Hello and welcome to the Deutsche Bank Deposit Receipts Special Investor Conference, DBV. I'm Zaf Aziz from the Deutsche Bank team, and pleased to announce our next presentation will be from Belite Bio. Before I introduce our speaker, a few points to note. Please submit your questions in the questions box. Also, all of those presentations will be recorded and can be accessed via the Deutsche Bank website, adr.db.com. At this point, I'm very pleased to welcome Belite Bio, a trade on Nasdaq, and using the symbol BLTE. Thank you. Hello, everyone. My name is Nathan Mata. I'm the Chief Scientific Officer for Belite Bio. We're based in San Diego, California. At Belite Bio, we are currently advancing through phase three clinical development and oral once-a-day treatment intended for two different but somewhat related macular diseases. The first is a juvenile inherited macular dystrophy called Stargardt disease. The second is an age-related macular degeneration called age-related macular degeneration, as most people know, AMD. We're specifically looking at the advanced form of dry AMD, which is called geographic atrophy. In both of these diseases, the accumulation of toxic byproducts of vitamin A is implicated in disease progression. Our oral once-a-day therapeutic is intended to reduce the accumulation of these compounds and slow disease progression. By way of introduction to the management team you see here, our founder and chairman, CEO, is Dr. Tom Lin, extensive experience in biotech over multiple therapeutic areas. Also, we have our CMO, Dr. Hendrik Scholl, who has extensive experience in Stargardt disease. In fact, he was the principal investigator of the largest natural history study of Stargardt disease called ProgStar. We'll be talking about that more in a moment. Of course, he's participated in numerous clinical studies of various investigational therapeutics in both Stargardt disease and dry AMD and geographic atrophy. I, of course, am Nathan Mata. My background is as a research scientist. My PhD is in neurobiology. I also have a master's degree in biochemistry. Many years ago, I was involved in the development of the first animal model for Stargardt disease. We found that in this animal model, that if you modulated the amount of vitamin A going into the eye, you have a significant effect on slowing the progression of pathology in this animal. We then advanced that into clinical studies where we are today with Tinlarebant, our oral once-a-day therapeutic that again targets these toxic byproducts of vitamin A. I'll get more into the mechanism of action in a moment. Finally, our Chief Financial Officer, Dr. Hao- Yuan Cheng, who's had extensive experience, capital market experience, has done numerous IPOs, advancing biotech companies through early and late-stage clinical development. Here's an overview of our pipeline. As I said, we're looking at two specific diseases. Stargardt disease is the juvenile inherited macular dystrophy. Geographic atrophy is the age-related macular disease. In Stargardt disease, we've completed a phase two clinical trial. This was a two-year trial enrolling 13 adolescent subjects that had Stargardt disease. We saw a significant slowing of lesion growth and some very promising safety outcomes that gave us encouragement to go into phase three clinical development. In the phase three clinical development, we've initiated a trial called Dragon. This study is a two-year phase three study that enrolled 104 subjects, aged 12 to 20 years of age. That study is nearly completed. In fact, we have an interim analysis that was recently conducted in February. We've had a very positive outcome from the unmasked look at the data vis-à-vis our DSM-B. I'll talk more about that as we get into the later in the slide deck. We also have ongoing a second trial, phase III trial called Dragon Two. This is also a two-year trial, a phase II three trial that's recruiting 60 subjects. This geography is a little bit more specific. It is focused on Japanese subjects, U.S. subjects, and U.K. subjects. It was designed really to take advantage of a recent designation we got from the Japanese regulatory authority called SAKIGAKE designation or Pioneer Drug designation, which is essentially like a clinical breakthrough status that allows us easy access and quick access to Japan regulatory authorities. Hopefully this will allow us to be the first approved drug, first approved oral therapeutic anyway for ophthalmology in Japan. Finally, in geographic atrophy, we have a phase three study, which is also a two-year trial called Phoenix. This study is enrolling up to 500 subjects and recruitment for that study is ongoing. We expect to close that enrollment sometime this summer. A little bit about the drug, as I mentioned, it is Tinlarebant. This is a novel oral once-a-day tablet that's designed to target a protein called retinal binding protein 4. This protein is the sole protein that carries vitamin A from the liver to the eye. It has no effect on other extrahepatic target tissues. Basically, by targeting this protein, we are limiting the amount of vitamin A that's going into the eye. Because these toxic byproducts are derived from vitamin A, our hope is that by targeting this protein, we have an effect on slowing the accumulation of these compounds and therefore slowing disease progression in both Stargardt disease and geographic atrophy. Most importantly, one of our differentiators, in addition to being an oral therapeutic, is that we're going after early stage disease, a disease at this stage that's not mediated by inflammation. This is very important because the approved therapeutics for GA, for instance, target inflammation, which occurs really in late-stage disease. There's really nothing for early stage disease in geographic atrophy and certainly no oral therapeutic. Of course, there's no approved treatment for Stargardt disease. We have a number of designations based upon our orphan disease status. Fast track designation, rare pediatric disease designation, and orphan status in the U.S., EU, and Japan. As I mentioned, pioneer drug designation in Japan as well. We have a very strong patent family, 14 active patent families. Most of these are matter of composition patents, not expected to expire until at least 2040 without any patent term extension. If we look a little bit into the overview of the drug, we start with the market. There's a huge market opportunity, of course, in advanced dry AMD. You can see that on the right-hand side because again, this disease is more prevalent in the elderly and it increases dramatically with age, as you can see there in the graph. Stargardt disease is a more limited opportunity because it is an orphan disease. However, it is the most common inherited retinal dystrophy known. It's approximately one in eight to 9,000 subjects. In the U.S., we have roughly about 55,000 to 60,000 subjects. And in China, we have about 109,000. We point out China because our Dragon trial, the phase III trial that we're going to talk about today, was heavily recruited in China. We have a large demographic of Asian patients in the Dragon trial. Going into the mechanism of action now, this is the vitamin A processing in the visual cycle. Basically what you're looking at is a schematic of the back of the eye where the bottom is your bloodstream and then the layers of tissue above are the back of the eye. This retinal pigment epithelium is a layer of tissue where all the enzymes that convert vitamin A into a light-sensing chemophore reside. Right above that is the photoreceptors. This whole process begins on the lower right-hand corner with the admission of vitamin A. You see here abbreviated AT-roll. That's the chemical abbreviation for vitamin A, all-trans-retinol. In the liver, vitamin A, all-trans-retinol, binds to retinal binding protein 4. This larger protein, transthyretin, binds to that, creating this very large molecular size complex, which gets dumped into the circulation. This is important because this complex resists filtration in the kidney. It allows a high steady state level of vitamin A in the blood. Most importantly for the eye, it's important to note that the eye expresses a receptor for retinal binding protein 4 that other tissues do not express. Consequently, the eye has a unique preference for delivery of vitamin A bound to retinal binding protein 4, whereas other tissues do not. Those tissues can uptake vitamin A from other sources or from other carriers. In the eye, there's a receptor-mediated process which allows vitamin A into the eye. It goes through a series of enzymatic reactions where it's eventually converted to rhodopsin in the retina. Light activation of rhodopsin liberates all-trans-retinal. This is a very reactive form and a very toxic form of vitamin A. It has to get out of the retina through some mechanism or it'll actually start destroying cellular membranes. The way it gets out is through an active pumping process mediated by a protein called ABCA4. ABCA4 is essentially an enzymatic flip base, which grabs the aldehyde from inside the retina and flips it outward, availing it to another enzyme for further detoxification back to alcohol and re-entry back into the visual cycle. This is the normal processing of vitamin A in a healthy, unaffected eye. In patients with Stargardt disease, there are genetic mutations that affect the function of the ABCA4 protein. Consequently, the aldehyde cannot be removed from the retina as efficiently. It lingers within the retina where it can condense upon itself, forming diamonds of vitamin A that are called dysretinoids. These are the toxic byproducts of vitamin A that I spoke about earlier. The most abundant dysretinoid that has been identified in ocular tissue, human ocular tissue, is known as A2E. In cell-based culture studies and in animal studies, this molecule has been shown to kill retinol tissue through diverse mechanisms. It is very well accepted in the medical and scientific communities that in Stargardt disease, the sole reason for pathophysiology and eventual blindness is because of the accumulation of A2E and related dysretinoid molecules. What does Tinlarebant do? Tinlarebant works in the liver to compete with retinol for binding to RBP4. It does not allow that larger protein transthyretin to bind to it. Consequently, what you get is a very small complex of retinol binding protein 4 with our drug bound to it. Because it is so small, it gets readily filtered through the kidney. The net effect of that reduction of retinal binding protein 4 through the kidney is, of course, a reduction in the overall level of the wild-type complex that's normally used to deliver vitamin A to the eye. Of course, when that complex goes down, the level of it, the amount of vitamin A going into the eye goes down. All the retinoids cascading downstream, including the toxic dysretinoids, will also be reduced. This is the mechanism where we intend to slow disease progression by going after these toxic dysretinoids, which are early stage pathogens, if you will, in the progression and the onset and progression of Stargardt disease. An interesting point about these dysretinoids is because of their vitamin A composition, they actually fluoresce. You can actually see them in clinical pictures when you look at the progression of the disease. Here's a good example of a patient with Stargardt disease on the top. We also have a patient with geographic atrophy on the bottom. I mentioned earlier that these diseases share a common pathophysiology. I'm going to show you that right now. We're looking at the baseline image of this Stargardt patient in the upper left-hand image. We're going over about four and a half years annually, looking at the growth of those lesions. These black areas you see here in the image are dead retina. This is tissue that's never going to be restored. Peripheral to that tissue, you see this intense zone of autofluorescence. That's where the dysretinoids are. As you traverse through time over 12, 24, and finally out to about 57 months, what you see is those areas of atrophy, these black areas, spread into the autofluorescent zone. The autofluorescent zone continues to expand outward in a centrifugal manner to accommodate that dead retina. What this tells you is that wherever you see dysretinoids, you will eventually see dead retina. We see a similar story here in the patient with geographic atrophy on the bottom. You see here in the baseline image, a small central area of atrophy. Peripheral to that atrophy, you see these little punctate areas of autofluorescence, sort of like satellites around a planet. These little punctate areas of autofluorescence are dysretinoids. If you look carefully over time, you see they gradually become atrophic lesions. They go from little bright autofluorescent dots to these darkened areas of atrophy. Again, wherever the dysretinoids occur, that's where you can see retinal pathology. We believe that in both these diseases, we can go after these dysretinoids with the same therapeutic that is Tinlarebant to reduce retinal delivery to the eye to get rid of these dysretinoids. I want to talk a little bit about the phase two Stargardt trial that we completed recently. I mentioned it was a study that enrolled 13 adolescent subjects. These subjects came in with the early stage of disease. They only have the autofluorescent lesions. They have not yet converted to atrophy. The autofluorescent lesions are referred to as QDAF. That stands for questionably decreased autofluorescence. They convert to the atrophic lesions, which are known as definitely decreased autofluorescence. In this study, we want to watch the conversion of one lesion type to the other. More importantly, in those subjects that convert to these atrophic lesions, those DDAF lesions, we want to watch the conversion, the growth of those lesions, because that is the endpoint for approval in Stargardt disease and also in GA to slow the growth of the atrophic lesions. You can see here the various criteria for the study, the location, the open label nature, two-year duration. Of course, in the very bottom, the key inclusion criteria. As I said, these are adolescent Stargardt subjects that have been clinically and medically confirmed with Stargardt disease. Here is a pharmacokinetic and pharmacodynamic profile of the drug. Through dose finding studies, we have determined that a 5 milligram daily dose is effective to reduce retinal binding protein 4 by about a mean 80% reduction relative to baseline. We put here in the graph the target threshold of greater than or equal to 70% RBP4 reduction because in a prior clinical study with a different drug that was also an RBP4 antagonist, we determined in GA patients that a 70% reduction was effective to slow lesion growth. We could actually see that in patients that achieved this reduction. In our trials, we are dosing to a level of RBP4 reduction that's at least a 70% reduction or more. You can see here with our 5 mg daily dose, we're getting about a mean 80% reduction. This is RBP4 shown in the red line, reduction expressed as a percent decrease from baseline. The blue line shows you the increase of Tinlarebant. You see a very nice correspondence between increase of Tinlarebant and decrease of retinal binding protein 4. Once Tinlarebant reaches a steady state, we also see a steady state reduction of RBP4 until we withdraw the drug at month 24. You see a rapid clearance of Tinlarebant from blood and a rapid rebounding of retinal binding protein 4 right back toward the baseline value, demonstrating the reversibility of the pharmacodynamic effect. I mentioned in this trial, we want to look at the growth of lesions, one lesion type to the other. For those patients that grow these incident atrophic DDAF lesions, we want to measure the growth of those lesions. I want to start by saying that over two years, despite all of these kids having very severe genetic profiles, five of them, that's 42% of the cohort, never converted to atrophic lesions. In fact, their disease was essentially static. In those seven subjects that did spawn those atrophic lesions, which is shown here in the red line, the growth of those lesions was significantly lower than the growth rate predicted by the natural history study of Stargardt patients of the same age range. This data was gathered by our CMO, Dr. Hendrik Scholl. It shows very clearly, and these are from the ProgStar data I mentioned earlier. You can see here we're getting about a halving of the growth rate in our Tinlarebant cohort versus the adolescent cohort from natural history. We believe we're seeing a very prominent treatment effect against lesion growth, which again is the primary endpoint for approval. You can see here on the right-hand side, the tabulated numerical changes in lesion growth. Now we'll go to the Dragon trial. This is our phase three placebo-controlled study, which enrolled 104 patients with Stargardt disease. In this study, we had pre-specified an interim analysis in which our DSMB, the independent DSMB, would take an unmasked look at the safety and efficacy data during the interim. The interim analysis, by the way, was triggered when the last patient reached the 12-month visit. They had their 12-month assessment vis-à-vis the imaging, et cetera, the clinical lab work, all that stuff. They looked at the data. If in fact there was a trend for efficacy that fell within what is so-called, statistically speaking, a promising zone, that is a window of conditional power, which indicated a trend for efficacy, we would be allowed to add 30 additional subjects. We would do that to increase our chances for observing a statistically significant effect in lesion growth by the end of the two-year study. The DSMB, I should back up a little bit, tell you a little bit about the trial designs in Dragon and Dragon 2. I mentioned it earlier. These trials are designed identically. There are only three real differences in the trial design. These are shown in the top three rows: the number of subjects, the global nature in Dragon versus the more specific geographies in Dragon 2, and of course, the randomization. You can see here, these again are adolescent, clinically and medically confirmed subjects with Stargardt disease. Here is the interim, some of the interim data that the DSMB provided for us. You can see on the left-hand side, the mean age, height, and weight of these kids. These are teens and preteen subjects of normal height and weight. You can see on the right-hand side, the distribution for sex, approximately 62% male, 38% female. Of course, the race distribution, as I mentioned, favors the Asian population because we did heavily recruit in Asian countries. Of course, following then the Caucasian patients from Europe and North America. As I said, the DSMB took a look at the data to determine whether or not there was a trend for efficacy. If there was, they would tell us to admit up to 30 patients to improve statistical significance for the second year of the study. If they told us not to add patients, we could either be on the futile side of that what's called promising zone or on the overly efficacious side, which would be very promising. After the DSMB looked at the data, they told us that no modification of the study is required and that we should continue without a sample size increase. That tells us we're not within the promising zone. We're on one of the other sides of that. They added a comment, and this is very important, that they recommended that we submit the data for further regulatory review for drug approval. That comment would not have been made if in fact we're on the futile side of that promising zone. For we're very optimistic and encouraged that in fact we're probably on the overly efficacious side. We've been doing just as the DSMB has suggested. We're out basically getting meetings with various global regulatory authorities to share this data and communicate the relevance of this study for Stargardt disease patients. Importantly, the safety profile was also outstanding. The drug was very safe and well tolerated. The overall withdrawal rate at the time of the interim, which by the way, I should mention, there's approximately 70%-75% of the data available at the interim because of the staggered nature of the trial. Over that period of time, less than 10% of subjects withdrew from the study. This is blended. We don't know which was due to active or which was due to placebo. The fact that you're seeing overall less than a 10% dropout when roughly 75% of the data has been collected is very promising. More importantly, the ocular adverse events, which we anticipated because of this mechanism of action, the withdrawal due to those ocular adverse events was less than 4%. Only four subjects withdrew because of ocular adverse events. Visual acuity was stabilized in the majority of subjects with a loss of less than three letters under both standard and low lumen. We are very encouraged by both the efficacy and safety profile of the drug. Here is an overview of the treatment emergent adverse events. The ocular aids that we anticipated based upon the mechanism of action are called xanthopsia. This is basically a light-triggered event. When patients are exposed suddenly to a bright light, there will be this startling of color in the visual field, which lasts a few seconds to maybe a minute. It is mild. It is transient. It is due to having limited vitamin A in the eye. Delayed dark adaptation is the opposite manifestation. This triggers when subjects transition from a bright light to a darkened environment. They will require more time to adapt to that diminished light environment because again, they have less vitamin A in their eye. This is not night blindness. It is a delay in the ability to fully accommodate to dim light. Night vision impairment is a more severe exacerbation of delayed dark adaptation. This is when the dark adaptation can go as much as, let's say, 20 minutes. You can see on the right-hand side the frequency and number of patients getting those AEs. Finally, a non-ocular AE that we get is headache. This can happen when patients are using or straining to use their visual acuity while they're experiencing these ocular AEs. This is the visual acuity data that we have for the trial. As I said, the visual acuity was stabilized under both standard and low luminance. On the left-hand side, you see standard luminance. Basically, this is data from an eye chart. Patients are reading letters on an eye chart. And depending on what the number of letters they score, they get a score. We're comparing that score to baseline. You can see over time, there really is no change in these patients' scores over 24 months. This is stabilized visual acuity. This is important because Stargardt patients do gradually lose vision. The fact that we're stabilizing it over two years is actually a very promising trend, especially when you consider that we've likely slowed lesion growth based upon what the DSMB has signaled. On the right-hand side, you see the same type of scoring, but now the light luminance is reduced by about 100-fold. We're looking at the ability of these patients to see under low luminance light. You can see initially there's about a two-letter drop. But then subsequent to that initial drop, there's no further drop in low luminance throughout the rest of the duration of the study. So again, we've stabilized vision under both standard and low luminance condition. And finally, a little bit about our phase three trial in geographic atrophy called Phoenix. It's important to note that in geographic atrophy, these are older patients, heavier, of course. Retinal binding protein 4 levels actually increase in patients that are getting older and fatter. But it turns out from a PKPD study we did in healthy adult volunteers intended to match the higher age range and BMI of GA patients, we can use the same 5 milligram dose because it produces the same pharmacokinetic and pharmacodynamic profile as it does in adolescent subjects. So we can use the same dose. It's the same endpoint. It's the same trial duration. In fact, everything about the clinical trial design in Phoenix is identical to the Stargardt disease trials, except for two things. One, of course, the indication being GA instead of Stargardt disease. Two, the higher number of subjects will be enrolling based upon the higher prevalence of the disease in the patient population. The reason this is important is because if in fact we have promising signals coming out of our Stargardt trial, we believe that will predict promising outcomes for the GA trial because again, everything about these trials is essentially identical. Of course, the mechanism of action we believe will be effective in both these indications because in both patient populations, we do see the accumulation of toxic byproducts of vitamin A as implicated in disease progression. With that, I'll close. I'll thank you for your attention. Of course, I welcome any questions. There is a question about, are we using AI in our drug development or clinical trial process? No, we are not. There are advantages for using AI for doing some enrolling and screening. We did not need that this time. You never know, in the future that could be implemented. There is another question asking, will Tinlarebant be delivered via intravitreal injections? The answer is no. That is one of our differentiators. We are an oral therapeutic. We do not need to get into the eye. In fact, Tinlarebant never gets into the eye, never crosses the blood retinal barrier. Its effect is purely systemic. It reduces vitamin A delivery to the eye. Once vitamin A delivery to the eye goes down, all the other processes are natural. No, we will not be delivering intraocular intravitreal injections of our drug. We have another question regarding the capital position and cash burn rate. Will we need to raise capital to complete the Tinlarebant trials? The answer is no. We have about a four-year cash runway. We have enough cash to complete all of the clinical trials I told you about today, except for a second confirmatory GA trial. We have enough to do all the two phase III trials, Dragon 1 and Dragon 2, as well as the Phoenix trial in GA with no problem. Another question, can you outline the road to profitability in base and best-case scenario? Yes. We expect to get approval in Stargardt disease in 2026 and certainly no later than 2027 based upon the data as it looks now. There will be premium pricing for Stargardt disease for a number of years until GA gets approved. If and when GA gets approved, of course, that price will come down. We see profitability in the short term, certainly. Because of the large-sized market of GA, it will not hurt us that much to reduce the price down to match basically a Sifovry or a Zirva intravitreal injection. Anywhere between, let's say, $20,000-$30,000 a year. We see both in base-case scenarios where we have high profitability with Stargardt and best-case scenario where we have profitability from both these indications, it looks very, very promising for Belite Bio. Another question about milestones to the end of 2025. Quite frankly, the primary focus of our role here at Belite Bio is execution. We believe the data are going to speak for itself. We're already engaging, as I mentioned, global regulatory authorities in the U.S., Japan, China, Switzerland, Australia, and of course, Europe to share the interim analysis data. There's a small unmasked team on our side that's communicating the data to these regulatory authorities. We're focused on getting an expedited approval by the end of 2025, either through approval of the first phase three study or approval of first phase three study with a commitment to complete the second study, Dragon 2, and provide that as a confirmatory phase two three study to support our drug approval applications, both NDA, PMDA, as well as MAA in Europe. I think we've reached the end of the questions there. Let me just take a quick look. Yep. I believe that's it for now. Happy to wait for any additional stragglers.