IBD Drive Time: Parambir Dulai, MD, on Best of ECCO 2022

In this episode of IBD Drive Time, Parambir Dulai, MD, joins our host, Dr Ray Cross, to discuss the best of ECCO 2022 in diagnostics, natural history, and epidemiology.
 

Parambir Dulai, MD, is an associate professor of medicine in the Gastroenterology and Hepatology Department at Feinberg School of Medicine at Northwestern University, in Evanston, Illinois.

TRANSCRIPT:

Welcome, everyone, to IBD Drive Time. I'm Raymond Cross from the University of Maryland and I'm delighted to have my friend and colleague, Parambir Dulai, from Northwestern here as our guest today. Parambir is going to present the best of ECCO 2022 in diagnostics, natural history and epidemiology. Parambir covered this for the IBD Education Group of which MedEd Consultants is a parent company. And we're going to go over the three best abstracts. Parambir, welcome to IBD Drive Time.

Dr Dulai:  Yeah. Thank you so much, Ray, for having me. It's a pleasure to be talking about these 3 abstracts. So the first one I'm going to go over is antibiotics as a risk factor for older-onset IBD. And we've been seeing an increase in prevalence and incidents of new-onset new diagnosis of IBD after the age of 60. And it's thought to be largely related to environmental exposure. So I think these authors did a really good job of trying to link together whether antibiotic exposure and risk of IBD is something to consider for some of these older patients. So they used the Denmark Nationwide Registry, which is a population registry, and they identified people who were 60 years of age or older from 2000 to 2018 and looked at the associations between antibiotic exposure, number of antibiotic exposures and risk of developing a new diagnosis of IBD after adjusting for age, gender and the time period.

What was most striking was they identified over 2 million people with over 22 million years of person-years of covered lives. And they saw a very clear and consistent association with a 1.6- to 2.4-fold increased risk for IBD with any antibiotic exposure in these patients. What was really nice about this study was that they were able to actually separate out the number of antibiotic prescriptions and the type of antibiotic prescribed and they saw a linear, incremental effect where the more the courses you got — if you had 1 course, you had about a 1.3-fold increase risk, 3 courses about 1.7, and 5 courses of antibiotics was a 2.4-fold increased risk for developing IBD. And this was most significant with fluoroquinolones, but all antibiotic exposures were associated with an increased risk. They also showed that this was most predominant if the antibiotics came within about a year before the diagnosis, showing some causality in terms of time and effect. So this was a really nice study really linking together antibiotic exposure and older onset IBD development.

Dr Cross: And you don't think the antibiotics, Parambir, was GI symptoms and they're being treated for GI symptoms and that's why there's an association? It doesn't sound like it.

Dr Dulai: No, and it was a great point. And they made a pre-emptive statement in their abstract to show that it was for both GI and non-GI antibiotics. So even antibiotics that are often never prescribed for below-the-diaphragm infections were associated with an increased risk for IBD. Notably though, fluoroquinolones were the most significant, and in this population, they were able to look at different indications and you can imagine fluoroquinolones for upper respiratory pneumonia were one of the dominant indications.

Dr Cross: So really interesting. So I don't think the message here for clinicians though, like a patient from last week who saw an integrative medicine specialist and she has H. pylori and they told her not to take antibiotics because they're bad for your microbiome. I don't think that's the message you're trying to send here, correct?

Dr Dulai: No, I think the point here is to be a bit more thoughtful and judicious with antibiotic prescriptions in some of these older patients. And I suspect the patient population that it's most relevant in is some of these aspiration pneumonias that we sometimes see where it's questionable if they really need antibiotics, and some other indications where sometimes antibiotics get a little bit overprescribed in older patients because we're not quite sure what's really going on.

Dr Cross:  Or hopefully my mom's not listening to this, but when she wants antibiotics for a common variety cold, that's not antibiotic stewardship.

Dr Dulai:  No. Yeah, exactly.

Dr Cross: All right. Terrific. Let's go to the second abstract.

Dr Dulai: Yeah. So the second abstract switches gears a little bit looking at diagnostics. So this is digital oral presentation 11, “Early intestinal ultrasound predicts endoscopic response to anti-inflammatory treatment,” and shows drug-specific response to biologicals and tofacitinib in ulcerative colitis. So this is pretty cool because I think ultrasound is not really well adopted in the United States, but it's growing in adoption across the world. So what they did in this longitudinal prospective study of about 50 ulcerative colitis patients is they looked at ultrasound measurements in relationship to clinical scores, biochemical parameters, and baseline and follow-up endoscopy in people who are largely starting infliximab, vedolizumab, or tofacitinib. And what was nice here was that they showed that there was a very clear association between ultrasound measurements, the bowel wall thickness, and endoscopic measures of disease activity or the endoscopic Mayo score, a very nice linear association between the two demonstrating an ability to identify disease inflammation.

And then they show this really nice linear trend in reduction in bowel wall thickness for on ultrasound with treatment. What was cool here was that they actually showed some differential responsiveness of ultrasound based on treatment with infliximab or tofacitinib, which we sometimes think of as a little bit maybe more rapid-acting compared to vedolizumab at week 2 and week 6. What was, I think, most helpful for clinicians that's going to come out of this is that they identify bowel wall thickness of between over 3 to 3.5 millimeters to have an accuracy of over 95% for predicting the presence or absence of endoscopic inflammation. So they provide some actionable measurements that you can use on these ultrasound measurements to actually define whether somebody has active disease in the clinic and be able to make treatment decisions on the spot as opposed to what we're currently doing, which is follow up endoscopy every 4 to 6 months.

Dr Cross: So yeah, there's a lot of research now looking at ultrasound for both ulcerative colitis and Crohn's, and I know that Bincy Abraham at Houston is doing some of this and Marla Dubinsky for pediatrics at Mount Sinai. So assuming that this gains momentum in the United States, how do you think that we would use this? Would you have it completely replace a sigmoidoscopy or colonoscopy? Or would you get a sneak-peek postinduction around week 8 to get a sense of what's going on? How do you envision this will be utilized if it is utilized at all?

Dr Dulai:  Yeah, I don't think it's going to replace that baseline endoscopy or that confirmatory endoscopy at 6 or 12 months because there still is some misclassification with this. But I think where it's probably most helpful is that 2- to 8-week timeframe postinduction or postinitiation, where you can get a sense of is the thickness actually changing? And do I need to be worried about this patient over the next coming weeks? Because we can't keep doing lower endoscopy every couple weeks just to see how they're doing, but we all have those patients that are a bit higher risk, we're a little worried about them. We want to make sure they're headed in the right direction. With a marker, that's a bit more accurate than with CRP or fecal calprotectin. And for them to show such a dramatic change in the first 2 weeks and responsiveness to change I think is a really important aspect of this for point of care testing.

Dr Cross: Yeah. And particularly my understanding, Parambir, correct me if I'm wrong, is that you can get this information with a handheld ultrasound. So if we can figure out in the U.S. how to make that reimbursable, get an RVU or something to be able to do that and interpret it very quick in clinic, then you could imagine there could be training programs where people like you and I could do this in practice.

Dr Dulai: Yeah, it is. You can do this on a handheld ultrasound. I mean hospitalists use handheld ultrasounds all the time. I don't see a reason why we couldn't do this. And with drugs like tofacitinib, where you can go up to 10 milligrams without too much difficulty because it's priced the same, you see somebody, you look for their thickness and you can dose escalate if you need to, or at least get their drug level and pre-emptively think about dose escalating biologics.

Dr Cross: Excellent. So before we go to the final abstract, I just want to remind our listeners that IBD Drive Time is sponsored by the Gastroenterology Learning Network and Advances in IBD.

So Parambir, why don't we talk about the final abstract?

Dr Dulai: Yeah. So the last abstract is the first virtual chromoendoscopy artificial intelligence system to detect endoscopic and histologic remission in ulcerative colitis. So there's a lot of growing interest in artificial intelligence as it applies to endoscopy and histology, really stemming off of the growth in adenoma detection rate, some of these new platforms that are available in a natural opportunity for that is to apply it in IBD because we're using endoscopy so much.

So this was a study where they took the PICaSSO, which is the Paddington International Virtual ChromoendoScopy ScOre, and looked at whether this PICaSSO system when done virtually on endoscopic and images could identify endoscopic inflammation and histologic inflammation when done with chromoendoscopy. So they took 496 endoscopic videos from 235 patients that were broken up into a training set of about 250 videos and a validation set of about 60 videos. And then they used another 153 videos separately as a test cohort.

They looked at one, the ability of this PICaSSO system to quantify and classify endoscopic activity using the UCEIS, the Ulcerative Colitis Endoscopic Index of Severity, and then separately, whether it could use the virtual chromoendoscopy part of it to actually predict histologic remission and the histologic activity. What was really striking here when you look at the results was that in the validation cohort, the system very accurately predicted the presence or absence of endoscopic remission with a specificity of 94% and an accuracy of 92%. When they used the virtual chromoendoscopy part of it, not just the white light part of it, it really didn't add much in terms of incremental sensitivity and accuracy. So it looks like you can apply this on a white light system. And what was also important was that they were able to predict histologic remission with an accuracy of 92% and specificity of 85%.

And again, the virtual chromoendoscopy part of it didn't really add too much in terms of accuracy or specificity. And so this I think brings forward the ability hopefully to be able to do virtual endoscopic and histologic assessments real time to guide quantification and provide real time clarification both endoscopic and histologic remission in our UC patients.

Dr Cross: And potentially, this could be used in lieu of essential reader for clinical trials and so forth, right?

Dr Dulai: Yeah. So I think the practical applications are 1, optimizing efficiency for clinical trial recruitment both when you have somebody you're screening, but also potentially banked videos for somebody because sometimes we get these patients where we know they have active inflammation, their endoscopy was done a week ago, but we can't use those videos to necessarily qualify them for the clinical trial. So if you have that endoscopic recording done, it's much easier to potentially qualify them or classify them. And 2, there's still a lot of uncertainty or unknown in terms of what the community practice level variability is in actually classifying disease activity, and whether this is another opportunity to optimize that, just like applications for artificial intelligence to optimize abnormal detection rates.

Dr Cross: Yeah. And I was thinking that when we talk about endoscopic, I guess we're using endoscopic improvement for a Mayo 0 or 1e, endoscopic remission for a Mayo of 0. And we have some data at our institution that typically we're happy with a 0 or 1. Although we found that the Mayo 1 do have a higher relapse rate than the 0s, and maybe histology might be a tiebreaker there to try to predict a Mayo 1 that's a high-risk and low-risk. And I wonder if a system like this may be able to tease that out for us to identify a patient that you might monitor a little bit more tightly after that good sigmoidoscopy or colonoscopy. It's all speculative.

Dr Dulai: Yeah. No, I mean I actually completely agree. I think when we do these endoscopes and we see a little bit of something in the rectum, it's not clear if it's really the ulcerative colitis or the PrEP effect, and having an ability to know real-time that the histology is actually normal would help us understand you really don't need to change much here. You can let them go with the diagnosis of endoscopic remission and not make an adjustment.

Dr Cross: I'd like to see a technology like this used in some of our patients that have those fields of pseudopolyps. And for community providers, it can be really unsettling because you feel like you can miss something. But something like this may help reassure providers that they're not missing an adenomatous lesion in that field of pseudopolyps.

Dr Dulai: Yeah. Yeah, and I that's appropriate because I think that's really where some of the technology becomes really cool and probably where chromoendoscopy and that virtual chromoendoscopy aspect might be more incremental, which we didn't see it here, but for differentiating adenomas, pseudopolyps, inflammatory pseudopolyps, I think that's where that would be really cool to see how it applies.

Dr Cross: All right, Parambir, time for the fun question. So tell us something about yourself that the audience may not know.

Dr Dulai: Yeah. So something fun about myself. In high school, I thought I wanted to become a professional boxer. So I actually trained a little bit. I wasn't very good so I decided to go to med school instead. So I actually appreciate you not beating me up too hard today with some of these questions.

Dr Cross: Wow, that's very, very cool. It's not as cool as Ben Click having green hair and being a skate rat, but it ranks highly, Parambir. Parambir, this has been great. I learned a lot. I'm sure that the listeners learned a lot. I wanted to thank you for joining us and hopefully we can have you back again soon.

Dr Dulai: Yeah. Thank you so much for having me, this was a lot of fun.