IBD Drive Time: Ben Cohen, MD, on AGA Guidelines for Biomarkers in IBD

Dr Cohen discusses with host Dr Raymond Cross the recent guidelines published by the American Gastroenterological Association on using biomarkers to monitor inflammatory bowel disease.

Raymond Cross, MD, is a professor of medicine and director of the IBD Program at the University of Maryland School of Medicine in Baltimore. Ben Cohen, MD, is cosection head and clinical director for Inflammatory Bowel Diseases in the Department of Gastroenterology, Hepatology, and Nutrition at the Cleveland Clinic in Cleveland, Ohio.

TRANSCRIPT:

Any views and opinions expressed are those of the authors and or participants, and do not necessarily reflect the views policy or position of the Gastroenterology Learning Network or HMP Global, its employees and affiliates.

Dr Raymond Cross:

Hello everyone. I'm Raymond Cross from the University of Maryland School of Medicine and welcome to IBD Drive Time, and I'm very happy to have my friend from Cleveland Clinic, Ben Cohen, here to talk about the new AGA guidelines on biomarkers and Crohn’s. Ben, welcome to IBD Drive Time.

Dr Ben Cohen:

Thanks so much for having me.

Dr Cross:

So Ben, uh, as I mentioned, the AGA just published our new guidelines on the role of biomarkers in the management of Crohn's. So what biomarkers should providers be using and, any that they shouldn't use?

Dr Ben Cohen:

Yeah, so I think, you know, this work was really important as we move more towards a treat to target approach. And I think we all know that symptom-based assessments alone don't correlate well with endoscopic assessments. So it's important for us to have these noninvasive ways of assessing disease activity. And that's really the point of the guidelines and there's been a lot of wide variability in practice in terms of how we use them. So hopefully these guidelines will help streamline that for providers. So the way we chose what we were going to look at was the guideline panel and technical review team took the biomarkers that are most widely commercially available in the US and to use as surrogates for endoscopic activity, and those would be C-reactive protein and fecal calprotectin. In the UC guidelines, we also looked at lactoferrin, but that that wasn't as relevant for the Crohn's disease guidelines. And we didn't look at any of the biomarkers that are solely used as prediction tools. And then the one other biomarker that we did look at but didn't make any recommendations specifically on was the endoscopic healing index, also known as monitoring. And while we assessed that there really wasn't enough real world evidence to make any specific recommendations around it. So we said that that's a gap area and hopefully we'll have some more research on that in the future.

Dr Cross:

Yeah, and by the way, I should have congratulated you on being part of this. So I, I've been able to do this once with the AGA and it's a really cool process and I learned a lot. So congratulations for being part of that. And I don't know if you agree with this, but one of the things that I do in practice with the residents and fellows is I try to dig in and figure out, okay, back when they were symptomatic and had this endoscopy, what biomarkers were done and did it correlate with what I saw? Because I think that you'll agree that there is no perfect biomarker for a given patient. You really try to have to do the work to find out which one reflects their disease activity

Dr Ben Cohen:

A hundred percent. I mean, if throughout the—we talk about the concept of benchmarking the biomarkers against the endoscopic activity, which is still the gold standard, and up to 20% of Crohn's patients won't mount the C reactive protein, for example. So if you know from the start when somebody has endoscopic disease activity, that they don't have an elevated CRP or calprotectin, then it may not be a particularly useful biomarker for you. We don't, in the guidelines itself, it's not part of the specific recommendation, but in the implementation considerations, we talk about this idea of benchmarking. So I say the same thing to my trainees and I'm always, especially when I first see a patient and they may have disease activity, making sure we know which biomarker is going to be useful for them.

Dr Cross:

And I was going to ask you this a little later, but I think it's relevant now. So there's been discussion that calprotectin is probably better for chronic GI disease and maybe not so good for small bowel disease. And was there any discussion in the document about disease location in Crohn's and perhaps one biomarker being better than another?

Dr Ben Cohen:

Yeah, I mean that's a really interesting point and I think I went into it with this thought that, for example, calpro doesn't work in small bowel disease and it's primarily going to be a chronic disease, but in actuality, there were more studies saying that there was no difference based on disease location than studies, that said there was a difference. At the end of the day, there wasn't enough data, high quality data, for us to make a specific statement about specific locations or the degree of elevation of the different biomarkers, but this is something certainly to look out for in the future. But we were all quite surprised in the review that we didn't see as a big difference by location.

Dr Cross:

That's super important, I think. So what are the threshold targets we're using for CRP and fecal calprotectin?

Dr Ben Cohen:

So the CRP level that we looked at in the literature was 5 milligrams per liter, which I think is a common cutoff. I mean, now there may be different labs that have a different threshold for normal, and you can go by what your specific lab is, but 5 is the most commonly reported. So that's what we looked at. I would say that it's important to try and use the same assay each time you check, so there's some consistency across that. In terms of calprotectin, we evaluated different cutoffs. So we looked 250, 150, and 50 micrograms per gram. And ultimately the guidelines centered on a cutoff of 150 for association with endoscopic disease activity. There may be scenarios where you could use a different cutoff and have good performance, but we wanted to main maintain consistency so that it's easy for the end user or the clinician to remember. And 150 was the cutoff on the UC guidelines and that had the most widely wide applicability for Crohn's disease as well. Now there, the one exception to that, which we'll probably talk about, is post-operative Crohn's disease, where we use the cutoff of 50 micrograms per gram.

Dr Cross:

And so I don't, I, correct me if I'm wrong, so, but I think what you're doing with that 150 as opposed to 250 is you're maximizing your ability to roll out inflammation, understanding that it's pro, it may not be the most specific level.

Dr Ben Cohen:

Correct. I mean, and again, it depends on the scenario we're talking about. So if we're talking about a patient who's in remission, then we want to rule out endoscopic activity and you're going to prioritize a high sensitivity or low false negative rate. Whereas if you're talking about a patient who has active symptoms, then it's a confirmatory test of endoscopic activity and you want to have a lower false positive or specificity.

Dr Cross:

So well said. And you know, when you look at that, we all have that slide, that PowerPoint slide for STRIDE 2 where it talks about the different endpoints. But what I really like about that on the x axis is the timeline of when to expect to see those various outcomes. So what do you think is the optimal time to evaluate for biomarker response?

Dr Ben Cohen:

Yeah, so if we're talking about a patient who has active disease and we put on a treatment, I use that same STRIDE 2 paper and that table and that PowerPoint slide. Whenever I talk about the different drugs in the guidelines itself, we said 2 to 4 months after starting a therapy. And that pretty much fits with the timelines. Obviously different drugs are going to have different onset of action and you may expect an earlier response, for example, with a JAK inhibitor than you may with an anti-integrin, for example. But for the most part it should fall in that 2 to 4 month range.

Dr Cross:

Yeah, so just basically post induction.

Dr Ben Cohen: Yep.

Dr Cross:

Great. So I thought the guidelines were a little confusing based on when, and I'm not blaming you, but when and when not to use endoscopy to confirm inflammation in patients without symptoms with symptoms and based on severity of symptoms. So do you think you can concisely summarize that?

Dr Ben Cohen:

I could definitely summarize it, whether it's going to be considered concise is another story. I think the first thing I would say straight off that is we mentioned throughout the guideline is that all this is shared decision making in terms of implementation of the guidelines and the guidelines don't specifically recommend in favor of or against the biomarker-based strategy over endoscopy, for example. So there's always going to be a role for endoscopy, and endoscopy remains the gold standard. And you know, when a patient or provider wants a greater degree of confidence in the finding, then endoscopy is a very reasonable approach. But it was definitely a challenge for us to formulate the guidelines. And the reason for that is that the performance of the biomarkers depends completely on what your pretest probability is for suspecting endoscopic activity. So, you know, we recommend symptom assessments using the PR O2, assessing abdominal pain and frequency score, and this is going to be the basis for saying whether you have a high probability of having endoscopically active disease or a low probability.

I think the best way to go through this is to consider the 2 different scenarios. So let's think first you have an asymptomatic patient who's in a remission and if they've had an endoscopy in the last 3 years that confirms their endoscopic remission, then the biomarker performance is really strong, particularly with calpro. In that situation, if you have a normal biomarker, we don't recommend a colonoscopy with the normal biomarker. However, if the biomarker is elevated in this patient who's otherwise asymptomatic, had had a previously confirmed endoscopic remission, that's when we would suggest an endoscopy to resolve the discrepancy between the biomarker and the patient's symptoms. But we leave a little wiggle room there. So if the patient doesn't want to do the endoscopy right there to repeat the biomarker in a shorter interval, 3 to 6 months, and if it's still elevated at that time to do the scope, if the patient hasn't had a colonoscopy in the last 3 years that have confirmed their endoscopic remission, then the performance of the biomarkers is not very strong and pretty much in all situations the panel will recommend that they have a colonoscopy rather than relying on biomarker performance for that type of patient.

So the way I think about this simply is I would make sure that my Crohn's patients have a scope every 3 years, probably making sure that they remain in an endoscopic remission and between those scopes, if they're asymptomatic and they have normal biomarkers, I'm comfortable just monitoring the biomarkers along with their symptoms. Now the other scenario is obviously we have a symptomatic patient and this is where the severity of the symptoms can impact the pretest probabilities and the recommendations. And I think this is where some of the confusion comes in. And as I mentioned earlier, we tried to prioritize having a low false positive of the biomarkers in these scenarios. So if you first think about a patient who has mild symptoms, pretty much we recommended endoscopy both in the case of elevated symptoms as well, elevated biomarkers as well as normal biomarkers.

And that's really because there's a low sensitivity and specificity for the biomarkers when there's mild symptoms and a lower pretest probability. The one exception I will say to that is if you have a patient who had a recent treatment adjustment and a partial improvement in their symptoms, but maybe their biomarker didn't normalize, so you start the patient on whatever drug, they're feeling a lot better, but their calpro remains like 250 let's say, then you could consider just tinkering with their dosing of the drug without necessarily doing an endoscopy in that situation because it's a relative improvement in their symptoms that they've had in that time. Now the other scenario for the symptomatic patient is they have severe symptoms, and in this case, biomarkers have a really good specificity for detecting endoscopic activity. So if you have a patient who has severe symptoms with elevated biomarkers, we would recommend undergoing a treatment adjustment.

You don't necessarily need to do the endoscopy at that point. Of course, there could be other reasons to do the endoscopy in terms of prognostication, you know, if you're thinking about surgery and wanting to know if somebody has deep ulceration. So there's always a potential role, but I think I feel comfortable, personally, severe symptoms, elevated biomarkers, that I need to do something different with their treatment and not necessarily wait until I can get them a scope. And then if the biomarkers are normal but the patient has severe symptoms, we would recommend an endoscopyin that situation.

Dr Cross:

You did summarize that pretty concisely because it was a, it was a long question. And I'm glad, I think validates what I do. This just came up today in clinical practice where a patient with Crohn's colitis, we have a good baseline colonoscopy on that patient, was started on a small molecule, he's now postinduction and has a little bit of symptoms and is still on a little bit of prednisone, but the calpro's over 3000. So I didn't feel obligated in that patient that I needed to necessarily look again. Now I think with Crohn's it's more complicated because you could have stricture development and someone who's deeply immunosuppressed, there still is the chance of CMV or colitis dysplasia. So there's all these other factors in our heads when we're making these decisions, but you know, I offered extended induction of the small molecule versus switching. The one rule I have, and I don't know if you agree, I'm not going to make 2 switches without looking, so I may make one switch if you're symptomatic and the biomarkers all align, but a second switch, now I'm really cycling, I'm definitely going to demand that  I take a look or have something even more objective perhaps in the biomarker.

Dr Ben Cohen:

I agree completely and, and we tried to talk about this a little bit in the actual guideline document in terms of what's like a high-stake switch versus a low-stake switch, right? So if you're, like, if it's a matter of optimizing your current therapy, then it's sort of a little bit lower stakes and I feel comfortable doing that without doing an endoscopy. When we're talking about switching drug class, I think it's helpful oftentimes to have the endoscopy to really guide us in potentially what we're going to use. Also, it may be helpful in terms of helping the patient decide whether they're going toward another medication or this is a situation where a surgery may be relevant for them.

Dr Cross

Yeah, I agree. And how, for a patient who's doing well, how often should providers be checking the biomarkers?

Dr Ben Cohen:

So we said every 6 to 12 months. So I think, and so generally speaking for my patients on advanced therapy, we, we see them every 6 months. So I think doing a symptom assessment and biomarker assessment along with their other standard of care labs that you may do for monitoring makes sense at those times.

Dr Cross:

And before we go to the last two questions, I just want to remind the listeners that we are sponsored by the AIBD network and we are now on Apple Podcasts and Spotify. So if you go to Gastroenterology Learning Network and you look for IBD Drive Time, you'll be able to subscribe. So you won't have to wait for the email to hear the podcast. So, I think maybe we talked about this a little bit, maybe the most controversial part of this would be a recommendation for use of biomarkers in post-operative Crohn's. So, can you tell us what those said and is it going to change your practice?

Dr Ben Cohen:

Yeah, so I agree these were definitely provocative and different from the post-op guidelines that were published a few years ago. The first key point I would say about this is that the recommendation is only in the first year after the surgery. So that, I think that's a key point that the data we were able to assess only replies to the first year post-op. The other key take-home point I would say is that C-reactive protein does not perform well in the post-operative setting. So one change I've personally made is I'm not routinely necessarily checking CRP in the post-op setting and then the simple rule for the audience is that calprotectin less than 50 is a strong cutoff for having no endoscopic activity in the first year after surgery in a patient who we would consider low or intermediate risk for endoscopic recurrence.

Now, how do we define risk? That's the bigger question, and we did that using the traditional risk factor assessments for post-op. So, for the audience, low risk for post-op recurrence would be an older patient over the age of 50, a nonsmoker, somebody who's had their first surgery for a short segment of active disease or have fibrostenotic disease. And then a long disease duration before they progress to surgery. And then the high risk factors for occurrence would be younger patient under age 30, a smoker, having had multiple prior surgeries. And the way we classified low, intermediate, high risk. So we said a high-risk patient is somebody who we define as not being on any post-operative prophylaxis with those high risk factors I just said. And, and for those patients, they need to have a colonoscopy 6 to 12 post-op and not rely on any biomarker, uh, monitoring as their post-operative recurrence assessment.

So that's very important. However, any patient who's on post-op prophylaxis, even if they have risk factors or those patients who are low risk, not on prophylaxis, this is the scenario where you could potentially use the calpro monitoring if they have a calpro less than 50 to defer that colonoscopy in the first year after surgery. If it's elevated above 50, then of course they should have a colonoscopy to confirm endoscopic activity and guide any treatment adjustments based on that. And you know, is this provocative? Yes, but I think it's still important because we know from looking at data in like our post-op registry, that there's a lot of patients who don't undergo any monitoring, including a colonoscopy post-op. And it could be that they live far away, they're not near somebody who can provide a colonoscopy or some other reason. So at least we're providing another option for how we can risk stratify a patient, get people to colonoscopy who are going to need it. And again, it's a shared decision making discussion. So even if a patient's on post-op prophylaxis, if it's somebody I'm seeing who has multiple prior surgeries, they've cycled through 3 drug classes already, even though by the guideline we've classified them as an intermediate risk, I'd probably still do a scope on this person in the first year because I think the stakes are much higher for them.

Dr Cross:

Yeah, I think that's reasonable. I mean, what I've been doing with calprotectin based on the POCER study is I've been checking a 3- month fecal calpro and if it's under 100, I would wait until 6 months. Now most of our patients been in clinical practice are going to be that higher risk patient, right? But I think what this would say if the calpro is really low, you know, for that lower risk patient, you could potentially extend that push that colonoscopy out another 6 months and do it in a year. Because we know that people can get anastomotic recurrence, I'm not sure that those 2 as I'm not sure how much they're going to elevate their calpros. And so I think that's reasonable. And as you said, like if they're just going to adopt biomarker, at least they're doing something right. Yeah. Because we know we're not doing enough treat to target endoscopies. So, I think that's reasonable.

So, fun question, Ben, tell the audience something about yourself that they may not already know.

Dr Ben Cohen:

Yeah. This, this is the hardest question— I hate talking about myself—but the one thing I would say is that I am a big live music fan. So, uh, and I actually hang out a lot with my surgeon here, Stefan Holubar, and we go to concerts together. I've seen the fun fact would be that I've seen Dave Matthews band probably like 120 times.

Dr Cross:

Wow.

I heard a rumor that you and Stef were going to go to Vegas to watch U2 in the Sphere, is that correct?

Dr Ben Cohen:

We are going to see U2 in the Sphere, although we're missing the party for that, so I don't know that I should be saying this on the public forum..

Dr Cross:

Okay. All right. Well, sorry about that, but if when you dothat, if you're allowed to take pictures, I'd appreciate a picture from inside. That would be great.

Dr Ben Cohen:

Yeah, I definitely will.

Dr Cross:

Alright, Ben, this is great. Congratulations on, on this work and hope to have you back soon.