Hi, everyone, and thank you for joining us today to kick off, or not to kick off, actually. We're in the midday already. It's our pleasure to have Belite Bio with us today, who's developing an oral treatment for Stargardt disease and expanding into GA. We're happy to have CMO Hendrik Scholl with us, and CFO Hao-Yuan Chuang. We've watched the Belite story from the sidelines for some time, and it's great to see you approaching the finish line now on some of these difficult retinal conditions, Stargardt specifically, which is a rare disease, but also expanding into GA. Why don't I give you the floor for a quick intro to your lead programs and how you uniquely fit into these diseases, and we'll launch into questions from there.
I'm happy to kick that off. It's a pleasure to be at the Stifel meeting. My name is Hendrik Scholl. I'm a Retinal Specialist by training, trained in Germany and the U.K. Was professor at the Johns Hopkins University, Wilmer Eye Institute. At the time, we developed a large worldwide natural history study that informed the clinical development for Stargardt disease. At the time, I had no clue that I would join Belite Bio later. I have worked with Belite in various capacities, including Scientific Advisory Board Member and DSMB member for many years, and two years ago, joined Belite. Belite is in the process to develop new therapies for an unmet medical need, specifically in the field of retina. We have two main indications, namely Stargardt disease, where there's absolutely no treatment available. I can explain about Stargardt disease in a little bit.
A very similar disease, maybe the older brother if you will, the very old brother, namely geographic atrophy, affecting people at an old age, with a very similar disease type, namely macular degeneration, where simply the photoreceptors in the RPE degenerate. This is the situation that we are in. We completed a registration trial called DRAGON in Stargardt disease and presented the top-line data December 1st, and since then prepare for submission of our NDA to the FDA, which will be completed within a month from now. We anticipate approval, going through a priority review with the FDA in the first quarter of next year. This will be the first therapy ever for Stargardt disease, which, as I mentioned, is a completely unmet medical need. It's a severe disease, and it's a pediatric disease. It's not so rare.
It's not as rare as, for example, RPE65-associated retinal dystrophy. It's 70 times as prevalent, affecting about 50,000 patients in the United States.
Great. Thanks. Maybe we can, just to get everyone on the same page, talk about the mechanism of tinlarebant and why it's so well-suited for Stargardt and GA. It's a very different approach than the complement inhibitors and some other drugs that are in development, and even unique that it's oral. That's the other benefit there, but maybe you can talk to us about that mechanism.
I would, to just support what I say, share a few slides on the mechanism of action. Tinlarebant indeed is an oral compound. It acts on retinol delivery to the eye. What we see here is the so-called visual cycle, which is quite an important process. It's a miracle of nature. This is how we are connected to the world. To be clear, rhodopsin, it's a chemical molecule that you see on top, and this allows us to see through a chemical reaction. If there wasn't the so-called visual cycle, the regeneration of the photopigment, we could only see once in our life at birth, and then it would be dark. This visual pigment needs to be regenerated, constantly regenerated. It's made out of retinol or vitamin A.
Retinol can enter the eye through the RBP receptor, and this is pretty much eye exclusive, meaning that vitamin A getting through the eye needs to be bound to RBP4 in the systemic circulation, and as such, can then enter the eye and play a role in the visual cycle and allow the visual pigment to be regenerated. In Stargardt disease, ABCA4, which is part of this visual cycle, is dysfunctional and will not allow to flip out the waste out of the cell, if you will, and this leads to the buildup of so-called toxic bisretinoids, including A2E, that are toxic to the photoreceptors and the retinal pigment epithelium and lead to photoreceptor and RPE cell degeneration. This exactly correlates with blindness, because if there is no photoreceptor, there is no vision. With tinlarebant is an oral drug.
In the systemic circulation, binds to RBP4, then RBP4 is no longer available, meaning that we reduce delivery of retinol or vitamin A to the photoreceptors, and that in turn leads to a reduction of the toxic buildup in the RPE and photoreceptor. This is the mechanism of action. Yeah. It's extremely specific because the eye expresses this RBP4 receptor. You have a systemic drug, but it has a very specific effect on the eye.
That's great. You have tested in DRAGON, and you tested a 5 mg oral dose once daily. Again, it's unique that an oral drug can reach the retina and doesn't require intravitreal injection. What is some of the earlier work that you did to give you comfort in the dose that you chose, and that it was sufficient to impact the retina?
I'm not sure if it still shows. Difficult to hear with my. Is it still showing a slide here?
No.
Okay. I share again. Let's see if that does me the favor. Okay, I hope this can now be seen. I go to presentation mode. Obviously we have run single and multiple ascending dose studies before deciding on the final dose of 5 mg per day. We give 5 mg per day, and this was done first in adolescent subjects, but also to prepare for the PHOENIX trial in AMD in the elderly. This is actually data on the elderly. It's exactly the same, meaning that if you give 5 mg per day, you see that the target, namely RBP4, is being reduced underneath the threshold of 70%, which is our target, within a few days. When you stay at 5 mg per day, it stays there, very comfortable at -80%, meaning only 20% remaining.
Then when you discontinue tinlarebant, then it goes very quickly back to baseline. Also important because it just shows it is completely reversible. Anything that happens is reversible, which is important. The safety profile in the registration trial was extremely favorable. Still, if there was ever a problem, you simply discontinue the drug and it gets back to normal. This 5 mg dose has been used in the phase II clinical trial in both of our phase III clinical trials in Stargardt disease, exactly the same dose, and same dose is being used in the PHOENIX trial in geographic atrophy.
Okay, great. Obviously, this correlates into progression of lesion. Are you seeing any kind of reversal, or is this primarily to slow the disease progression?
I think it's important to say that it's absolutely impossible to regenerate photoreceptors. There has been research over decades with stem cell therapy, trying to regrow photoreceptors in the retina, but this has not been successful. There's nothing really on the horizon that would indicate that this is going to change any time soon, meaning that any neurodegenerative disease, that's not only true in retina, it's also true in neurology, including Alzheimer's disease or Parkinson's disease. The neuronal cells that have been lost are lost, they are lost forever. The aim of the therapy is to slow down, ideally stop progression. This is something that has been tried over the years. The first successes in GA with the two injectables, with treatment effect sizes of 13%, 21%, and 14% when we look at the OAKS, DERBY, and GATHER2 trials.
Now, with our compound in Stargardt disease, a therapeutic effect of 36%. It's two to three times as large. It's not 100%, meaning that even under therapy, patients would not regain vision.
I see. I guess by nature of your administration, are you looking at a different patient population than was being investigated in the earlier GA trials or what's being investigated in other GA trials right now? Have you tried to go early in treatment, or is it hard to get a good separation when you go earlier in treatment? How are you thinking about the population who'd be most appropriate for this treatment?
Now we talk AMD, obviously, so we switch from Stargardt disease.
Oh, I'm sorry, yes. AMD
Yeah.
I guess it's somewhat relevant.
Yeah. We have to be clear about what we discuss.
Yeah.
About what we discuss.
Yeah. Sorry about that.
Yeah, do you want to discuss AMD, or want to discuss Stargardt disease? I'm happy with both.
Well, I guess it ends up being a similar question because you have Stargardts where you are going to have patients who have not advanced to the point where they're seeing like major vision obstruction. Could you potentially have this treatment earlier in disease?
Yeah. The first answer is yes. The second is the following, and that actually is the same in AMD also.
Yeah.
It's just that in Stargardt disease, when the disease kicks in, it's a continuous process, and vision is getting worse and worse. The question is, what is the population that participated in the DRAGON trial in terms of early or intermediate or late disease? There's no agreed nomenclature for Stargardt disease, but it is clear that Stargardt disease is a progressive disease. It starts relatively early, and patients have 60 to 80 years ahead of them after the diagnosis of Stargardt disease, meaning that because it's progressive, regardless of when you start the therapy, you will always have an effect on vision. Some people think when you reach the level of legal blindness, then it's too late. It's not too late.
ABCA4 is one of the very few disease genes in this space, and retina is the leading tissue in the human body being affected by mutations and dysfunctional genes. One of the few disease genes that can lead to complete blindness, so no light perception at all.
There's a big difference between legal blindness, which is the level of 20/200 in the better seeing eye, according to the U.S. legislation, and no vision at all. In our opinion, tinlarebant will be helpful at any stage. It will help in the future, patients to retain 20/20 vision for longer if they still have 20/20 vision. When they have lost vision, allowing them to lose vision at a much slower pace. Third, it will allow patients to not go completely blind. Even if they have lost significant vision in the past. I think this is important. AMD is a bit different. In AMD, we have a continuum of what's called early AMD, intermediate AMD, and then we have two forms of late AMD, wet AMD and geographic atrophy or late dry AMD.
Intermediate AMD is not a target in the U.S. currently because it's not being recognized as a disease as such. It's a precursor of the two forms of late AMD. Likely, if tinlarebant shows impact on slowing down lesion growth in GA, it will likely also prevent the formation of geographic atrophy, meaning it could also work in intermediate AMD. That is something that we will explore in the future.
Okay, great. Just in terms of the regulatory endpoint, the lesion size for Stargardt measured by the DDAF. Is that just a difference in nomenclature between Stargardt and GA? You're essentially measuring the same thing, or are you looking at something a little bit different?
Yeah. I share that again, and obviously, this can be seen. It's actually a nomenclature thing.
Mm-hmm. Yeah.
When you look at the fundus autofluorescence image here on the left, in the very center, you have this area that is relatively black, and we call that DDAF, definitely decreased autofluorescence. This is actually a term that was coined by the ProgStar study group that I led a decade ago, meaning that their autofluorescence is much reduced, similar to the optic nerve head, ONH, that is shown on the left of the image. These areas can, number one, be detected and be outlined. The growth of these areas, and you see the growth over two years when you go from the left to the right panel, on the expense of the grayish area that you see to the right of the lesion, can be precisely measured and is an approvable outcome measure by the FDA. Is that the same as geographic atrophy in geographic atrophy?
Because in geographic atrophy, we do not have that sophisticated nomenclature, but the answer is yes. It shows an area of much-reduced autofluorescence, and in this area, the retinal pigment epithelial cells and the photoreceptors have died away. It can be very precisely measured. This is shown here. We have shown that in the ProgStar Study Group, that over this period of two years, it's exactly the same period as the DRAGON trial. The progression rate was only 40% or so compared to AMD. AMD progresses faster.
Namely, 0.74 sq m per year. This is the endpoint that was used in the DRAGON trial.
Okay, great. That's very helpful. Just to follow on the DRAGON trial. You also have a DRAGON II trial. Can you talk about this second trial and filing plans, and how rapidly you can move this forward?
Yeah. I think it's important to understand that DRAGON II is not the second registration trial to seek registration in the U.S. It has a completely different origin. We seek approval based on one trial, namely DRAGON. This is being agreed upon with the FDA in all our discussions with the FDA, although they would never commit that they would grant approval on one trial because they say it's a review issue. We anticipate that with this one trial, plus the confirmatory evidence that we will submit, we will get approval for Stargardt disease, of tinlarebant for Stargardt disease in patients being 12 years or older. The DRAGON II trial originates with Japan providing Belite Bio with a Sakigake designation. Japan was too late to participate in the DRAGON trial, but definitely wants to bring tinlarebant to the market as the first country in the world.
That's the explicit meaning of the Sakigake, or breakthrough designation.
Obviously, they want to see data in Japanese patients. That led to the birth of the DRAGON II trial. Because we invited patients to participate in Japan in this registration trial with essentially the same design as the DRAGON I, we also wanted patients waiting for participating in the trial in the U.S. and U.K. to also participate. Therefore, we launched DRAGON II as a registration trial in Japan and allow sites in the U.S. and U.K. to also participate. That led to 73 patients participating in the DRAGON II trial.
Great. Will the FDA want to pull any of that data into your filing, or it's completely separate?
It's completely separate.
Okay.
Obviously, the integrated summary of safety includes all patients being exposed to tinlarebant, and that includes all patients being exposed in the phase I, the phase II, DRAGON I, DRAGON II, and even PHOENIX. Because in the PHOENIX trial, we have 530 patients.
Yeah.
Many patients have completed the first year, they were exposed for more than a year to tinlarebant 5 mg per day, meaning we have a very extensive database for safety.
out of all of these trials.
No, that's great. Okay. In the phase III, you showed clear separation and high statistical significance. Was there any other?
Yes
benchmark for clinical meaningfulness that you're comparing to? Are there any other secondary measures of note to show disease modification as well?
Yeah. I just remember when I showed this slide on DDAF, the other area close to DDAF, which was also decreased in signal but not as black, is called questionably decreased autofluorescence. Not as black, simply questionable. Questionably decreased autofluorescence. We also looked in the DRAGON at areas, when you sum together, called DAF, decreased autofluorescence. Decreased autofluorescence means that areas are included where the pigment, the melanin, the photoreceptor had not completely died away. Meaning, if we find a treatment effect on DAF, that would mean that tinlarebant not only allows to slow down the dying away of cells, but also cells becoming sick that then later would die away because they turn into DDAF.
That was the question, therefore, it was our key secondary endpoint. What we found was that the treatment effect on DAF was essentially the same size, it was 34%, and it also reached statistical significance, meaning that we have a key secondary endpoint that shows strong confirmatory evidence of what we find in the primary endpoint. We are not used to that anymore in ophthalmology. Over the last two decades, pretty much everything was monocular. Injections-
surgery-
Yeah
gene therapy, what have you, devices even. Now with Belite Bio, we have now a systemic drug. It obviously should work on the other eye.
Yeah.
In a clinical trial, you have to optimize everything for your study eye, but you can also look at the fellow eye. We found statistically significant treatment effects in the fellow eye for both the primary endpoint and the key secondary endpoint. In our opinion, there can't be any stronger confirmatory evidence than what we currently have.
Okay. No, that's good to know. I imagine as a systemic drug, you are affecting the fellow eye. How do you decide?
We should.
Yeah.
We should. I think it's nice.
Yeah. There's no reason you shouldn't.
in AMD, it's more complicated in AMD because I just mentioned earlier, there are two forms of late AMD. There's also wet AMD. The other eye may have or could develop wet AMD over the trial. In Stargardt disease, the two eyes are fairly similar. Not identical, though. Meaning that in the fellow eye, some patients had no lesion or a lesion that was too large to be included in the analysis.
Yes.
Meaning there was much higher variability. I mentioned that because it is not a given that you would find a statistically significant effect also in the fellow eye, but we did.
Yeah.
It simply has to do with this very high efficacy that was found in the trial, that that was also detectable in the fellow eye.
Okay. That's good to know. Just one question on the safety. Stargardt's patients obviously are much younger than GA patients. What are some of the biggest SAE concerns given that they're going to be taking this presumably their entire lives for 60 to 80 years, as you said?
Absolutely. Actually for a very, very long time.
Okay.
Here's the thing. At the end of the day, it will simply be a patient education issue. By the way, it's the same with wet AMD. You cannot tell your patient, "Stop by a couple of times, we inject, you'll be fine." You have to tell the patient, "You have a chronic disease. This is how the medication works, that we have to inject you every month, and you likely will be fine." With tinlarebant, you have to tell your patient, "You have to stay on that drug chronically. You have to take that pill every day." That pill leads to some, not a complete, but some reduction of the substrate for the visual cycle, meaning that when you change from bright to dark, it will take a bit longer. Just a bit longer. By the way, patients don't care that much. What patients care about is cone vision.
They want to read, they want to recognize faces. They want to be mobile under normal light conditions. This is really what patients care about. You have to educate them. The final threshold, how sensitive you can get in the dark is not affected. How much time it may take to get to that threshold is affected. It's slower. That's the mechanism of action. The cone system also can, as I say, show adverse events. It was called xanthopsia or Essentially, it's a discoloration of the visual scene when you get from dark to light for seconds or minutes. Meaning that when I look at the screen, I take the tinlarebant, I come out of the dark, I look at the screen, I can see your face, I can see your name, but maybe you would appear a bit yellowish for a couple of seconds or minutes.
Is that important? No, you have to educate the patients. You have to tell them, "This is what can happen. You have to be prepared." They do not freak out, it's all fine. These are the two main side effects with the tinlarebant. Systemically, there's essentially none. There was essentially none. We had six SAEs, four of the six were in the placebo group.
Got it. Okay, that's good to know. Just moving to GA really quickly.
Yep.
What are the global filing plans? How are you thinking about the study endpoints? I know for GA, for the complement inhibitors, they were never approved in the U.S. because there was no establishment of functional benefit. What are your thoughts around that and the opportunities for tinlarebant for GA?
It will very much depend on the effect size that we find in the trial. It may get us in a very convoluted space. Because there may be several thresholds if we just as good as the injectables, but we are oral. What about pricing? We don't want to affect our primary market on Stargardt with the same compound. What happens if we have triple the effect of the current injectables? That may have an impact on future pricing, and that will lead to a completely different calculation. For the latter, I would just defer to Hao-Yuan Chuang, our Chief Financial Officer, to provide you with a bit more color on that.
Definitely, we are devoted to develop treatment for both Stargardt and GA. We just need to see more data to be able to evaluate the commercial strategy.
I see. Okay. For the GA program, are you thinking about a different patient population than what's historically been studied?
The answer is yes and no. Our inclusion criteria indicate a bit earlier disease stage than what was studied with the injectables. Essentially, anything that can slow down neurodegeneration would be helpful. Likely, by the way, the two approaches are complementary. Meaning, at the end of the day, you could take both. It's just that an oral compound is much more convenient to getting monthly injections into your eye.
What would be that patient who would get both?
If I was a patient, I would go for both. I would definitely go for the tinlarebant. I may be biased, but the oral compound is so much more convenient. It will take years. There will be investigator-initiated studies, I'm sure.
Yeah
to explore the combined effect of those two approaches, but this will take years. For this specific moment in time, it is pretty clear that if we find a significant efficacy, this will be the standard of care because it is an oral compound.
Okay. We're just about out of time. Thank you for touching on the commercial plans. One last question. If you do in fact see the major effects in GA, you're considering keeping the same dose, or are you potentially creating a new brand and thinking about how that affects the other population, the Stargardt population?
Well, I think, again, it's going to be data-driven. It's probably too early for us to comment about that yet.
Okay. Great. Well, unfortunately, we are out of time, but thank you for that overview. That has been very helpful and exciting to see something reaching the end stage for Stargardts finally.
Thank you. It was a pleasure. Thank you for having us.
Great. Thank you.
Thank you.