All right. Well, thank you for joining me this afternoon. Thank you so much for the opportunity to present. I'm going to be speaking a little bit more in detail regarding some of the targets that we just heard about. I think we've been hearing a lot about treat to target. And so some of the questions that come up is really which targets should we be using in inflammatory bowel disease? And one of my areas of interest, my name is Reena Khanna. I'm a gastroenterologist in London, Canada. And one of my areas of interest are really looking at these targets that we have and trying to understand some of the operating properties of how they translate into clinical trials and to our clinical practices. And so I'll hopefully explain to you where endoscopy fits in, but where all of these targets actually fit in, in the treatment of IBD. So these are the areas that I was hoping to cover over the next 20 minutes. And I hope I'll be able to give you an overview. So this is a slide that we just saw, and you'll see I'm sure many times throughout these presentations, but we know that Crohn's and ulcerative colitis are these chronic inflammatory conditions where there are cycles, repeated cycles of damage and repair. And those repeated cycles lead to deposition of fibrostenotic tissue, ultimately leading to complications such as strictures, fistulas, resulting in surgery. And that cycle repeats over and over. Now, there's been a lot of emphasis in those early treat to target studies, looking at what if we treated earlier before that damage happened, and how do we actually use our endpoints to guide when to escalate therapies to prevent these adverse outcomes. And so there have been many studies that have looked at this. We saw two of the treat to target studies already listed, and there is actually a third, REACT. And so REACT actually demonstrated, this is the first cluster randomization study in inflammatory bowel disease. It's still the largest study conducted in IBD with nearly 2000 patients conducted in the community. And this study compared an early treat to target algorithm versus a standard of care, which was that conventional paradigm. And it actually showed improved outcomes over two years, decreased surgery, hospitalization, major adverse outcomes. However, when we looked at symptomatic remission over those two years, we see there's actually no difference between these two arms. And one of the features that has really come out and highlighted in recent years is that unfortunately symptoms do not correlate with what's happening at the mucosa. And there's been more and more emphasis around how do we better capture what's happening at the mucosa so that our patients can avoid adverse outcomes. Because based on REACT, if we treated based on symptoms alone, we wouldn't be doing our patients a service. Now there are several lines of data that have suggested that healing the mucosa is actually really, really important in Crohn's disease. This is one example which has demonstrated that patients who heal the mucosa are less likely over time to require a resection compared to patients who don't actually heal the mucosa. And so if symptoms don't tell us about mucosal healing, are there other tools that we can actually use instead? And the other thing to highlight is treat to target is pushing us towards this as well, that we need other targets beyond symptoms. And so which targets can we fit in? And I think particularly during the pandemic when many endoscopy units and non-urgent procedures were canceled, we were forced to look at which targets are available to us. And I think as clinicians in practice, so I'm a clinical trialist and a clinical practitioner, sometimes those targets from clinic do apply, from trials apply to my clinic and sometimes they don't. And so understanding the limitations of those targets is vital to make sure we're doing the best job we can for our patients. So what potential targets can we actually use in those treat to target algorithms in Crohn's disease? And I think these are all things we all know. The first thing we can use are symptoms. Now, traditionally in trials, we've used the CDAI. It's never really translated well into our clinics, but we know symptoms in general. I'm speaking to the CDAI here, but symptoms in general we know do not correlate very well with disease activity and inflammatory bowel disease. And these are just some of the many reasons why symptoms are not a good index of what's happening in the mucosa. I'll give you a few examples. This is just one study. In this particular study, looked at CDAI scores in patients with Crohn's disease and patients with IBS. And in this study, and I'm sorry if it doesn't depict as well as I'd like, it shows that that brown bar, which is IBS actually scored on average higher than patients with inflammatory bowel disease. So the score is not, it's a little subjective and it's not specific to the disease we're actually trying to study. There's also many studies that have shown, I'm showing you the CDAI again, but many of the symptomatic scores do not correlate very well with what we see at endoscopy. So you can see the two don't seem to line up at all. It's like a snowstorm of dots. And therefore we know symptoms really don't give us a good picture of what's happening at the mucosal level. The other thing we all know is the CDAI and all symptomatic scores, HBI, PRO exclude, they were never validated in patients that we see every day. So patients with ostomies, patients post-operatively, patients with fistulizing disease were never validated for any of these indices. And so we've turned recently to this PRO2 and PRO2 was one of my fellowship projects. So although I have a soft spot for it, I can tell you that it's still not a validated index. And the PRO2 is derived from CDAI and still has all of the inherent flaws associated with symptom assessment. We also know that there are now so many different definitions for PRO2, and it's hard to know in our clinical practices, which definition we should be applying to our patients. Stride now gives us some guidance around that, but we still know that symptoms are not the whole answer. So we've started turning towards biomarkers. And so how can biomarkers start helping us? And we know that biomarkers have so many different roles. They help us to verify that there is actual disease activity. When you're trying to separate concomitant IBS from IBD, we can use them as a surrogate marker for efficacy of a drug. We can use them as part of our treat to target algorithms, and they can also help with risk prognostication. However, there are limits with our biomarkers. And so let's talk about a few. We've typically used CRP, and we know CRP is important in many different inflammatory conditions. And we know that CRP is not specific to IBD. So it goes up with many different conditions. So how can CRP help us? It can certainly help us in many acute settings. We know many of our indices, like the Oxford scale, many of our scales for acute illness can actually, this can be predictive of what's happening at the mucosa. However, if we're going to be using CRP, we have to take it with certain caveats. We know it can go up with many other different diseases, and we have to understand a little bit about immunology. And so we know in order to produce the CRP, we do know about 20% of patients with Crohn's disease will never produce a CRP. And the reason for that is CRP requires IL-6. IL-6, when activated, stimulates the liver to produce CRP. And up to 20% of patients with Crohn's do not have IL-6 as part of their pathogenic immune pathway, and therefore will not ever produce a CRP irrespective of how unwell they actually are. The other thing to keep in mind, we see in the middle of the screen here, IL-6 figures prominently. And so if we're going to look at a CRP after initiating therapy, it really requires that we understand how the drug works. Now, we just saw a lot about the new molecules that are coming down the pipeline in IBD. And if you have a molecule that directly inhibits IL-6, you will see a very, very rapid decline in CRP. For example, the stikinamab. Now that doesn't tell us about mucosal healing, but it is a marker of target engagement. On the other hand, we have molecules like betalizumab, which work at the tissue level. They don't inhibit IL-6. So we may not see an early drop in CRP, but that doesn't mean it's not working. In that case, CRP might be more reflective of a later healing process. So CRP can be helpful for the different drugs. We just have to understand what it's measuring and what exactly it is telling us, depending on which molecule we're using. Now, fecal calprotectin has really taken on a role in inflammatory bowel disease, partially because it is much more sensitive and specific to what's happening at the bowel. We know this measures neutrophilic activity in the stool, which is a much better reflection of what's happening in the bowel. Of course, there's false positives, there's false negatives that we can get. But overall, we know that this has become very important for us to use in those treat-to-target algorithms. Now there are lots of data that have shown that by such as COM, which we just saw in the last presentation, which shows that when we use biomarkers such as fecal calprotectin in treat-to-target algorithms, it actually leads to better outcomes for our patients. So by treating patients with those biomarkers, as opposed to just standard of care, the authors were actually able to describe greater rates of mucosal healing at the end of 48 weeks. So it does help us make more targeted decisions in our patients. Now, it is important to remember though, the operating properties of these indices. So I won't ask you to look at this whole entire slide. This is a meta-analysis looking at the operating properties of fecal calprotectin in Crohn's disease. We also see a similar meta-analysis done in ulcerative colitis, and this was conducted by one of my friends and colleagues, Dr. Mahmood Mosley. The punchline from both of these is that the operating properties are about 90% sensitive and 90% specific. Now, I can tell you, as someone who's done a fair amount of work in index development, that is phenomenal for a clinical index. However, in my practice, I'm happy to make changes along the road with an isolated fecal calprotectin. So if we're on the treat-to-target pathway and we're going to make some dose adjustments and make some changes, I think fecal calprotectin is a phenomenal tool. In someone who has failed multiple biologics, and we have very few options ahead of us, I'm very reluctant to make a change, a large change, like a change of class of therapy based on a fecal calprotectin alone, because although the operating properties are robust, we would still be wrong approximately 10% of the time. And in my opinion, that's just a little bit too high, particularly in our sickest patients. And so we do still rely a lot on endoscopy. I think this is really in many ways has become one of the benchmarks of how we actually assess the mucosa. And so one of the areas I've worked a fair bit in is looking at the endoscopic indices we have in Crohn's disease to see how they actually perform and to look at their operating properties. Now, up until a few years ago, the indices we use were not truly validated. Now, when you when you talk about the term validated, it's very specific in the index development world. And there's several different criteria and index has to meet before it can actually be considered a validated tool. Now, we've gone back retrospectively and tried to validate these indices, and they are phenomenal in many ways, but do have certain drawbacks. Now, we all know that endoscopy is very important in patients with IBD, and that endoscopic appearance is very important. We know that patients who have very severe lesions on that index colonoscopy are more likely to have poor outcomes in the long haul, including colectomy, and even the development of distillizing disease. So that endoscopic appearance does carry some weight in terms of what happens to that patient in the long haul. We also see that healing the mucosa, the corollary of all of this actually leads to better outcomes. So in the EXTEND trial, which looked at adalimumab for mucosal healing, patients who attained early endoscopic healing by week 12 were much more likely to actually maintain that healing over a year compared to patients who did not have that early mucosal healing. So healing that mucosa seems to be very important for longer term outcomes in our patients. Now, clinically, in clinical trials, there have typically been two indices that have been used, the CDEIS and the SESCD. Now, the CDEIS, which I've depicted here, assesses five segments of the colon, the rectum, the sigmoid, the transverse, the right colon, and the ileum, and there's four different parameters that are assessed, the presence of deep ulceration, superficial ulceration, surface involvement of disease, ulcerated surface, and then you come up with a score by totalling all of that, and then you actually also look at the number of segments that were assessed, and then you look for ulcerated stenosis. So this is a very, very cumbersome scale to use. It actually requires long division, and I have to say I don't think most people use this in their clinical practices because it's tedious, and it doesn't make a lot of intuitive sense. When you add long division to any index, it's really hard for a lot of people to imagine the difference between a 40 and a 38. What is that clinically relevant or not? So in order to simplify the CDEIS, there were several studies that looked at that, simplifying it, and that is how the SESCD was born, and so the SESCD is very similar. You look at those same five colonic segments, and there are four items to look at. Ulcers, the percentage ulcerated surface area, the affected surface, and stenosis, and that comes to a simple sum. So you take those five segments, four, so the 20 values you have, and you get a simple sum, and so it is much simpler and has really become the index that is used most commonly in clinical trials. So one of the areas that I've done a fair amount of work in is actually looking at the operating properties of these indices, and I can tell you they perform very well. So two very fundamental properties are intra- and inter-rater reliability, so how well people agree with each other and how well people would agree with themselves if they graded the same colon over and over and over, and so these scores are considered remarkably high, almost perfect by index development standards. Now what is the issue with the SESCD and the CDEIS? They are developed as continuous measures, meaning they are a scale that goes across a range. In clinical practice, in clinical trials, we dichotomize these indices. We say, are patients eligible for the trial or not? Are they in remission or not? And when you do that, these operating properties drop almost in half, and so the indices are actually not very robust when we dichotomize them the way we use them in our clinical practices, and this is why we often see in a lot of different guidelines people talk about absence of ulceration, because that seems to be something that's easy for people to agree on and actually seems to carry some weight in terms of long-term prognostication. And so although we think it's really important to grade colons and have common language between all of us when we perform endoscopy, it's also really important to understand some of these limitations of the endoscopic indices. Now there are areas where people tend to disagree on how to score the SESCD, and these are the four that seem to come out as being the most disagreed-on lesions, and so there would be lesions between segments or lesions at the anastomosis, and the biggest source of disagreement is which segment do you apply it to? So if you saw something between the…at the ileocecal valve, do you count it towards ileum or do you count it towards cecum? Now it's a small difference, but it actually can be very important, because you would…if you assign it to the termina ileum, you're actually assigning an entire new segment, which elevates the score substantially. The other area of disagreement are anal lesions, and this is because the…the anus typically has some redness and is typically a little bit narrowed, and so we know that patient… it's often not viewed quite as well. Superficial ulcers, people tend to disagree on depth of ulcer and stenosis, and part of…it seems to be another area that people disagree on the most, and so there are standardized scoring conventions that can help eliminate some of this disagreement, but if you are going to implement standardized scoring at your center, it's probably really important that you standardize across endoscopists, particularly for these items, because they will lead to the greatest discrepancy in reporting of endoscopic lesions. Now there's been a lot more interest, given that Crohn's is a transmural disease, looking at transmural healing. Endoscopy will really only give us a glimpse of the mucosa, and shouldn't we be looking for healing deeper in the bowel? And so this is something that has really taken off and a lot of people have looked…looked into, and so we do see several studies looking at agreement between endoscopy and MRE, and they do seem to agree fairly well. I think what we do realize is there are standardized scoring indices looking at MR enterography, but the exact definitions for remission, for what's considered mild, moderate, and severe are still being worked out, and so I think this is something that's probably very important and may take off in the future, but I think there's still a lot of questions on how this applies in clinical practice. And if you practice in a jurisdiction like me, in Canada, where we do have wait times, it's not always very accessible to have something like MRE, and therefore in a lot of jurisdictions we're starting to see ultrasound being something people look at, where gastroenterologists are actually looking at the bowel wall to look for edema to make real-time changes, and so I don't think that's been adopted everywhere around the world, but it's certainly something that's up and coming and can give us that picture of transmural healing. I do think this is…transmural healing is an area we're going to see expand in the future. Now, I've mentioned already some of the newer things, and I think the other area that's really coming to the forefront is histology. We're seeing this in the clinical trials, so now many molecules are including histology, although the most recent phase 3 studies are looking at histology, and I think this is an area that regulatory bodies have really taken note of, and for mucosal healing, they also want to see some of that healing at the tissue level. In Crohn's disease, there's a lot of unanswered questions. Which index should we be using? Where do we biopsy? How many places do you biopsy? Do we use the worst score, so the biopsy with the most inflammation? Do we take some sort of average or aggregate? And so, I think this is an area of great interest, and intuitively, it feels like if we heal deeper to the tissue level, it should be better for our patients, but I think for primetime use in our clinics, there's still a lot of questions and probably more questions than answers at this point. Now, I'm going to summarize by saying this is where we are. I think we do still use a lot of symptoms because symptoms are so important to our patients. I think for endoscopy, people have really, in trials, are using this SESCD score. Absence of ulceration is another score that is another very simple tool that I think a lot of people still rely on. It's important to note that a lot of the definitions of the SESCD are still subjective, that what constitutes remission, what constitutes mild, moderate, severe, are not truly grounded in validation science, and I think imaging and histology are up and coming but are not quite ready for primetime. Now, I'm going to turn very quickly to ulcerative colitis, and I think there's a lot…we've covered a lot of ground in Crohn's, and so I think this will be a much shorter section to discuss. I think the main things that we've seen in ulcerative colitis are really symptoms, endoscopy, and, of course, histology has taken on more of a role in the ulcerative colitis space, and so I think in terms of symptoms, we know of all the symptoms we use in ulcerative colitis, symptoms do track better with what's happening at the mucosa than it does in Crohn's disease, and the two symptoms…the symptom that seems to have the greatest correlation with what's happening at the mucosa is rectal bleeding, and that's why I think STRIDE really pushes us towards looking at rectal bleeding and stool frequency. Those are things that patients care about, and it actually helps us as physicians gauge what's happening at the mucosa. Now, in terms of endoscopy, I think the Mayo score is something we're all very familiar with. I think every endoscopy unit is mandated, it seems, to have a picture of the Mayo score. We see it everywhere, and it's very intuitive. We see the Mayo 0, which is a normal colon, Mayo 1, the hallmark features being loss of vascular patterns and erythema, Mayo 2, the hallmark feature really being friability. We sometimes see absence…we see absence of those vascular markings and possibly some erosions, and then a Mayo 3 with that hallmark being that severe or spontaneous bleeding as well as frank ulcerations. Now, these are very intuitive and very easy to score, and I think many places will report this in their reports. There are things to keep in mind, though. Scarring, pseudopolyps, mucus do not contribute to a Mayo score. Now, those are things that will not likely heal with therapy and so are not included in the active inflammation scores. It's also really important to realize the ulcers on top of a pseudopolyp are considered ischemic and not inflammatory and do not contribute to a Mayo score. So, there are standardized rules even for reporting a Mayo, and that can lead to differences in how we all kind of talk about things. I will say I do still think the Mayo score itself is important, but so is the description. And one of the limitations of Mayo is that we look at ulceration in itself is considered a Mayo 3, and so if you have 90 percent ulceration of the rectum and then you reduce that down to one ulcer that is 0.6 centimeters, I think most of us clinically would think that was a clinical victory. However, according to the Mayo, both of those would still be scored a Mayo 3. And so, I think the description is as important as reporting an actual Mayo score because there are some limitations, even though we know it's fairly robust. Now, there's lots of data that tells us that Mayo is important and actually, again, tracks the really important outcomes in our patients, including absence of colectomy or avoiding a colectomy over time. Now, where does histology fit in? Histology is really gaining a lot of momentum, and so there's lots of studies recently that have shown that you can have an absolutely normal colon by several different indices, Mayo, Barron, or just what's considered a normal endoscopy, and still up to 40 percent of those patients will actually have histologic activity. And so, it seems that what's happening at the mucosa doesn't always track with what we're seeing histologically, and so people have really wondered, is histologic remission even more important than what we're actually seeing with our eye? And so, there are lots of emerging data suggesting that histologic remission may actually improve outcomes, such as hospitalization and colectomy, things that are very important to our patients themselves, and we also know that histology can be reliable. So, when we look at those inter-rater reliabilities, it can actually be an important tool as well. Now, in terms of histology, I think there are still some things to discuss. There's lots of indices you'll hear about. There's the GABO score and the RHI and the NANSI, and so which index, where to biopsy, how many biopsies, and I think the most important thing is getting commonality in reporting from pathologists is still something that remains to be done in the IBD space. And so, where are we with ulcerative colitis? I think in trials, we're still using the Mayo score and the modified endoscopic score of the Mayo…sorry, the Mayo endoscopic score, and so in clinic, I think STRIDE and a lot of people will look to this rectal bleeding stool frequency endoscopy using that Mayo with a description, I think is still really important, and I really do think histology is going to probably take off in the next few years. I think this is really where a lot of the science is pushing, and this is where the trials are actually leading us in our clinics by helping us to define what definitions lead to important clinical outcomes. So, I think outcomes is actually a really interesting area. It's kind of interesting in the IBD space that we don't truly have a gold standard, and I think in the next few years, all of the phase 3 studies and all of the ongoing investigator-initiated studies are going to give us a better picture of how we can best serve our patients. So, I guess I might be one of the few people who are actually passionate about…I think really passionate about outcome measure validation, and so thank you very much for indulging me with this opportunity to speak with you today. I really am grateful for the opportunity to have been able to join you today, and I thank Beau for the invitation. So, thank you very much.

inflammatory bowel disease

treat-to-target strategies

mucosal healing

biomarkers

endoscopy

validation