 Thank you so much. Our next speaker will be Dr. David Rubin. David is an expert in the treatment and assessment of digestive diseases with special focus on inflammatory bowel disease and high-risk cancer syndromes. David is a professor of medicine, and in the last week has become the acting chief of the section of gastroenterology at the university. Today David will speak to us on the topic ethical issues in inflammatory bowel disease. David Rubin. Well thanks Mark, and congratulations on 25 years. This is always a wonderful opportunity to catch up with old friends, and I'm honored to share one particular element of the ethics in my career, and I'm always grateful to the McLean Center for training me to do this. So recently I was asked to be a peer reviewer for a clinical trial that had been performed in my area of expertise, which has to do with inflammatory bowel disease, and it raised an issue that I had been discussing in the past about the relevance of placebo control, but this trial actually drove it home in a unique way. So allow me to educate you a little bit about these types of trials and why this is of particular importance in both, and I hope you'll agree with me that it raises some important ethical issues not just in clinical trial design, but I think also potentially in a variety of international study designs in human subject safety. I do have some disclosures that I think are relevant, although the take I have on these clinical trials is not one that I think any of the industry partners that I've consulted for would be very happy about, and the main study I'm presenting I had no conflict related at all. So inflammatory bowel disease is relevant to all of you, and if you haven't heard about it or know anything about it, I'll just tell you that there's one and a half million people in the United States who suffer from either Crohn's disease or ulcerative colitis. Later this weekend I'm on my way to New Zealand, which interestingly reports the highest incidence of inflammatory bowel disease in the world. We're not quite sure why that's happening, but it's rising in those parts of the world, and there's some very interesting observations that are going on. These are chronic incurable symptoms. Even ulcerative colitis after surgery has its own set of problems, and we are trying to treat these with medicines in most situations. Just to remind those who know a little bit about this, but for everybody, ulcerative colitis on the left is a condition that affects the large intestine, and Crohn's disease on your right is a condition that can affect any part of the intestinal tract. Most commonly affects the small bowel and a little bit of the colon. People with suffer from urgency, bleeding, diarrhea, weight loss, abdominal pain. It can be a very disabling disease, and some people refer to it as an invisible disease because they suffer tremendous problems functioning socially, and they have a lot of symptoms, yet people look at them and don't necessarily know why they would have problems or what's going on. The reason I tell you that is to focus a little bit on what clinical trial endpoints have been focused on, and they are focused a bit on symptoms and a variety of other measures. Treatments have been developed that are basically a step-wise approach to management, where we start in the upper left here, the amino salicylates, and we move through a variety of increasingly immune suppressive strategies. The rest of my presentation is really going to focus on amino salicylates. Most patients with ulcerative colitis will respond to amino salicylates and are patients who are appropriate for those types of therapies. Our goal is to induce remission, so people feel well. We will be satisfied with response in clinical trials, but that's really not acceptable when you're taking care of patients in the real world. And after we get people feeling well, we want them to stay there forever. That's what's called maintenance, of course. I learned from Mark, and I always do this now in many of my lectures, is reflecting on history of our field, and I think others have done that already today very nicely, but Nanna Spartz actually developed the first 5-ASA therapy, and she was doing it to develop a treatment for arthritis, and it was incidentally observed that it also made people with bowel inflammation feel better. The problem was that when you gave people lots of self-accelerating, there were lots of side effects. So over the years, many people have developed analogs, and they focused on the active ingredient, which is what we call 5-ASA, and a bunch of different delivery systems for these drugs. They work well. 50 to 70% of patients will have some response, and between 15 and 40% will achieve that remission in a short period of time, and they're very safe. So we like these drugs. And there have been many trials now that have been done that have been placebo controlled using these therapies, and the consistent finding in the trials has been that 5-ASA is superior to placebo in ulcerative colitis. So the problem has been when do you stop doing placebo controlled trials? And there are pros and cons in this situation that I know this audience is quite familiar with. What I hear most commonly has to do with the fact that it's a cleaner study design. You can recruit a fewer number of patients to get your results, so you're not necessarily going to have to spend as much money or time. And the other problem that has been raised has been, well, the placebo rate is between 10 and 30%. Therefore, there is some equipoise here because there's a reasonable likelihood that a patient suffering from this condition that we don't know the cause will respond to placebo, so we should be okay with this. On the other hand, it's, of course, not preferred by patients, even if they're explained what this means, and there is also the concern about harm. So as I've started to rally against this in clinical trial designs, especially with particular therapies, I've been looking for better ways to communicate about this. And when this trial ended up in my e-mail to be a peer reviewer, I thought it was of particular importance. Now in clinical trials for ulcerative colitis, they use this complicated index that includes a variety of symptom-based scores on the left that the patients tell the investigators about, and a couple things that the doctors do on the right. And the one I want to point out to you is the flexible sigmoidoscopy score. So we grade the degree of inflammation. You don't have to be a gastroenterologist to know that on the left it looks less severe or significantly inflamed than on the right. And we use that as an additional marker, which we might argue is a more objective marker of the degree of inflammation when we treat people. So what has been a focus of interest has been how reliable is an index like this in a clinical trial. For instance, everything on the left is more subjective. Patients tell you about it and we rely on them to communicate to us so we understand what might be going on. And the things on the right are the physician-reported ones. In particular, physician-global assessment, we would argue, can be quite subjective, although we'd like to think we're experienced enough to know what that means. And the flexible sigmoidoscopic scores, you could say, well, that sounds more objective. Why would that be a problem? I'm about to show you why. Now in historical trials, I'm only going to show you one slide. We consistently see that drug is better than placebo. These are older, classic landmark trials in our field. And over the years, people have replicated this study design for every new delivery system with the same active ingredient. This study that I was asked to review was similar to those, but it was done in Belarus, India, Turkey, and the Ukraine because they were developing what would be called essentially a generic formulation of this therapy that they wanted local regulatory approval for. It was a multi-center double-blind randomized controlled trial looking at this therapy compared to placebo in the same population of ulcerative colitis. And they included an endoscopic inclusion requirement. In other words, they thought, you know, the way we're going to make sure people get in this trial who are really inflamed and don't have something like an irritable bowel will be to make sure they all have those endoscopic findings that I showed you pictures of. At the end of the trial, after 281 patients were recruited, it was a negative trial. In fact, they missed their primary and their secondary endpoints. And everyone said, well, how could this have been? Why did this therapy that's just like all the other ones not beat placebo in this population? Well, it was interesting because in our field, we've been moving towards centralized reading of more objective markers. And the study design included a priori, a central reader looking at the endoscopic scores. And you know what? When the central reader who had already been trained and experienced people had agreed that this person was appropriately qualified, found that a third of the patients included in the trial had their endoscopic score upgraded by their investigators in order to get into the trial they thought or something happened where these patients would not have been eligible. In other words, they either didn't have any inflammation or they had very mild inflammation and they weren't eligible for the trial. And that we know is associated with a higher placebo response. So in this a priori methodology when they re-ran the analysis, even with a third less patients where you think there might not be enough power, they now hit all their primary and all their secondary endpoints in reanalyzing the study. Very interesting question. So one of the points here has to do with understanding investigators' intentions to get people into trials or maybe this was just a learning curve issue. The investigators didn't really know what they were looking at and they were scoring things incorrectly. Well to their credit, they also did a number of other analyses. The site investigators did not make the same mistake upgrading or downgrading the scores at the end of the trial. In fact, they were right on the money with the central reader. And to their credit, when they looked at the site investigators reading of the endoscopy scores at the beginning and at the end of the trial, there was no evidence of a learning curve. So we're left with wondering about these investigators and what they may have done to get their patients into the trial. It's a big question about clinical trial design and clinical practice of course. So what might have happened? So in my editorial that I ended up writing with this study, I broke it into unintentional and intentional violation of good clinical practice. Unintentional meaning the clinician, the investigator wants to get the patient into study to help them get their medicines. Some of these countries have difficulty getting patients' care and they wanted to try to get them some additional care and perhaps help their disease. The intentional problems that we might be worried about would include trying to reach your recruitment quota, understanding that they get paid a lot of money. Here it's a lot of money in the U.S. but in those countries recruiting one patient into one of these trials by pharma can pay for your research assistant for the entire year. So it's quite significant. And there are other external pressures we all know about like authorship and also just meaning quotas and other things, external pressures. In addition there's consequences of enrolling these patients and I think we would understand that. When patients are enrolled who don't have active inflammation, it's been consistently seen in our field they have higher placebo responses and lower therapeutic response rates and missed primary endpoints. There have been other examples in our field but never one where we had this built in centralized reader to understand what's actually happening. So I'm going to end by just summarizing why I think this is more evidence that we need to do a better job not only in understanding how we recruit patients into studies but as we use more objective measures of disease activity we also should be moving away from placebo controlled trials and all sort of colitis which is the argument I made in my editorial. We have to understand that the more we've done these and it's the same active drug over and over again compared to placebo I really don't feel any longer that there is appropriate equipoise for a study design of that kind. And as I've learned from Laney Ross and from Mark over the years there are better ways to do this even if they end up costing more in the short term. And we've been arguing now for either add on trials or the term we are now using of course would be head to head trials with comparative effectiveness. So I hope you followed me through this whole story of this particular disease state and some of the troubling things that went on in this trial design but I thought it was a really illustrative example of how I've applied my ethics training to the field that I'm in. So thank you very much. There was time. Are there any questions for Dr. Rubin on clinical trial design for the ulcerative colitis and the use of placebos? I don't see any. David thank you for getting us back on time.