David Rubin, MD, discusses how to tailor treatment for inflammatory bowel disease patients.
DR. DAVID RUBEN: Hello, my name is Dr. David Rubin and I'm a professor of medicine at the University of Chicago Medicine. Today we're going to be speaking about tailoring treatment for the inflammatory bowel disease patient-- getting the drug and dose right. And as you can see, this is really a moving target. What we've come to appreciate in the world of inflammatory bowel disease is the importance of individualizing our assessment of a patient and choosing the right therapy for that individual. The reason this is so important is because there is an acknowledged heterogeneity of inflammatory bowel disease. We now recognize that despite our classification system of Crohn's disease and ulcerative colitis, we've come to realize that in fact, there are probably 50 or 100 different types of diseases all overlapping. Some of which have similar clinical characteristics. In addition, we recognize that there's a large primary non-response to therapy, a large secondary loss of response even after a patient responds to therapy, and the disease changes over time. This is most evident in Crohn's disease where we've acknowledged that many patients have progression from inflammatory to fibrous stenotic phenotypes, but is increasingly appreciated in ulcerative colitis as well. And lastly, and very importantly, our patients change over time. There are a variety of characteristics related to our patients that change and will certainly affect our choice of therapies, so it's easy to say one size does not fit all when we think about the complexity of managing IBD. But what makes it more challenging is understanding how we can actually use this in clinical practice. So where would we like to be? Our optimal use of therapy for inflammatory bowel disease should include using therapies at the right time, meaning not too early and not too late. We recognize that, in general, therapies work better when we use them earlier in the disease course, but if we use them too early we might be exposing patients to therapies they may not need or risks that they shouldn't need to be exposed to. In addition, if we use therapies too late we may not have the response that we wish. In addition, understanding prognosis has become critical to the management of these patients. We also would like to use the right dose. More and more we appreciate the complexity of pharmacokinetics and pharmacodynamics in our patients. We'd like to use just the right amount of therapy. Not too little and not too late. And of course, understand how much we should be using and change that over time depending on how the patients need to use it. We want to use the right interval. In other words, we'd like to dose properly. This is particularly the characteristics of our biological therapies. We want to choose an interval that makes sense for stable maintenance of remission over time, and we'd like to emphasize to our patients that remission means no breakthrough between doses of their treatments. We'd like to use the therapies for the right duration. This is increasingly an important topic and one that I'll return to, but the concept of using therapies forever is really not something that we talk about anymore. We want to use our therapies when we need them, and we want to appreciate when, if possible, we can de-escalate or even withdraw certain therapies in specific patients. And of course, we want to balance efficacy and safety. Understanding the changing risk-to-benefit ratio among our patients is a key issue in discussing therapies with patients. In fact, this changes when the patient responds to therapy and is now feeling well, compared to how they might be feeling about risk and benefit when they're sick and you're choosing your initial treatments. And of course, we like to balance the right cost. We want to make sure that we're doing this in a cost effective manner and thinking about not just cost to our individual patients, but also costs to society as we continue to use more and more expensive therapies. So why haven't we reached the ideal treatment use in our inflammatory bowel disease population? Well first, our existing classification system isn't specific enough to direct therapy. Another way to say that is that despite calling things Crohn's disease and ulcerative colitis, we're not granular enough in the way we describe these diseases to choose therapies differently based on some of the general features of these diseases. There are some exceptions that I'll describe in a moment. Secondly, our goals for managing patients are not correct. They are largely symptom based, or they have been historically, and now that we finally have started moving towards more objective measures of disease control, and very importantly have treatments that can achieve those objective measures, we can now start moving towards chronic care and chronic management. Which includes keeping people healthy over the long term and avoiding complications. So we can move from crisis management to chronic care. The third problem is that our therapies just don't work. Remember that we're treating inflammatory bowel disease not because we understand the cause-- unfortunately that's not the case-- but rather because we've come to appreciate the downstream effects of whatever processes is going on that's driving these diseases. So in other words, we're choosing targets that are downstream in inflammation. And by choosing those targets and using therapies against those targets, we're not actually getting at the underlying source of the problem. Therefore, many of our therapies may not work, or after they've worked for a while for a variety of different reasons, our patients may lose response. And lastly, there is a well described disconnect between the patient and the health care provider. In other words, the health care provider may have specific goals for the patient, and the patient may have specific goals, and they may not be the same. In addition, despite describing to patients what our goals are related to remission, stable remission, sustained steroid-free remission, patients often don't quite understand what that means and may settle for feeling better but not achieving complete remission. And this is a very important problem that we face and one that we have to do a better job describing and explaining to our patients. So we have made great progress but not quite enough yet. So if you think about our clinical observations and our treatments, and break it down into our standard classification of regional enteritis, or Crohn's disease and ulcerative colitis, and our treatment strategies over time that have gone from just trying certain therapies to more evidence based, to targeted therapies over time, you can appreciate some of the changes that have occurred. Across the clinical observations, you can appreciate that we've gone from just the early days of clinical descriptions and careful observations that led to the classification system as we currently have it, to the explosion of genetic information that has emerged in the last decade, to some immunologic phenotyping related to serial types and immunotypes. And the current emphasis on understanding our environment, which includes the microbiota of the guts. On the treatment strategy front we've gone from desperation-- when people were dying from these diseases and we were trying anything to try and save their lives-- to the early phase of evidence based therapy that has emerged, to today's modern management using phase III clinical trials. To some of the disease modifying strategies that we're currently employing and moving towards what's called treat to target, which has to do with achieving a specific target of control. And I'll come back to that in a few minutes. So when we classify our diseases we think about identifying sub types that would actually change our treatment options. And when we think about our treatments overall, we want to identify effective, sustainable, safe therapies, and even cures for some patients. And although we're making progress, we're not quite there yet. And until we define what these goals are, we're not going to be able to achieve them. So what are the factors that influence disease control? Well, there's a variety, and we can break them down into patient factors, therapy related factors, and disease factors. If we start with the disease related factors, we can knowledge the lag time before a patient is diagnosed. This is most typical in Crohn's disease where patients may have symptoms or even sub clinical inflammation for a long time before they actually come to medical attention, develop symptoms, or have a diagnosis. Unfortunately, this puts us behind the game when we want to try to treat these patients. Secondly has to do with the general phenotype, the genotype, and then more recently, the microbiome, the immunological characteristics of the disease, and even other such things as elevating a CRP, which doesn't occur in all patients but is a very specific disease factor that can influence and help us choose therapy. The second would be to focus on our therapy related factors such as, what has the patient had in terms of prior resections, previous failed therapies, are they steroid responsive and have they been exposed to steroids previously? Is there a dose response to certain therapies? What's the half life of the specific therapy the patient's being exposed to, and what are the delivery mechanisms? When we start to factor in all the different characteristics of our different therapies, especially as we come to understand the pharmacodynamics of our biological therapies, we can appreciate that there are some very specific distinctions between therapies that affect our patients. And we'll come back to that in a moment. Our patient factors are related to the patients. How big are they in terms of their BMI? Their gender actually influences their likelihood of responding, and maybe their outcomes as well. Their nutritional status is quite important. Pharmacogenomics related to how they will respond or not respond to specific therapies. And also whether or not they're at higher risk for side effects with therapies, and whether or not they're taking their therapies as prescribed. And a very important factor in the inflammatory bowel disease world is whether or not they are a smoker, a lifelong non-smoker, or an ex-smoker. So our treatments have been aimed primarily at observations and theories, and unfortunately, as I mentioned, not the cause of the disease. We have, for the most part, focused on immune modification. The breakthrough that steroids could treat IBD certainly saved many lives, but it's had the downside of many side effects and not being a sustainable or effective maintenance strategy. Thiopurines and methotrexate are another option, and, of course, anti-TNF biological therapies have been the most significant influence and greatest advance in our therapies in the recent years. Now, there have been other therapies developed and are in development for inhibition of other cytokines, and recently the development of leukocyte trafficking in inhibitor therapies. There's been great interest as well in microbiota manipulation-- recognizing that the gut flora are certainly modified in some patients. As well as some patients describe the onset of their inflammatory bowel disease after they've had an infection, or after they've been on antibiotics. We recognize that perhaps another treatment strategy would be to modify the microbiota as a way to control the disease. Unfortunately, the evidence for this has been limited and not as successful as we'd like so far. But this includes some evidence for antibiotics, especially in Crohn's disease. Prebiotics and probiotics have limited usefulness but have been studied. There's great research going on right now looking at fecal transplantation, but so far without any specific results that can guide us further, and a variety of other things. But of most interest to our patients has been this concept of using diet to change their disease, and the concept that diet, in fact, we know changes the microbiota. But also because patients so strongly believe that their diet must somehow be contributing to this disease since when they eat they have symptoms, and the diet in their food intake certainly can be affecting their bowel. Unfortunately, we haven't made great breakthroughs in the diet area, but there is ongoing research that's of great interest. Another major treatment strategy, of course, is surgery, and many of our patients still require surgery for their diseases. Whether it's resection of fiber stenosis in Crohn's disease, or resection when there's fulminant disease that's not responding to medical therapy, surgery remains one of the most effective strategies we currently still have for inflammatory bowel disease. So why do patients with IBD, and whether it's Crohn's disease or ulcerative colitis, not respond to their medications? Well first, there's the primary non-responders. The drug is wrong, the mechanism may be wrong, or the diagnosis may be actually wrong. So perhaps we're treating an overlap condition, or we're calling it Crohn's disease or ulcerative colitis when in fact it's related to a completely different condition altogether. Another problem is that we often are under dosing, and that's a big challenge that we face. And then there's rationale to think that there's the wrong delivery system. Whether this has to do with oral 5-ASA or rectal therapies, or whether we're talking about IV or injectable therapies for a biological therapies, there may be some rationale that's reasonable and should be investigated further regarding delivery of our therapies. And lastly, of course, would be a primary non-response. Not just because the mechanism is wrong, but because the patient's allergic or intolerant to the therapy. And then there's a variety of reasons patients lose response over time. They have a change in their dose, either by the clinician or by themselves. There may be a change in the delivery mechanism which has to do with the bowel getting more inflamed or less inflamed and transit time changes. And there may, in fact, be some unintentional non-adherence related to changes in insurance, or changes in the availability of certain therapies which we know can lead to loss of response and cause big problems for our patients. So when we think about our historical treatment tracked strategies in Crohn's disease and ulcerative colitis, they really have been flawed. Waiting for patients to fail one therapy before moving onto another one is really a strategy that puts patients at more risk than benefit, and unfortunately, takes too long to get them well. The strategy, also thinking about a pyramid based strategy in which we use lots of therapy for patients with milder disease, and it's more common. Then we'll move up only for patients who have failed other therapies or who are sicker, is also a strategy that doesn't take into consideration prognosis or the need for different therapies in patients earlier in their course. So thinking about how we might do this starts with the understanding of which patients are highest risk for the worst outcomes. In Crohn's disease, this has been studied in a variety of ways. And I'll refer you to the specific study of a referred cohort of over 1,000 Crohn's patients. The three factors that were independently predictive of a disabling Crohn's course within the next five years included an initial requirement for steroids, the age at diagnosis below 40, and perianal disease at diagnosis. Smoking was of interest as well but missed the statistical significance in this analysis. What about in patients with ulcerative colitis? Well, there are a variety of clinical predictors of poor response or needing surgery. And this includes, of course, the clinically ill patient with bad ulcerative colitis. But of particular interest is the patient who has a low serum albumin. Not just because they're malnourished or they've been sick and developed a low serum albumin, but more and more because we've come to realize that some of these patients are actually losing protein in their stool. And the loss of protein in their stool is also a marker of a lower likelihood of responding to a protein based therapy like our monoclonal antibodies. So you can think about the clinical characteristics of your patient with ulcerative colitis, and have a good sense for which therapies may or may not work before you even start them. Another thing to keep in mind has to do with endoscopic severity. So we know now from many different studies, most of which have been retrospective, that the severity of the endoscopic appearance when you look with the scope predicts the likelihood of needing a colectomy or failing medical therapy. This, of course, makes sense to us in the era of endoscopy that we now live in, but the reality is that these have not yet been incorporated into many of our clinical grading scales. Nonetheless, when you scope a patient and see severe ulcerations, you certainly appreciate that this patient is sicker, this patient has a likelihood of a worse prognosis, and probably needs different therapies rather than starting at the bottom and giving people mesalamine as an example. What are some of the factors contributing to primary non-response, or loss of response, to TNF inhibitors? Well, there have been a variety that have been described. The first is treating a patient with no inflammation. That should make sense. So you don't want to be treating a patient with an irritable bowel with an anti-inflammatory therapy. The second would be using the wrong endpoint, such as having a patient who's actually healed and has scar tissue, but we're using more and more therapy because they're having symptoms from the scar tissue, not from the inflammation. Another part of this might be the patients having diarrhea related to bile salts, or rapid transit, or to vitamin deficiencies, or actually has an overlap of another diagnosis like Celiac disease. Perhaps the mechanism of inflammation is not TNF dependent. In a patient who has a normal CRP, or patients who have higher pANCA titers we know are less likely to respond to TNF inhibition. And there have been some nice studies showing that there are some polymorphisms related to response to TNF inhibitors. And I'll show you that in our future discussion in a moment. Lastly, the patient who smokes, unfortunately, has a higher risk for worst Crohn's disease in general, but is also less likely to respond to our anti-TNF therapies. The pharmacokinetic issues are more and more being described by our colleagues, including our colleagues in Canada, who are doing some nice work led by Gordon Greenberg, looking at some of the things that change over time as we give anti-TNF therapies and monoclonal antibodies. Patients who respond to therapy initially may not respond as well over time if they are losing response rate related to loss of the therapy in their stool. And separately, patients who respond well to high doses of therapy may not need as much therapy as they move into maintenance phase. There have been a variety of factors that have been described associated with pharmacokinetics of monoclonal antibodies, and this is quite important. And I've summarised how they affect the drug levels in red on the right. First of all, the presence of anti-drug antibodies will reduce the amount of drug that the patient is exposed to. Concominant use of immunosuppressants, for a variety of reasons, raises the drug levels. Having a high baseline TNF level-- not anti-TNF drug but actual TNF in the blood and possibly in the tissue-- is associated with lower drug levels. In theory, at least, based on the binding of that protein by the drug. Low albumin, as I've already mentioned, is associated with lower drug levels and having a high baseline CRP, elevated body size or BMI, and males more than females all will have increased clearance of drug. So you can start thinking about some of these factors even without having to order any advanced testing. Your male patients more likely to need dose escalation. Your male patient may need a higher drug dose. Your patient with a low albumin may need a higher drug dose, et cetera. This is a nice study from a group in Amsterdam that specifically looked at fecal loss and correlated it to the response and severe colitis. And what they showed in a small number of patients was that the more anti-TNF therapy, specifically infliximab, that was present in the stool, the less likely the patient was to respond to the therapy. This proof of concept study demonstrates that we should be thinking carefully about how to dose these drugs. Or even measuring surrogates of drug loss in the stool like other proteins that may be present as a way to predict who may need more drug or who may need just a completely different mechanism to control their disease. So in order to optimize therapies for IBD, we should be educating our patients first on what the goals of management are. Make sure they understand that our goal is not just symptom improvement, which is very important, but also stable remission and off steroids. In addition, if we're moving towards objective endpoints, we need to think about how we can communicate to patients that our goal is to try to reduce their CRP, or heal their bowel, or some other measure that we will define. We should choose our initial therapy based on how sick they are, as well as based on their prognosis and some additional factors. We want to make sure we don't under dose our therapies. If we're going to choose a therapy we should optimize it the best we can, and we should have a specific plan for follow up. In other words, we shouldn't say, let's see how you do, and leave it open. We should plan for specific follow up in the clinic and phone calls to make sure patients are moving through the response to their therapy, or moving on to something else. So how might we think about this? Well, with our 5-ASA therapies, we've long appreciated that there's a dose response, but I'll throw in that there's also a delivery response. So if you add oral and topical therapy together, patients, in general, do better. Whether they have distal colitis alone, or even more extensive colitis. For patients on steroids, we certainly want to recognize that they can work, but when they're not working patients should be off of them or come off of them quickly. And if they are working, we need to transition smoothly and quickly to a steroid sparing therapy. Our thiopurines, which have come under fire over the years recently for some safety concerns, as well as some of the comparative effectiveness studies, do have benefit in some patients. And using thiopurine-associated metabolites can show adherence and can demonstrate shunting profiles. When patients are shunting more drug to 6-MMP and may benefit from the addition of allopurinal, with specific concern and monitoring for safety. And our anti-TNF therapies. Therapeutic drug monitoring has been of great interest to us and started with the concept of approaching a patient losing response, but has move to the understanding that we can also use our anti-TNF therapies to predict the likelihood of response, and to understand who's having rapid clearance so that we can preempt a clinical relapse or loss of response. And there's a variety of other mechanisms to consider in our treatment strategies. Temporary bowel rest, or even diversion of the bowel, is still a strategy that can be an effective induction strategy. Using calcineurin inhibitors like cyclosporine or tacrolimus has a role in some patients, and our novel therapies, of course, that are emerging. So first, you have to recognize what your outcome measure may be for a specific patient. Is this a specific symptom that you know correlates very well to an objective marker of the disease, such as the scope appearance, or CRP, or is it growth and development, which is critical in our younger patients? Is it restoring the lab values-- so the patient's nutrition, their hemoglobin, they're CRP-- or should it be our mucosal healing for many patients? Should we be taking a look in patients despite their resolution of symptoms? Should we be taking a look with the scope in order to make sure that they're actually healing and adjusting therapy? In order to do this, we need valid indices such as the simplified endoscopic scoring system for Crohn's, or the emerging UCEIS, the Ulcerative Colitis Endoscopic Index of Severity. And the FDA has expressed interest, and there's ongoing work now looking at histology as another endpoint. And there's a variety of surrogates, such as fecal calprotectin which may be of interest as well. So depending on your individual patient, I would argue that you should have an individual outcome measure that specifically correlates to their disease control, and which together you can use to manage their disease more effectively. So a proposed algorithm for treatment of IBD focused on whatever your specific outcome might be, is the following. A baseline assessment of disease activity by endoscopy paired with a surrogate marker. We choose our initial therapy not just based on how sick they are, but also based on their prognosis. And after some defined amount of time-- between three and six months usually-- we would reassess their disease to assess whether they've achieved the control or the target that we identified. And if they have, they enter a follow up cycle in which we are monitoring them for ongoing control or loss of control. And if they haven't, we talk to the patient, we discuss our options, and if the patient is willing to proceed, we escalate our therapy in a defined way and keep going through this cycle until the patient has achieved our target. When the patient refuses or we run out of options-- this is where we are in our current practice for many-- clinical followup, clinical trials, or some other salvage therapy. So what we're talking about in the big circle is the treat to target, and in the smaller circle is disease monitoring. And in the disease monitoring phase is where we might consider de-escalation of therapy over time. So can this actually be done? On a recent paper published in the IBD journal in 2014, it was shown-- from the University of California in San Diego-- that in retrospect, patients who had treatment adjusted based on their endoscopic findings, in fact could then, in follow up, achieve some healing. And often, when they achieved mucosal healing endoscopically, they also achieved histologic healing as well. So this is a proof of concept study that suggests that in fact, yes, for some patients we can escalate therapy, even when they're feeling well, to achieve that level of healing. The next question, of course, is does this change outcomes? So what is this concept of de-escalation, and how can we conceptualize it? Well, in this cartoon what I've created here is on the left will be the inflammatory burden demonstrated in right. So in a standard patient who has active inflammatory bowel disease, who starts out with a lot of inflammation and a high inflammatory burden, with some treatment that's imposed upon them, hopefully that inflammatory burden comes down and is then staying down over time with some fluctuation. On the right represents therapy intensity. So our standard approach over all these years has been that you induce remission, and then maintain with some similar therapy intensity over time. In other words, what some have described the amount of therapy you need to get the patient into remission is what you stick with as long as it continues to work. The concept here is that the induction therapy continues at a similar dose as the maintenance therapy, and a lot of drug may be used over time in that patient. What we're talking about now is recognizing that the disease burden, or the therapy needed over time, may actually decrease because the inflammatory burden decreases. So this concept is that the maintenance therapy may be systematically decreased, or de-escalated, and therefore less drug is needed. What isn't well defined at all in our field yet is how long you would wait between induction, achieving maintenance and stable control, and then dose de-escalation. There are some examples of this, however. With steroids we've been doing this for years, so we use steroids to induce remission, and then we withdraw the steroids and leave them on a stable maintenance therapy. That's a classic example of intense therapy when the patient is sick followed by a different strategy for maintenance. For concomitant immune modulators with anti-TNF therapies, there have been a variety of different strategies described. Maintaining patients on both drugs seems to be appropriate for some of our patients, but we've learned that for patients who are immune modulator experienced and have failed those therapies upfront, we can potentially withdraw those immune modulators without an observable difference over time. Although the study that looked at this had to do with a small subset of patients and possibly wasn't powered enough to show us what we really want to know over the long term. And then more recently in an open label experience, patients had their anti-TNF withdrawn and were left on their immune modulator. So there are different strategies with these two types of therapies that we can consider. And even in 5-ASA, a recent uncontrolled experience in the study called, Momentum With Delayed Release, 5-ASA MMX, 4.8 grams per day was used for induction. And then patients were all converted to 2.4 grams per day for maintenance. Those who had mucosal healing and symptomatic response, so the so-called complete response, actually did fairly well with this dose de-escalation, or dose decrease strategy. So it is possible and there are different considerations available even today with some of our therapies. So if you believe in all this, and we move on towards treating to a target, we also have to acknowledge how we monitor patients. And this raises the concern about drift. So anybody who has a chronic disease that's in control needs to acknowledge that they may have drift over time. Meaning the disease may have a tendency to reactivate, or the disease may drift within some stable parameters. In order to understand drift we also have to understand what our threshold would be to make a change in the therapy in order to recapture those patients. This has been shown with a variety of different ways, but I'm showing you here the results of the story trial in which infliximab was withdrawn in patients on combo therapy. They used both CRP and calprotectin to monitor patients to see when they might need reintroduction of the infliximab. And they demonstrated that both CRP and calpro preceded clinical symptoms as a way to monitor the patients and know when they needed infliximab before they had a flare or before they had a clinical consequence of being off that therapy. Likewise, if I show you the results of this nice study by Sorrentino and colleagues, they were looking at patients who had an ileocecectomy for Crohn's disease. In the patients who had no recurrence after the surgery, they had a stable pico calprotectin at a low level over time. So calpro was used here post-op to predict the likelihood of having recurrence and maybe needing other therapies. Among the patients who had a recurrence but didn't respond to either mesalamine or infliximab that was re-introduced, you can see that their fecal calpro levels stayed high throughout, and they didn't have to be scoped on repeated occasions to know this. And of those patients who had recurrence but responded to infliximab, you can see how quickly the fecal calprotectin comes back down to near normal, or normal, baseline. So we can use something like a fecal calprotectin in some of our patients to monitor the patients over time, to understand when they might be drifting away from disease control when they need therapy, or even when they're responding to therapy. All getting away for scoping. The key message here, however, is that the patient has to have elevated levels to begin with, there have to be observable changes, and we have to come to terms with some of the reliability and variability issues with these lab results. But for the most part, this concept is an important one, and I think does apply to some of our patients. So the therapeutic drug monitoring concept is the last major concept here, and this is not just for anti-TNF therapies. We can use therapeutic drug monitoring to assess loss of response. We can use it to predict stable response by checking anti-TNF levels early. For instance, we could predict the likelihood the patient's going to be responding to the anti-TNF as long as out to 52 weeks in the ulcerative colitis trial in which this was looked at. And we can even use monitoring and checking drug levels early for preemptive dose adjustment so that we can keep patients under control. A nice study presented at DDW in 2014 demonstrated that if you were monitoring your patient over time, even when they felt well, you could adjust their infliximab level and therefore keep them under stable control, rather than losing control or having clinical relapse. And lastly, would be the concept of dose reduction to avoid toxicity. This certainly happens when we monitor CBC, or even a 6-thioguanine metabolites in our patients on thiopurines, but can occur with some of our other therapies as well. And so the concept of therapeutic drug monitoring is not just about our monoclonal antibodies, it applies to all of our therapies. And I would argue you're doing it already in your practice. So what's the future? The future of drug selection and dosing really has to do with much better ways to predict who will respond to which therapies, and how to do this in a way that we can communicate more effectively with our patients. Future approaches will include measurement of biologic targets in order to choose therapies wisely. It'll include the additional biomarkers for response to the therapy, and they may be therapy specific. There are more sophisticated use of combination therapy and de-escalation strategies that are being studied right now. Specifically the de-escalation concept is a good one to keep in mind and to look forward to the future. And there's the rapidly growing area of pharmacogenomics. Not just understanding which genetic alleles may be associated with response or side effects, but also, how do we communicate this more effectively to the practicing physician and to the patient? And then lastly, would be, how patients might monitor their own disease. Are there some objective markers for disease control that they could do at home that would change our ability to control their disease over time? So there's such a thing as the genomic prescribing system that's being actively studying in a variety of places. At the University of Chicago Medicine there's a project going on right now called the 1,200 Patients Project. And then the principal investigator is a colleague of mine in oncology named Peter O'Donnell. 1,200 patients have been recruited and have had their pharmacogenomics determined. Of these, many have been in the GI practice with me, and what we get is a dashboard that tells us a little bit about their pharmacogenomics. We get red light, green light, and yellow light related to whether or not they were tested for a specific allele related to a class of drugs or even a specific therapy, and whether or not we should or could or shouldn't prescribe that therapy. And then we can even click on links to get more information about these therapies. So you can imagine how this might influence your practice and make it much easier for you to understand which of your patients can respond and get certain therapies. If we look back to what I mentioned much earlier in this talk at the infliximab pharmacogenomics, there have, in fact, been some specific mucosal gene signatures that predict response to infliximab in patients with UC and other work that's going on in Crohn's. Recognizing this, if we can then translate this to that dashboard that's being used by Dr. O'Donnell and his colleagues, we can imagine starting to use these to say, you know what? This class of therapy is not the right one for you. We should be thinking about a different option altogether. And lastly, I'll refer you to work that Corey Siegel and Marla Dubinsky are doing, which is translating all of this into a way that patients might understand it. Where they take that prognostic information, they take the information related to their likelihood of responding to certain therapies, and they graph it out in a way that patients might look at this just like they'd look at their portfolio for investments. What amount of risk are they willing to take, what do the therapies currently offer them as individuals, and how might they incorporate this into their understanding of their treatment goals and discuss it with their health care provider? So I'll summarise that we've covered a lot of information related to the need for an individualized approach to the inflammatory bowel disease patient. There are, in fact, many available currently studied and understood clinical characteristics that can guide us. We recognize that one size certainly does not fit all despite our traditional treatment strategies. We can use clinical markers of disease severity to guide our initial therapy, and we should be doing that. We're moving towards a treat to target approach using a systematic assessment, and identifying objective markers of disease control that we can rely on and move towards with our patients engagement. And we are now more and more understanding the distinction between a sick patient and their disease burden and their therapy intensity. And understanding that as we move towards maintenance, we can reduce their therapy. We might even be able to withdraw specific therapies as long as we have a specific monitoring plan for that patient. So we would know when to re-initiate or adjust our treatments to prevent clinical relapse. Thank you very much for your attention.