulatory T cells because IL-23 is able to downregulate the function of regulatory T cells — which, of course, have an anti-inflammatory property. So I think this is a recipe — if I may call it like this — that if you are able to block the inhibition of regulatory T cells and thereby strengthen their expansion and their functional possibilities, this would qualify for a very good maintenance therapy because a very good functioning regulatory T-cell system is, I think, one of the key ingredients of maintaining remission in these patients. This is something that is very unique about this substance class. And I think that would, at least from an immunologic point of view, explain the very good and efficacious maintenance data that we see.

Higgins: Absolutely. As I began to see this phenomenon across the different trials of these durable drug-free remissions, it really reminded me of your 2019 paper in Gut on anti–tumor necrosis factor (TNF) therapies and how they can shift immune cells in the gut in nonresponders to increase their expression of IL-23 receptors during resistance to anti-TNF therapy. Can you tell us more about your work on this phenomenon?

Atreya: I think we started this work, and of course we are all concentrating on finding predictors of therapeutic response to therapies. This is very important and one of the main questions in the future, because this would tell us to use the drugs at utmost efficacy level and stratify patients to each drug according to their response probability. But I think another question, is why does an ongoing therapy fail in some patients? We know this from the anti-TNF therapy, immunogenicity, that pharmacokinetics (PK) plays a very important role, but if we really look at these data closely, we can see that in patients, for instance, with anti-TNF antibodies, there's a high proportion who have sufficient trough levels and no antidrug antibodies. So in theory, these patients must really respond very well to the ongoing anti-TNF therapy. But nevertheless, the therapy fails in a high proportion of these patients. I think this indicates that there's something happening on a molecular level that drives what I would call resistance to a biologic therapy.

The term "resistance" to a therapy was introduced first in the oncologic field. But we all know from our colleagues that cancer becomes resistant to an ongoing therapy, and by identifying the resistance mechanisms, you can then of course choose the next subsequent most efficacious therapy that you would then switch to. I think the work that we did is the first introduction of this concept into the field of immune-mediated inflammatory diseases, like inflammatory bowel diseases, because we have seen that in patients with ongoing anti-TNF therapy, there's an expansion of resistant T cells to the anti-TNF therapy. Normally when you initiate an anti-TNF therapy, there's induction of programed cell death — apoptosis. So the pathogenic T-cells that initiate and perpetuate the disease undergo programed cell death due to the efficacious therapy with anti-TNF therapies.

Now, what happens on these levels in patients who do not respond to ongoing anti-TNF therapy? We could see that IL-23 plays a very important role in these patients because it's produced on a higher level by macrophages in the mucosa. This then leads to the expansion of so-called IL-23 receptor–positive TNF receptor 2–positive T cells. The unique feature of these T cells is that they become resistant to this programed cell death mechanisms that anti-TNF therapy normally induces when it's efficacious in patients. So this expansion of these resistant T cells to anti-TNF therapy then changes the immunophenotype driven by IL-23 in these patients. So they are resistant to anti-TNF therapy. This then indicates, because IL-23 is the main driver of this resistance mechanism, that IL-23 might be a very suitable target then for the next therapy.

We have seen in phase 2 and phase 3 trials that there's a remarkable efficacy of this biologic therapy in this class of patients who have been exposed, for instance, to anti-TNF therapy. And I would say — and you've already mentioned the network meta-analysis, for instance — we have really seen that among biologic therapies, these IL-23–selective blockers seem to be the most efficacious therapy in the patients who have been exposed to previous anti-TNF therapy, for instance. This insight into the immunologic mechanisms that drive resistance to therapy then presents us with the next best target for a most efficacious therapy.

We are currently studying what the mechanism is that is driving the upregulation of IL-23 in these patients. We are still collecting data and examining tissue levels and tissue from patients who do not respond to ongoing anti-TNF therapy. Growing insights into this field would then help us perhaps also to find the best algorithm for therapies — for example, how can we switch from one therapy to the next with the highest probability of responding to subsequent therapy? This would perhaps create an individual therapeutic algorithm for each individual patient. I think that would be a next step toward an utmost goal of personalized therapy in each individual patient.

Higgins: It's really interesting that we're seeing more and more patients who have adequate drug levels and have previously responded to anti-TNFs but have developed resistance to anti-TNFs. Your work has shown, at least in a subset of patients, why that happens — that there's actually an immunologic switch to resistance. And the interesting piece of the anti-IL-23s is that they seem to switch people into a durable remission. So this idea that a patient with IBD is not constant — that they can actually switch and have a new immunophenotype over time — is really fascinating.

Clinically, a lot of us have seen the sequencing issue that if you have a patient who's been on an anti-TNF and the anti-TNF has induced a drug-induced psoriasis, those patients respond amazingly well to an anti–IL-23, suggesting that the TNF therapy might even prime patients to respond to an anti-p19 therapy. And the idea of sequencing in a thoughtful way may be the future of IBD therapy.

But another piece that people have talked about, particularly in relation to our oncology experience, is, would it make sense to do combination therapy if we know that treating with a TNF will produce some patients who develop resistance and prime that IL-23 pathway? Would it make sense to combine an anti-TNF and an anti-IL-23? I feel like that's what an oncologist would do. Do you think that's a possibility, or do you think sequencing might be a better solution?

Atreya: That's a very interesting question. Sequencing, I think, would be very important for us. For this, we really need to stratify the patients. We are still calling them Crohn's disease. But all of us know that there are so many different subtypes that have specific features, and these specific features are driven also by immunologic changes between the patients. What we need to do is longitudinally follow up each individual patient and investigate what happens in the immunologic landscape during therapy. So if a patient responds, how is the immunologic landscape changed and if the patient does not respond anymore, what has happened on the immunologic level? This would really help us target a sequencing therapy. They would really choose the best therapy and the most efficacious time for the individual patient.

Can we perhaps reach higher therapeutic efficacious levels with the combination of the monotherapies that we are using currently as our biologic therapies? I think at the recent conference at the European Crohn's and Colitis Organisation (ECCO), we have seen that the combination of anti-TNF therapy and inhibition of IL-23 leads to higher efficacy levels in patients with ulcerative colitis. I think the question is always, does combination therapy have to be a lifelong therapy, or can we de-escalate perhaps? And I think the window of opportunity for treatment success is always the induction therapy in each patient, because we know that this is the time point when we decide, does the patient respond or not? If a patient responds to combination therapy during induction, then we have a very good chance to perhaps de-escalate. And you have very nicely characterized the high efficacy level of IL-23 inhibitors to maintenance therapies.

So I would envision something like an induction therapy — a combination of anti-TNF and IL-23 inhibition. Then, if a patient responds on an efficacious level, and also on an endoscopic level and perhaps in the future in a histologic level, this would be then the prime group of patients in whom we could de-escalate the therapy and then probably continue maintenance therapy with an IL-23 inhibitor.

Higgins: And I think beyond a histologic level. Your work suggests that immunologic monitoring and following the immunophenotype are really important. It's a really interesting idea, because I think most of us think of anti-TNFs as putting a brake on inflammation. But as soon as we remove that brake, it roars back. Whereas what seems to be happening with the anti p19s is a switch in the immunologic phenotype in a subset of patients, and these patients might have much more durable response in part because of higher regulatory T-cell activity. Do you have any guesses or speculation on what might be immunologically different about the patients who achieve this immunophenotype switch?

Atreya: Growing knowledge is needed, and perhaps we can define something like a new endpoint. I would call it something like "molecular healing," because I think something's happening in these patients who respond to IL-23 inhibition that tells us that there's something happening deep on an immunologic level. We really need to investigate this. I think with these investigations, we are able to define what would happen best on a molecular level to reach the deepest remission that we can achieve with our therapeutics that we have at hand.

There are some patients for whom it's like a switch that you push that is able to shut down the complete inflammatory mechanism. These are patients that are doing very well, and this is the subgroup of patients whom we need to investigate to see perhaps what the immunologic situation was before initiation of IL-23 therapy and what happened afterward. Because there's a striking and remarkable effect that I've also witnessed during clinical trials.

Higgins: It's really amazing, to think about potentially not histologic or endoscopic monitoring, but immunologic monitoring of these patients. And it does seem like we could potentially induce these patients into a deep immunologic remission and then monitor their inflammation over time and potentially reinduce it with anti–IL-23 therapy on an as-needed (PRN) basis. So it would be more of an intermittent induction — more like cancer therapy than like maintenance therapy.

Atreya: I think that would be potentially possible because we don't have this immunogenicity problem which prevented us from using anti-TNF in the way that you just described. So I think that will be something like an on/off therapy and best combined with a marker, besides fecal calprotectin, that would predict even earlier an inflammatory reaction, perhaps on a molecular level. I think that would be a perfect combination for us and, of course, for the patient.

Higgins: Currently on a low-tech level, we would probably measure C-reactive protein or fecal calprotectin to monitor for this return of inflammation. But looking forward, would it make sense to monitor serum IL-22 or regulatory T cells in the colon, or possibly these resistant anti–lL-23 positive cells? What do you think, speculating about the future? What do you think is practical and what might be best?

Atreya: I think best would be a blood biomarker, because it will be easiest to look upon. But I sometimes think what is really happening is only seen on the tissue level. So the question is, of course, how can we get access to the tissue, or is there a surrogate marker for what happens in the tissue? Here we need a marker that is assessed before fecal calprotectin, because if the fecal calprotectin rises, there's already inflammation at hand and we need something before that. I think that will be something like a tissue marker and best perhaps with a surrogate parameter, perhaps like IL-22, which I think in some studies has shown a very nice and remarkable effect. In other studies, the results were not as clear. So much more translational work is needed in this field.

Higgins: It's really remarkable to see this kind of immunologic switch. We've seen it in patients with IBD receiving bone marrow transplants, and rarely with patients who get a very potent dose of thiopurines with very low TPMT or NUDT15 . They can't process it. It's sort of like a very mild bone marrow transplant. But to accomplish this immunophenotype switching with just a monoclonal is really remarkable. I think this is raising a new understanding of how we can control and manipulate IBD on an immunophenotype level, whereas I feel like a lot of our past therapies have largely just been putting on a brake.

Thinking about this beyond the p19s, where do you see this going in the future in terms of IBD therapy and thinking about IBD therapy?

Atreya: I think we have hopefully opened a new era of therapeutic efficacy, a new level. Understanding what happens on a molecular level will also then help us to see whether, in patients who experience disease flare under IL-23 therapy, re-initiation of an anti-TNF therapy would be efficacious based on the mechanism that we described. Could this be like a cycle of anti-TNF and IL-23, which could preserve remission in this patient? I think that will be interesting to see, going back to the sequencing concept that you already introduced.

This is something where achieving these new endpoints pushes us forward in defining new endpoints. And this is only possible because we are now able to probably attain them with the therapies that we have. I think this is really driving our field forward, and I'm looking very positively into the future because with these new substance classes, we're able to define new endpoints and go toward deep remission for our patients for a long time. Hopefully we are really entering that new era now.

Higgins: It's amazing to think about it.

Today we've had international expert Dr Raja Atreya discussing with us the unusual phenomenon of durable drug-free remission in a subset of patients on anti-p19 therapies, which seems to be due to an immunophenotype switch and deep immunologic remission, changing the way we think about therapy for IBD. What this means for future intermittent induction therapy, sequencing of therapy, and long-term monitoring remains a topic for future research, which I'm sure Dr Atreya will be deeply engaged with.

Thank you so much for joining us. This is Dr Peter Higgins for InDiscussion.


Efficacy and Safety of Mirikizumab in a Randomized Phase 2 Study of Patients With Crohn's Disease

Long-Term Safety and Efficacy of Risankizumab Treatment in Patients with Crohn's Disease: Results from the Phase 2 Open-Label Extension Study

Five-Year Efficacy and Safety of Ustekinumab Treatment in Crohn's Disease: The IM-UNITI Trial

Ustekinumab as Induction and Maintenance Therapy for Crohn's Disease

Risankizumab as Induction Therapy for Crohn's Disease: Results from the Phase 3 ADVANCE and MOTIVATE Induction Trials

Risankizumab as Maintenance Therapy for Moderately to Severely Active Crohn's Disease: Results from the Multicentre, Randomised, Double-Blind, Placebo-Controlled, Withdrawal Phase 3 FORTIFY Maintenance Trial

A37 Efficacy and Safety of Risankizumab as Maintenance Therapy in Patients With Crohn's Disease: 52 Week Results From the Phase 3 Fortify Study

Th17 Cells in Inflammatory Bowel Disease: Cytokines, Plasticity, and Therapies

Expansion of IL-23 Receptor Bearing TNFR2+ T Cells Is Associated With Molecular Resistance to Anti-TNF Therapy in Crohn's Disease

Comparative Efficacy and Safety of Biologic Therapies for Moderate-to-Severe Crohn's Disease: A Systematic Review and Network Meta-analysis

European Crohn's and Colitis Organisation

Mercaptopurine Therapy and TPMT and NUDT15 Genotype

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.