Maria Abreu, MD, on Disease Management Today and Tomorrow

Dr Abreu reviews how the field of disease assessment in inflammatory bowel disease is advancing and what new tools and techniques may be ahead.

Maria T. Abreu, MD, is director of the Crohn's and Colitis Center, professor of medicine, and professor of microbiology and immunology at the University of Miami Miller School of Medicine, in Miami, Florida.

TRANSCRIPT:

My name is Maria Abreu. I'm at the University of Miami, and I'm a gastroenterologist focusing on IBD. And I've had the opportunity to speak in a session entitled “Looking Into The Crystal Ball,” and the topic I was given was to discuss disease assessment today and tomorrow. And I tried to highlight that, of course, right now, you know, you can't give an IBD talk unless you show the STRIDE guidelines, which would show you the metrics of how one needs to have a plan for each patient, holding oneself accountable to making sure people have improved clinically and, have an improvement in symptoms, that that can happen in the short term.

But over the intermediate and longer term, that there should be deeper levels of improvement, and those include, of course, biochemical improvements and this thing that we call mucosal healing, and that mucosal healing, if you try to follow the field and play trivial pursuit, you'd make yourself crazy because that bar keeps changing. You know, what do we call mucosal healing?

For all intents and purposes, more recently, we think about it as having endoscopic perfection and also, histologic perfection. Even that's a pretty high bar. But some of the things that I've discussed in this particular conversation is that now we have intestinal ultrasound, so there's a deeper level at which we're assessing whether or not there has been improvement. Because as it turns out, even in ulcerative colitis, which we think of as a mucosal disease, if you're looking by ultrasound, you have hyperemia, you have thickening of even the muscle wall. And so, therefore, it stands to reason that for the person to have really full improvement, all those things need to get better. Imagine the dysmotility that might occur if one actually has not only the mucosa involved, but also these deeper layers of the intestine involved. Think about pain signaling and all these other things that our patients tell us about that we say, well, I don't know. Usually, you know, we don't treat anyone with things for pain in ulcerative colitis, as an example.

Then there are other, I think, really interesting things that we need to contemplate in terms of disease assessment. There are people that have clearly shown that the that people's heart rate and the amount of how they how they move and whether they're more sluggish or not based on Fitbit data and other kind of fancier things—these things might be telling us that someone is going to be having a flare. I was listening to something on television and service dogs, of course, can pick up whether someone has low blood glucose or, you know, have has other things. So there's obviously something intangible that that the body knows that something is amiss. And could that be a way to gauge that someone is heading into a patch of a flare? There are people who are developing fancy toilets that can sample stool, and maybe that's another thing that we will be using for disease assessments since at least American patients hate to collect stool for fecal calprotectin. Others have developed techniques that people can take a picture of their poop, and that apply some AI algorithm if you do that a few times to know, you know, really, is there blood, is it loose, and all these different characteristics. So I think that we need to keep a kind of an open mind about what we mean with respect to disease assessment.

People are also developing, you know, we talk about histologic score and all of these scoring systems, whether it's endoscopy, which, my friend Ryan Stidham did just a beautiful job discussing endoscopy and applications of AI. But there's also the pathology piece, right? We depend highly on very skilled pathologists in in referral centers with experience with IBD to give us give us high quality reads and tell us whether that person is has mild, moderate, or severe inflammation. Is it active, i.e., neutrophils, or is this chronic? And tell us about that specimen. Of course, in clinical trials, we have fancier scoring systems—Geboes, the Nancy scoring system, that now are being used to quantify how much inflammation there is so that one can say, yes, someone responded or didn't respond from a histologic standpoint.

But what if a computer could look at all these slides, and you could say to it, this is a person who got better. This is a person who had a little bit of improvement. This is a person who had no improvement at all. And let the computer tell you, okay, this is what is going on, and rate it. Marietta Iacucci, who is now at the University College Cork has really worked on developing AI for pathology. She's an advanced endoscopist herself, but, to look at these slides, and to have computer algorithms that will rate the inflammation. And so I think that I'd like to believe that our human pathologists are quite good, but in point of fact, especially when you think about in many places, this could be the answer to some things, which is to have to have AI-assisted pathology for IBD.

And then finally, the other the other kind of cool stuff that I shared is that investigators in Germany have used confocal endomicroscopy to again challenge the notion about how deep is deep remission. Right? And in their study, with confocal endomicroscopy, the patient receives injection of fluorescein, you know, IV fluorescein like you would for looking at the retina. But it's to highlight all the blood vessels. They can look then at whether that fluorescein is leaking in between epithelial cells as a marker of epithelial barrier dysfunction in states of inflammation. Of course, that barrier is leaky, even when it looks histologically normal, however, or even endoscopically normal, there could still be leaky barrier, and they can assess this with confocal endomicroscopy. And they found that those people that had no leaking barrier based on this endomicroscopy, those are the ones that were the most likely to be really have no flares and do clinically much better than even those patients that had endoscopic remission, suggesting that we can have functional ways to assess whether someone is truly in in the deepest state of remission. And, of course, we need our drugs to kind of come along for that ride to allow us to accomplish what we believe to be the highest levels of improvement, which include fixing this perturbed barrier function and, reducing the inflammation at both histologic and functional level.

So stay tuned, because I think there will be much more than only doing endoscopy that will define our best outcomes.