Everyone, we're back. It's my pleasure to introduce the Turn Therapeutics team, Founder and CEO Bradley Burnam, and Head of Clinical and Regulatory Stephen Hahn, to walk through the Turn Therapeutics story. I'll leave it over to Bradley to give an overview of the company and tell the story. Again, thanks for joining us.
Thank you for inviting us. Like Alex said, my name is Brad Burnam. I'm Founder and CEO of the company, as well as creator of the technology that we'll be discussing. I am joined by my very good friend and clinical trial expert, of course, Dr. Stephen Hahn. Today we're going to be talking both about trial design and enrichment strategies and how they can affect the labeling of the drug and ultimately the access, as well as a bit about the design of our trial. I'm going to turn it over to Dr. Hahn to discuss the first part.
Well, hello everybody. Thank you, Brad, very much. I really appreciate being here and the opportunity to talk about what I think is a pretty important topic. I'm going to focus first on trial enrichment. What we mean by that is really what the FDA has helped us define this as, which is it's the prospective use, meaning the definitions you use for enrollment in a clinical trial of any patient characteristic. That could be clinical, that could be laboratory, that could be a radiographic finding to select a patient population in a study or a participant population in a study in that allows for that drug effect to be more likely. So enriching increases the probability that you'll see a signal in a specific patient population and tends to de-risk the performance of the study in terms of having the intended outcome.
This is all driven by the biology of the drug and the interaction with the disease and the pathophysiology of the disease. So you start there from a scientific perspective. There are a couple of different types of enrichment. One is prognostic. So you can select patients who are likely to have a disease-related event, high risk, say, for myocardial infarction, heart attack, or stroke, or greater disease progression. Think of the variability in breast cancer subtypes in terms of breast disease progressing, so that the trial has enough signal that you can measure it in the clinical trial. Because if you can't measure that signal in the population of participants in the study, you don't have a study to perform. Then there's predictive enrichment, which we see a lot more of, particularly in the cancer area.
If you are addressing a specific mechanism of a drug, think of HER2/neu in breast cancer and HER2/neu associated drugs. You have a biomarker-defined subgroup, which allows you to enrich that patient population, increasing the possibility of seeing a treatment effect and decreasing the variability in the population. Again, all towards the end of trying to design a successful trial that shows how well the agent that you are thinking about is working in the patients. Now, next slide, Brad, please. What is really often not clearly recognized is that from an FDA and regulatory perspective is that how you design the trial and therefore the enrichment procedures that you use for it very much connect to the label.
Of course, the label is important to doctors and to patients for sure, because they want to know, if I have a certain condition with a certain set of characteristics, is this a drug that is likely to work for me? Then, of course, insurers use this because they want to know that they are paying for a treatment that works in a specific patient population. So they are going to restrict it often to what the label shows. Not always, but often. Then, of course, when we think about the total addressable market from a commercial perspective, how we define that enrichment, how we define pivotal clinical trials, really does end up being very pertinent ultimately to the label that is negotiated with the Food and Drug Administration, which is sort of the definitive answer after a drug is approved.
There are some classic trastuzumab for HER2-positive breast cancer examples of this. Around 87%, 88% of cancer drugs in the last two years approved by the Food and Drug Administration were enriched with predictive biomarkers. So a trend we are seeing now in other diseases as well. Next slide, please, Brad. If we move now to atopic dermatitis, think about the variability component of this and how enrichment might help or not. This all gets back to how dermatologists and other clinicians measure the severity of atopic dermatitis.
There is a way to measure lesion severity that is called the IGA, vIGA, which is the Validated Investigator Global Assessment. It is a five-point scale, higher being worse, lower being better. Then there is EASI, which is the Eczema Area and Severity Index, which is really about the extent of the disease burden, both important factors for a patient with atopic dermatitis.
What you see on the left is you could have a patient who, and Brad uses this example all the time, severe eczema as measured by the vIGA. In this case, it is on the neck on this figure. But think about hands where a patient has difficulty buttoning his or her shirt or using a fork and knife because of the severity of the atopic dermatitis. But a relatively low EASI score because it is localized to one area. That still, to that patient and the clinician, is moderate to severe atopic dermatitis. Not defined that way in clinical trials, but still clinically significant. If you go to the right, you can see here is a patient who has both a high IGA, so severe disease, but also a pretty extensive pattern of the atopic dermatitis across the skin. In this case, an EASI of 16.
Two very different, but important ways of measuring disease severity. That gets to the issue of how you would enrich a patient population or a subject population for a study in atopic dermatitis. Next slide, Brad. You can see here just with the approved systemic atopic dermatitis therapies, they all share one thing in common, and that is the bar of an EASI greater than or equal to 16. These drugs were assessed, subsequently approved, moderate to severe atopic dermatitis, but really based upon an EASI score that showed a pretty widespread extent of disease. Again, there is an associated IGA that is also pretty severe with these therapies. But what it leaves out is that patient to the left that I showed on the last slide, who has, to their mind, severe but localized atopic dermatitis and therefore clinically important. Next slide, Brad.
There is a gap that is created with the measuring system, and this is not just an atopic dermatitis issue, this is across many different diseases, where the way we measure the disease and measure outcome sometimes creates these situations where there are folks that have unmet medical needs. Currently, if you have a low EASI score, regardless of severity, the current treatment is with topicals, PDE4 inhibitors, JAK inhibitors, et cetera. Drugs that you can put on topically that address localized disease. These folks, in general, are not eligible because of the label insurance, et cetera, for systemic-related drugs, DUPIXENT, for example. Those patients on the right with a higher EASI score are typically treated with biologics. Really important here in terms of understanding the patient population.
If you had, for example, an agent that was effective for localized and extensive atopic dermatitis and really helped with moderate to severe, whether it was localized or extensive disease, that would be a way of enriching the population. If you go back to that previous slide of what the population definitions were for the systemic drugs, having a definition which also includes folks with high IGA scores but relatively limited disease would be one way of enriching the patient population in that drug. It all gets back to what is the biological effect of your drug, how does that interact with the pathophysiology of the disease, and then what did you see in your preliminary data that then allows you to make very rational and good decisions about enrichment of patient populations in pivotal trials. Brad, I think that is mine. Yep.
Thank you. Now I am going to talk a bit about our trial. The trial was designed from the outset as an adaptive trial, a two-stage trial, to be more specific. When we began this trial, especially with a disease like atopic dermatitis, which has so many different variables to measure. You have IGA, you have EASI, you have pruritus, you have POEM, you have as many assessments as you could imagine. We did not necessarily know from the outset which was going to be the best endpoint for measuring our drug, and also what would be the driver to the enrichment target. So we set aside those first 50 patients for an interim assessment, and we unblinded those first 50 patients without looking at patient 51 forward, which is being enrolled as we speak under a design that I will discuss. We said, "What is happening here?
When we introduce this drug to the human body that is affected by this disease, how do they interact?" Let the drug and the body actually cast a vote as to the design of the trial. We learned some very interesting things from those first 50 patients that helped us design the next stage. One of the enrichment factors that is very infrequently, if not never used, I have not seen it in other trials, is how much discomfort by virtue of the self-reported itch score and the self-reported pruritus score is enriched in these trials. We found that in our 50 patients that when a self-reported itch score was five or more, regardless of surface area, it can be anywhere from 1.1 to 50, we had a large amount of variability in our trial.
Across these four endpoints, which we consider the primary efficacy endpoints that we are interested in, with week four vIGA being the most likely registrational endpoint that we would be looking at. We actually got statistical significance in this week four vIGA, and this is public info. We have discussed this in other webinars, as well as pretty good separation across the board. As we went up and we looked at these nested subsets, these are each inclusive of the next. We go to six on the pruritus index for anybody that had at least an EASI of 1.1, but a self-reported itch score of six, we got even greater separation. Then finally, when we got to seven, we actually saw a pretty dramatic separation across the board in this first group.
It was a 25-patient subset, but we actually got statistical significance in two out of four endpoints that we were looking into in these first 50 patients, being week four vIGA and week four EASI-75, with pretty dramatic separation at even such high levels at EASI-90 and EASI-100 at week eight, showing potential durability. Looking at something like a six or a seven is not atypical in the trials that we see. For example, DUPIXENT's pivotal trials had at least a seven as a baseline mean itch. VTAMA, which is indicated for mild to severe, had a 6.7 of 6.8. Adbry, roughly a 6.5. We have decided that for stage 2 of our trial, for efficiency purposes, this does not rule out that we would use a lower pruritus score as an enrichment or an inclusion criteria in later trials.
We wanted to use a baseline pruritus of at least seven, but the trial was able to remain inclusive of patients across the disease extent index. I think of the EASI index as what is the extent of the inflammatory burden that the body is dealing with. Of course, when you have an index with a very large numeric variability, from 1.1 to 50, as we have seen in trials, you do not want to have imbalance inside of your analysis set.
We did, with Dr. Hahn and Dr. Stauch, our amazing statisticians' help, we created a prospective one-to-one stratification with what I like to think of as three buckets, they are called stratums, of course, where you are typically considered mild amount of EASI coverage, even though this does not account for the severity of their lesions, which would still be at least a three or four in all areas.
A patient with a lesser disease burden, somebody, for example, that severe hand eczema patient, which would usually be measured at about an EASI of four. The 1.1 to 7 group is in this bucket. Again, it is one-to-one randomized within this bucket so that we can have balance in this group. Then another 60 patients in this group, one-to-one randomized for what you might consider your moderate surface coverage. We think of drugs like VTAMA, and maybe we start branching into the systemics at that point as we get toward the 14 range or approximately. Then of course, we do want to have representation from our severe extent burden. We are enrolling people with an EASI of greater than or equal to 16.
Those patients tend to be a little bit more difficult to recruit, of course, but we wanted representation on the trial, so we are adding those at roughly a 15. It could end up being 16, depending upon the enrollment target and what we get to. As we discussed, these are the enrollment criteria for our trial. The baseline pruritus index is greater than or equal to seven. We have the separate strata for the EASI to maintain imbalance. We do not want to be comparing apples to oranges. If we have a bunch of low EASIs being compared to a bunch of high EASIs in the other arm, it is not going to create for a good balanced trial.
Everybody has a minimum IGA of at least three being enrolled in the trial, which means their worst lesion has to be at least a vIGA of three to be eligible. When we took the design that I just showed you, and we essentially overlaid it on the first 50 patients, what happens if we take those 25 patients that meet an EASI of at least 1.1 and a baseline self-reported pruritus of at least seven, what does this look like on the 50 patients that we have already unblinded? We have not unblinded patient 51 forward. Those are still mid-enrollment. But what does it look like on the patients that we have already looked at that are coming out of the same investigators? It looked pretty compelling as far as we are concerned.
You can see that week 4 vIGA was 61.5 versus 8 with a P of 0.01 statistical significance, which was very exciting to see. Week 4, EASI-75, we got a P of less than 0.05 as well, with pretty dramatic separation. What got equally interesting is seeing these week 8 durability numbers. When you have 46.2% of patients on the drug side reaching an EASI-100 versus 8 on the vehicle side, we like to believe that that is not an accident. We are hoping to see it again in the next stage of the trial. We of course, have not unblinded yet, but these are endpoints that we will be exploring. When you look at our stage 2, and when we look at our stage 2, we obviously are hoping that history repeats itself. Based upon the statistics, we are very hopeful.
That being said, it's trial design, and we're hoping to get the most information out of this drug that we can, out of this study that we can, and use that to further stages of the trial. I think that I will stop there and thank Alex for the invite. I thank Dr. Hahn for joining, and we can tune out.
Great. Well, Brad and Stephen, I really appreciate you taking the time to join us, and, yeah, looking forward to the data.
Thank you, sir.
Thanks, Alex. Thank you, Brad.