 Good morning. I'm sure we'll have some more people kind of trickling in but I thought we'd go ahead and introduce our speakers today and get started. Our first speaker today is Barbara Morosco. She's a clinical associate professor here at the Moran. She's going to be speaking to us about research opportunities here. It should be very interesting that we also have from the lower campus Joyce Mitchell. Oh here in the medical school. Thank you. Joyce Mitchell. She's the associate vice president for academic health information technology professor and chair of the department of biomedical informatics. She's going to be speaking to us about a clinical cohort finding system to help us with our research as well. So please welcome them. So I actually have Steve to thank for this opportunity and sort of the stimulus here to think about doing this. So when I sat I don't know if Mark's here. Yes. So when I was here Mark Grave Grand Rounds maybe about a month ago primarily focused on the residency program and hit upon a few topics on what he had heard from the residents in terms of research that was available here at the Moran. When I was joining or coming on board and talking to Randy we started talking maybe about six years ago. It was the first time I met Randy. I saw the opportunities here and I said wow this is just such a great place to do all aspects of research. I don't have a PhD. I've never really been in a lab per se working at the bench except now with Bala doing some animal studies. But there's so many types of research and it's really here. It's right here and Steve also opened up another whole opportunity opportunity that's here through further which is what Dr. Joyce Mitchell is going to speak about. So this is supposed to be interactive. It's supposed to be creative. It's really geared towards the residents medical students fellows. So what is research. Well it's a careful or diligent search. Studious inquiry or an examination especially an investigation or an experimentation aimed at a discovery and interpretations of facts. Revisions of accepted theories of laws in the light of new facts or practical application. It's the collection of information about a particular subject. So basically just ask a question any question. There's all different types of research observational research correlation research quasi experiments and true experiments. So what's observational. Well case studies human behavior studies primary characteristic is that the phenomenon are being observed and recorded. So you can even do a prospective observational study in triage. How many patients come in with corneal ulcers. And what's the inciting event. Is there a difference in patients presenting in the summer versus the winter. What about dry eye. Dry eye is supposed to be seasonal. In fact I was just at a conference this past weekend where I don't know how many of you are familiar with eye gait. So it's I ontoporesis for the treatment of dry eye. And they ran a clinical trial phase three. No surprise did not meet their primary endpoint. But what they did find was that the recruitment of the sites differed over the year. And the results for that site differed by season. Very observational. Informative correlation. This is a very busy slide. But basically it examines the co variation of two or more variables. So let's think about this. So research on cigarette smoking and you want to see if there's a correlation between lung cancer and cigarette smoking. There's two variables here. The variables the smoking and the lung disease. Do they vary together. Do they vary at the same rate. So somebody who increases smoking does that increase their rates of cancer. Correlation research can be accompanied by various techniques. You can do it retrospective. You can do it prospective. You can do a database analysis. You could do a cross sectional study. Just bring some patients in. Again I'm going to use triage. So patients come in with corneal ulcers. What's the correlation between the corneal ulcer and contact lens use. Over the past five years. What's the correlation between contact lens use and keeping in their contacts and corneal ulcers. Again correlation. It's observational. Nothing is manipulated. You're not treating. It's not interventional. And it's not necessarily casual research. Correlation research is often conducted as an exploratory or a beginning research. And again with Dr. Joyce Mitchell here when she talks about what can be gained from this further database that everyone has access to. It's really exploring correlations. Steve and I were talking about there's a lot of research now on sleep apnea and glaucoma. Do the glaucoma patients here at the university have a higher correlation with sleep apnea. The data is all over the place. Some people have found yes there's a correlation. Other people have not searched the database. See. Look at the diagnosis. Quasi experiments. I actually like this term. So they're similar to true experiments. But they use naturally formed or preexisting groups. So the subjects can't be randomized. And it's not a true experiment but it's not a junk experiment. An example would be lung capacity in old and young individuals. So there's a lot of confounding variables. You've got pollutants. You've got cigarette smoke. You've got demographics. You've got age. You've got where the patient lives. So there's a lot of differences between the groups that you can't control for. But again some of those differences could account for your findings. So you design you carefully assess the casual or causality with your quasi-experimental designs. And again in some ways it's similar to a correlation study. But you can control one of the variables. So maybe you can have the patient stop smoking. So in some ways it's partly interventional but you're not controlling as you would for a prospective clinical study. And this is the true experiment. So this is probably what most of you are familiar with. You have at least two groups. You have an experimental group and a control group. And each group will receive a level or change of the independent variable. So you've got an independent variable. You've got a dependent variable. And the dependent variable will be measured to determine if the independent variable has an effect. So let's put this into perspective. Glaucoma study. Your dependent variable is your IOP. You're going to see if your independent variable which is your glaucoma treatment is having an effect on your dependent variable. And then you're going to ask if there's a correlation with that IOP and progression over long term. So there's a correlation component. The control group is going to provide us with a baseline. And in true prospective randomized studies and I'll get into a little bit about that. You do have a control group because you want to know what is happening if you're not intervening with that independent variable. All subjects should be randomly assigned and be tested in parallel. Because again, as we saw with my example with the IGATE, patients were not necessarily treated in parallel. You had one site enrolling in June. The next site enrolling in December. So they were treated within the site in parallel. But they were actually treated at different time courses. And that actually affected the outcome in this dry eye study. And again, it was pretty a pretty major deal because they failed to meet the primary endpoint in a phase 3 study. Should it be conducted as a blinded or masked study? And it's funny. You can always tell the people that I joke about this. You may have heard me say this, but you can always tell the people who have worked in ophthalmology because we never say a blinded study. We always say it's masked. So what are the components to a controlled prospective study? Experimental or the treatment group where you vary that independent variable. You have your control group. You have an independent variable as I mentioned, dependent. It's a random assignment. And again, it could be double blinded or masked. Investigate or a single masked is another term. Various forms. You can have a clinical prospective study, which is interventional again, phase 1 through phase 4. And again, you can have non-interventional, and we spoke a little bit about retrospective case studies, database analysis, we'll go more into, and literature reviews. There's various levels of evidence. And when you think about the level of evidence, the tip of our pyramid is systematic reviews, which are reviews of what's already in the literature on published studies. So for instance, if you did a systematic review of all the Lucentus studies and came out with a conclusion on the efficacy and the AEs, that would be probably the highest level of evidence, because it's a conglomerate of studies. Below that again is your randomized controlled trial, going down cohort studies, case control, case series. And then at the bottom are editorial and expert opinions, which still carry weight, but their level is on the lower end. So what's an interventional clinical study? We call them randomized controlled trials, RCTs. And again, they're randomized, and that is the preferred way to assign participants to control or interventional. This eliminates and controls for variables. So you want each treatment group at the baseline, that patient will look very similar across treatment groups, age, race, demographics, baseline dry eye, corneal findings, IOP, et cetera. And then again you want to mask. So you want to remove the bias. You don't want to know that Mrs. Jones is actually getting treatment. You want to know or placebo. You want to know that you don't know what she's receiving. And again, it removes the bias, both of her and you. An assignment of a participant is determined by a formal process of randomization. I've been throwing around randomization, but it's really just a random numbering. And you can, this can be accomplished in large clinical trials, either through a IVRS or an IWRS, or even in the clinic if you have a small trial, you can just basically say, okay, patient one, group X. Patient two, group Y. Patient three, group X. And just assign them randomly. So whoever walks in the door. Chances are based on statistics and probability that if you have a large enough group of patients, they will fall fairly evenly across the treatment groups. And that's what you hope for. So why is randomization so important? Well, it's the preferred way of assigning participants, as I said, to control an intervention and it eliminates and controls for the variables. It also establishes similar patients across treatment arms. To test additional variables, you can stratify. So again, let me, I'll get back to a Glaucoma study. So I want to know if my treatment X works better than a placebo or a vehicle. But I don't know if it's going to work better if patients have IOPs from 21 to 25 or from 25 to 28. So I can stratify so I can have within my first treatment group, two groups of patients. The lower IOP and the higher IOP. And they'll both receive treatment X. My placebo, higher IOP, lower IOP. So within that treatment arm, you now have two groups and you're also asking the question, does the treatment work? But does it work differently between those two IOP levels? What's the placebo effect? And I was thinking of when Tony Adamus was here and was talking about the patients who were able to see better. And I think this happens all the time. You know, the patient thinks that they're getting better or they think there's something going on and they try a little bit harder or they're reading a little bit better. But placebo, when I looked into this, actually I was quite surprised because there really is some conscious belief in the patient that the drug is working. Or at least if they're getting the intervention. And there's a subconscious association between recovery and the experience of being treated. And we see this all the time. Again, even in glaucoma studies, the patients who receive placebo across the board have a two drop in a two millimeter drop in IOP. So when you look to see if your intervention, very early glaucoma studies, if you're studying your drug against your vehicle, just assume that your placebo, your vehicle group, no active, will have two millimeters of mercury drop. And this has been well reported, well studied. So again, if you're looking for a benefit of your drop, you have to almost look at four to six because your placebo will have some effect. And researchers, again, have demonstrated that it actually may be a response from the brainstem. So there is some subconscious and conscious components to placebo. So again, very important to mask. So how do you control the placebo effect? Well again, patients do do better in clinical trials than they would just coming into your clinic. If they are receiving a placebo in a controlled trial, maybe they're more responsive, maybe they're more attentive, maybe they're following directions better in these clinical trials. Maybe we've also weeded out the non-compliant patient, which is another reason why patients always do better in clinical trials. And I sort of mentioned this already. So basically when you're thinking about clinical trials, recognize that there's going to be a placebo effect, accept it, and then just plan for it in your statistical analysis. So just to briefly go over some different phases of clinical trials. So a phase one is generally a dose-ranging study in healthy individuals, and you're really looking for safety. Generally open label, although it could be masked, and you can measure formical kinetics of the drug. So you're looking to see if your drug is actually getting absorbed, what the exposure is, and if there's any safety concerns. Phase two, I like to think about that now you've taken that drug and you said, okay, it's safe in healthy individuals, let's put it into a diseased patient population. Because you never know if the safety will be different in that patient with the disease. And again, I think about our ophthalmology patients, a drug that's really irritating to the cornea that you're going to use for a keratitis patient may be well tolerated in a healthy white, quiet eye, but actually could be very irritating and very intolerant in a patient that has an ocular keratitis. Generally primary endpoint is safety, although you do want to see an efficacy signal. So you'd like to see that your drug is having some effect and is doing somewhat of what it's supposed to in that disease population. Phase two B's are generally larger, and these help you to inform the phase three, which are the largest studies, which are generally the regulatory studies. So your phase two B is statistically powered for an efficacy single, and again you want to inform your phase three. So here's where you're going to know how large is your effect. What are you going to need to do to power your study to actually get a signal, a statistically significant signal. And then also what's your comparator. And your phase two B will actually help you determine that. And again, your phase three, obviously everybody I think for the most part is aware in just very, very large studies. We tend to be small in ophthalmology compared to cardiovascular. Some of those studies are thousands of patients, three, four thousands patients, thousands of patients. There are also millions of dollars, and we tend to be several hundred patients. Again, phase three, you want it to be more representative of real world. Sometimes the criteria are a little bit relaxed. They may not be as stringent in terms of the patient that you're studying. And chronic conditions can have very, very long phase three studies. Even if the primary endpoint is only a three month or six month, a lot of times these patients are followed for years out work, because the FDA really wants to see if you're treating chronically what's the safety over time. And then phase four, these also tend to be long studies. They're open label. They can be epidemiologic. They can be observational. They also can be massed studies. It could be extensions. But it's really, again, to capture real world experience. So take the patient out of the clinical study, remove the placebo, remove the, you know, the clinical parameters of being stringent and the compliant patients and really see what the real world experience is with the drug. And we've seen this with BIOX, BIOX had a safety signal, had a black box. Other drugs actually have been pulled from the market. And, again, it's really to capture what is that experience beyond that clinical trial population. So what are the research opportunities here at the Moran? Well, they're tremendous. And they're traditional clinical trials. And we've got, we're fortunate enough to have Dr. Bernstein here leading the clinical research, as well as Deborah Haverson, who manages the clinical trial. So, again, they're definitely the go-to people if you want to hear or learn more. You can do investigator-initiated research. So IIRs, you can actually apply to a pharma company. And you can use a compound that's in development or is already approved. And you can look at novel new treatments, or you could just do an additional study in the same group of patients. So I'm trying to think what comes to mind. Think about, I don't know, erythromycin. I'm starting to think of everything that I was talking about today. So you want to look at erythromycin. I don't even know if it's approved for angular blepharitis. I don't know. I don't think it is. Okay. Divitis. So it's a perfect example. So angular blepharitis. How many times do we use angular blepharitis to treat, or actually erythromycin to treat angular blepharitis? You could do a study looking at the use of erythromycin in treating angular blepharitis versus maxotrol, gentamycin, exactly. If you do go to a pharma company and want to use a drug off-label. So, for instance, if you want to use Lucentus for the treatment of post-tribeculectomy fibrosis studies. A lot of times, if they're going to allow you to do the study and or give you grant money to do the study, it has to make financial or commercial sense for them. So it has to be an area that they would be interested in looking at or pursuing, perhaps. Then it makes much more economical business sense to them. But it actually is a very neat way to do research with drugs that you have at your hands. So it doesn't need to be an experimental drug, it can be something that's actually in the clinic. So retrospective chart reviews. Again, just ask a question. And this came up recently with a patient that I had who was on Durzal. And she had a very, very pronounced IOP spike that was very, very difficult to treat. I don't know if we really know. We know that Durzal will cause an IOP spike, and we've got cataract surgeons using it routinely, post-operatively. But how rapid is the spike? What's the degree of that IOP spike? Is it more difficult to treat? What's the onset? Is it at four weeks? Is it at six weeks post-operative treatment? There's so much here that we don't know about this drug. And again, this is a drug that you have at your fingertips. It's just been approved. It's just been marketed. And it's a very, very strong steroid. It's worth studying. Look through your charts. If this is a drug that you use routinely, follow your patients. Just ask some questions about it. Or again, the medical students, the residents, ask some questions about it. What don't we know about this drug? We don't know a lot. We know very little. And again, retrospective chart review is a great way to look and say what's happening or what had happened to these patients. And you can always use historic data, too. That's another neat thing. There's a lot of historic data on the rate of IOP elevation in steroid use with dexamethasone, with prednisolone. So you've got a historic database that you can compare it to. What's the most common ocular pathology at the triage? Again, I wonder about the patients that, guys, see a lot of corneal ulcers there. And there's a lot of patients sleeping in their contact lenses. Are there more younger patients sleeping in their contact lenses? Is that why they're having more corneal ulcers that present? And is there a greater incidence of corneal ulcers here at the Moran than, say, in another large urban city? Literature review. There's been a lot of recent interest in, again, I'm just going to throw out some ideas, caroidal thickening in AMD and also in glaucoma patients. And what is the day to tell us? It's sort of all over the place. We know that if you measure caroidal thickness, you're getting a measure of caroidal blood volume and ocular blood flow. Well, is there a correlation between AMD? Is there a correlation between glaucoma? You could review all the present literature and do a literature review, publish an editorial, publish a review of the literature. So I wanted to pull up one example. David DeMille over the summer had come to me and said he was curious as to what patients looked like when they entered a glaucoma clinical trial and did various inclusion or exclusion criteria affect the outcome? So again, thinking about that IGATE example, patients were enrolled. Oh, another interesting finding that they found was that during the study patients that had decreased corneal sensitivity actually makes sense, had worse outcomes in the fact that they did not notice as much of an improvement. Now, they did not stratify or they did not set inclusion criteria based around corneal sensitivity, but this was something they found. So obviously, if you're conducting a dry eye study and you want to know if patients feel better, you have to enroll patients who can actually feel their corneas. And this is what they learned retrospectively. So David had said, you know, well what about when you enroll glaucoma patients? You're talking about IOP, you're talking about demographics, you're talking about baseline IOP, you're talking about prior medications, diagnoses. So he was really instrumental and just did a very thorough literature search on the phase two and the phase three studies. He had some criteria by what he used to define the studies he looked at. And we ended up with four papers. So three of which have been accepted already. And it was pretty cool. You know, it really was just a question. It was just a question. Database analysis, and this is where here I'm going to introduce perfect timing to Dr. Joyce Mitchell. So again, Steve was working in triage with me one day and he said, did you know that there's access to this database that really just has a wealth of information? And we can actually ask questions. We could say what's the correlation between glaucoma and sleep apnea? We could ask the question between AMD and diabetes in the Utah patient population. So that was pretty much my introduction. And Steve has been instrumental in again introducing this to me. And I thought it was so valuable that I wanted to introduce it to you. So I thank Joyce very much for being here. I'm going to let her introduce herself. But with that, Joyce Mitchell. Thank you.