Belite Bio is a clinical stage biopharmaceutical company focused on developing novel therapeutics for retinal diseases, including Stargardt disease and geographic atrophy. The company's lead asset is tinlarebant, LBS-008, a novel oral RBP4 antagonist that has initiated rolling NDA submission. Thank you both for joining.
Thank you.
Thank you for having us.
All right. Well, to start, could you walk us through the mechanism of action and why targeting RBP4 is the right approach for both Stargardt and GA?
Happy to. Both diseases are extremely similar in phenotypic expression. It's just that Stargardt disease starts very early. It's a pediatric or disease of the adolescent patient. While age-related macular degeneration, therefore the name, is age-related and is a disease of the elderly. Both diseases affect the macula. In both diseases, what you observe clinically is that deposition of cytotoxic bisretinoids influence the disease progression. We can see that clinically when we use fundus autofluorescence imaging, which is a standard of care imaging in a typical retinal clinic. We see these compounds lighting up underneath the retina. These are the drivers of disease.
We have seen that in geographic atrophy already at the time with Frank Holz in 2007 in the largest natural history study at the time in age-related macular degeneration, that if you have increased autofluorescence around lesions, this is a major driver of disease progression. When we get to Stargardt, in Stargardt, we have a very precise understanding of the disease. The disease is monogenic. It's inherited. It is due to mutations in a gene called ABCA4. It's an ABC transporter, A4 subfamily. ABCA4 plays an important role in the so-called visual cycle. What's the visual cycle? Very briefly, I keep it brief, but it's complex. The visual cycle is very well understood. The visual cycle is the regeneration of the photopigment. If we look at light, if we sense light, our photopigment can sense that.
In that moment, it falls apart, so to speak, and it liberates all-trans retinal into the photoreceptor. All-trans retinal is derived from vitamin A. It's very toxic and has the propensity to react with itself to form dimers or bisretinoids, the most important being A2E. ABCA4 function as a flipase to flip that out of the cell into the visual cycle to get it regenerated. If ABCA4 is dysfunctional in Stargardt disease, all-trans retinal accumulates. It's very toxic itself, and then reacts with itself and forms these A2E depositions in the RPE, in the support layer of the retina, driving the toxicity in the disease. Now the question is, what can we do about that? Number 1, we may repair the gene. This is not our approach. Which is very complicated to accomplish that. You could also try to disrupt the visual cycle.
This has been tried. You would kind of induce a retinal disease if you did that. We do not do that. Tinlarebant is not a visual cycle modulator. Tinlarebant aims to reduce vitamin A in the photoreceptors. It reduces substrate of the visual cycle, therefore having less compound to be developed into toxic depositions in the RPE, and therefore having a therapeutic effect. How do you now reduce vitamin A content in the circulation and only have an effect in the eye? Is that possible? The answer is yes, and the reason is the RBP receptor that is exclusively expressed in the eye. Only the eye is dependent on sucking in vitamin A through the so-called RBP receptor, retinal binding protein receptor.
Therefore, we can target, with an RBP antagonist, vitamin A bound to RBP4, therefore reducing RBP retinal in the circulation, and thereby only reducing vitamin A in the photoreceptors, but not in the systemic circulation because the other organs, extrahepatic organs, have other routes of access to vitamin A independent of RBP4. Therefore, tinlarebant allows to very precisely, almost with the precision of gene therapy because of the RBP4 receptor, to reduce the content of vitamin A in the photoreceptors and therefore reduce toxicity in the eye without having any safety concern for any other organ.
Great. Makes a lot of sense. Maybe just flipping to the regulatory pathway for a quick second. You recently initiated a rolling NDA submission for Stargardt. What are the key data points from the DRAGON trial that gives you confidence in the filing, particularly speaking to the primary endpoint of lesion growth?
Yeah. Having a primary endpoint for an indication like Stargardt is key, and it was very kind of you to mention the fact that I led the largest natural history study in the space called ProgStar, progression of Stargardt disease. ProgStar was instrumental to establish the natural history, obviously, in Stargardt disease and develop and implement a primary endpoint in consultation with the FDA at the time. In 2011, we met with the FDA in person at the time in Bethesda. The FDA recommended for the drug development in Stargardt to implement a natural history study and to look at outcome measures very similar to, at the time, was done for geographic atrophy, namely to establish atrophy measured from funded autofluorescence images as the primary endpoint for clinical trials in geographic atrophy. So that's exactly what we did at the time.
We implemented the ProgStar study when I was professor at the Wilmer Institute at Johns Hopkins as a worldwide natural history study, and we measured lesions on autofluorescence images, multicenter in a central reading center over the period of two years in six monthly visits. And we could establish DDAF, definitely decreased autofluorescence, very similar to what we see in GA as GA lesions on autofluorescence images, as lesions that can be precisely measured over time, and they enlarge at a predictable rate. Also important, we could show that with data from ProgStar are correlated with visual function long term. The limitation of visual function, if I may mention that, because you could ask, why don't you simply go for visual acuity, is that visual acuity loss also established in ProgStar is very slow. It's only half a letter per year on average in Stargardt disease.
With an eight-letter intersession variability, it's almost impossible to find a visual acuity delta deficit over time in a reasonable amount of time in a reasonable cohort of patients. Therefore, we believe that with the FDA, we found an approvable endpoint that will lead to success in our market authorization application with the FDA. In the DRAGON trial, we used DDAF as the primary endpoint and found that with five milligrams per day of tinlarebant, we found a progression rate of only 0.38 square millimeters per year, while the placebo group showed a progression rate of 0.59 square millimeters per year. This was highly statistically significant using an unstructured covariance matrix that was pre-specified. We found a p-value of 0.0033 and a much better covariance structure for the dataset we had also discussed with our DSMB at the time, namely an autoregressive model covariance structure.
This allowed to arrive at a p-value that was smaller than 0.0001, so triple zero one. This was highly statistically significant. Importantly, in the DRAGON trial, our key secondary endpoint decreased autofluorescence, which indicates tissue that is not as degenerated, as unhealthy, if you will, as the tissue within DDAF lesions also show the treatment effect that also reached statistical significance. Therefore, we are very confident that with that efficacy signal, we have a strong dataset for our application. I would also like to mention the safety data. With tinlarebant, we had only six SAEs in our clinical trial of 104 Stargardt subjects over two years. Four of them, by the way, in the placebo group. Only two of the SAEs in the treatment group. None of them was related to the treatment. The ocular AEs were related to the mechanism of action.
Mostly were mild, mostly resolved while on treatment, meaning that with that combination of strong efficacy, excellent safety and tolerability, and supporting evidence from natural history data from the ProgStar study, we should be in a very strong position with our market authorization based on only one trial.
Got it. Very helpful. Do you think as you think about physicians and the key endpoint that they think about, are there any differences between how they might view that? You mentioned visual acuity versus the regulators.
Yeah. In a typical retina clinic nowadays, clinical care is heavily image based. These clinical images, such as fundus photography, OCT, autofluorescence imaging, are being used for wet AMD, for dry AMD, but also for IRDs. Currently, if a patient with an inherited macular degeneration comes into a doctor's office, the typical situation is that the patient would get retinal photography, autofluorescence, and OCT imaging, and that would remain exactly the same when tinlarebant hits the market. Doctors are used to judge on the retinal health based on these images. They use these images to make the diagnosis, and they would also use these images to monitor disease progression. Meaning that when tinlarebant gets to the market, the standard of care would remain very similar with the only big difference, obviously, that patients would be on treatment, would receive tinlarebant to slow down the progression.
The progression would be judged on fundus autofluorescence images, and there would be no change in standard of care needed for that.
Got it. Then I know you touched already a little bit on the confidence that one DRAGON trial should be sufficient for regulatory approval. As you think about the DRAGON II trial
Yeah
What are some key objectives you're looking to accomplish with that?
Yeah. You should keep in mind it's originally designed for Japan. In Japan, a tinlarebant designation or breakthrough or pioneer drug designation to allow tinlarebant to be marketed as the first drug in the world in Japan for Japanese patients. In order to fulfill the requirements for a successful PMDA submission in Japan, we initiated the DRAGON II trial to collect data in Japanese patients. Since Japanese patients could not already participate in the DRAGON I trial, we needed a second trial. We took an opportunistic approach with the DRAGON II trial, running a trial with essentially the same design to allow patients that have vaded in the U.S. and the U.K. to participate in the DRAGON program to participate in the DRAGON II trial.
The DRAGON II trial is not intended as a second trial to support our submission to the FDA, it could, in a given scenario, support our application if needed as a post-marketing commitment.
Got it. Very helpful. As you think about, I guess, submission to the FDA, the label that you potentially are looking to get, do you think it would be a broad label, younger pediatric patients if needed, what would be the subsequent pathway to that?
Yeah. This was discussed with the FDA, it is I think an easy case to argue for us that Stargardt disease is the same disease regardless of the age of onset. It's due to dysfunction of ABCA4. We also have natural history support for claiming that Stargardt disease of 12 years and older, which would be our intended designation for tinlarebant, is supported by natural history, which shows that the progression rate in ProgStar, for example, of the age groups six plus until 50 years was essentially the same. Meaning that there should be no difference if we intervene early or late for patients having successful and safe treatment being on tinlarebant. Maybe a comment on how early and late we can and should treat.
If you think about Stargardt, there was a reason why we enrolled patients with lesions defined on autofluorescence images in the DRAGON trial, namely because we needed something that can be tracked over time. The question is the generalizability of the data to patients that are even earlier in disease stage or later. In my opinion, there is very good generalizability, meaning that patients that are early in the disease stage have not yet developed lesions or not lesions as large as needed for being enrolled into the DRAGON program, would definitely benefit from the drug because we can or could keep them for longer on very good vision. Also, patients that reached relatively low levels of vision, like 20/20 to 20/200 in the better seeing eye, which would correspond to legal blindness in the U.S.
These patients still have a lot to lose, meaning that they will continue to lose down to 2400, 2800 to even NLP, no light perception. ABCA4 belongs to the very few genes in the disease space of IRDs that can lead to no vision at all, no light perception, and we would like to avoid that or at least delay that. Meaning that the addressable patient population out there is large, in my personal opinion, includes almost all patients affected by Stargardt.
Got it. Then maybe just moving to post potential approval, what do you think about the competitive landscape in both Stargardt and GA across modalities and where this might fit in?
Yeah. We welcome competition and we welcome the search for a therapy, a new therapies for Stargardt disease. I've seen these patients for more than 2 decades, and there has been nothing for these patients, and they go blind, and the only thing that you can help them with is low vision aids and obviously advice not to supplement vitamin A as an example. We certainly welcome the competition in the space. When the gene was discovered in 1997, this was actually the start of my residency. People thought now that we have the gene, it may only take a few years, and then we will have gene therapy to target Stargardt disease. It turned out to be so difficult. 10 years after the approval of LUXTURNA for RPE65 disease, there has not been a second gene therapy in the whole space.
We have learned how difficult it is to develop gene therapy, and still most of the competitors are trying to develop gene therapy for Stargardt. There are a couple of very important problems. It is very difficult to target photoreceptors as opposed to RPE. Transduction efficiency is variable. You induce inflammation. You can only deliver the gene through surgery underneath the retina. You have to detach the retina, in Stargardt, specifically the macula, surgically. You should never detach the retina otherwise, obviously. In Stargardt, you have the specific problem that ABCA4 exceeds the packing capacity of AAV, and therefore, makes it even more complicated, for example, to use two vectors or to try to use intent technology to glue two proteins together in order to have a compound that could compensate for the loss of function of ABCA4.
Long story short, there is very little competition currently, specifically in the gene therapy, but also in the stem cell arena of therapy approaches for Stargardt disease.
Got it. Maybe switching gears just to kind of starting to talk through commercial opportunity, maybe patient population size. How do you think about the size of opportunity for Stargardt? Are there areas you'd bifurcate them or kind of divide them out? Just maybe more broadly there, we can dive into a couple specific nuances.
Yeah. Until recently, it was very difficult to actually know about the affected patient population in major countries like the U.S. Population-based studies like the Beaver Dam in the U.S., Rotterdam in Europe, or Blue Mountains in Australia provide us with good estimates for common diseases like age-related macular degeneration or diabetic retinopathy, but really nothing for rare diseases like Stargardt disease. Until recently, maybe about a decade ago, genetic databases have become available, and we know about the prevalence of carrier rates of mutations in target genes like ABCA4, plus we know the penetrance rates of these mutations in Stargardt disease.
Taking that together and research over the last decade allows us to estimate the patient number, taking all races in the race mix in the U.S., Caucasians, African origin, East Asian origin, into account to arrive at a patient population of between 49 and 57,000 patients in the U.S. That means all patients of all ages. It also means all patients not necessarily diagnosed and certainly not necessarily genetically confirmed. At least we have a good estimate of the total population of Stargardt patients in the U.S. Now we get to the fraction of patients that carry a diagnosis. That number is presumably pretty high because it's a severe disease that would not remain without any diagnosis, but it doesn't necessarily be the precise diagnosis.
When it comes to genetic confirmation, that is another hurdle. We anticipate a relatively high price tag for the drug. Clearly the reimbursement system will demand genetic confirmation. I personally believe that the standard of care must include genetic confirmation for Stargardt disease. This is one of the efforts we are currently having for our launch to get as many patients as possible genetically tested in order to maximize the number of patients that can benefit from the treatment.
Yeah. Maybe, Paul, are there any specific initiatives you can talk about this time, or is that something for later?
Yeah. Maybe I pass it over to Hao-Yuan.
Well, first of all, they are existing labs that provide the testing. One of them is actually fully sponsored by the Foundation Fighting Blindness. I think the other one does probably screen for slightly more IRD. However, the insurance will cover that majority of the cost as well. I think to Hendrik's point, we see that the genetic testing become more and more available nowadays, hopefully with a treatment available, that's going to be additional incentive for the patient to be incentivized to take the testing. For us, it's really more about on education about the disease and about helping the physician and the patient understand what are the testing available to get the test they can. Sometimes people are not taking the test, where to go to the testing, and what's a consequence of that.
I think in the past, there has been some misunderstanding or worries that people think that, "Oh, if I get the testing done, but there's no treatment, does that affect any of my insurance premium?" That kind of worries, right? With the treatment coming or with the treatment available, that will really change a lot of that kind of worries. What we're doing, of course, before the drug is approved, all we can do is just continue to bring the awareness of the disease and make people to be aware of the testing.
Got it. I guess in addition to kind of education and maybe zooming back out in terms of your guys' preparation for commercial readiness, are there other initiatives you're taking to make sure that you are ready?
Well, I think at this point of time, we have done several research, pricing research, patient journey research. We look into different database in the U.S. to get a good sense about how many patient are already diagnosed with Stargardt disease and genetically confirmed. That's why we plan to have maybe a commercial day event in September to share what we know. We do see that Stargardt is a huge unmet need. Like Hendrik said, patients, definitely most of them will seek help, and we see U.S. retina specialist community is very well aware of the disease, and there's a majority of those retina specialists do prescribe genetic testing. We do not worry too much about the awareness from that community.
Maybe more education awareness is required for the general ophthalmologist or low vision optometrist, just to make sure that they could be aware of when they see certain symptoms of the patient specialist.
Very helpful. I guess as you position ability to fund a lot of the commercialization you have ahead of you, can you just talk through that a little bit?
Sure. We do see that launching for Stargardt disease in the U.S. is something that we can do, given it's a rare disease. We do believe that we have a team that experienced about IRD. At the same time, we do sit on $800 million cash. Our budget for commercialize for this in the U.S. probably is somewhere about $300 million. With our existing pipeline, probably cost about $150 million for the next three years. In total, we're thinking about somewhere like $450 million-$500 million is going to be the burn rate for this year and the next two years. With $800 million cash, we think we are pretty comfortable that we should be able to launch this in the U.S.
Great. Then you guys may also be eligible for PRVs. Does that factor into your thinking as well?
Well, yes. With the rare pediatric disease designation, if the drug is approved, we should be qualified for PRV. The calculation that I just shared has not included that potential additional $100 million-$200 million proceeds from PRV.
Maybe just final question. Obviously, you mentioned pipeline and kind of other indications. Anything you want to comment on in terms of ability to continue to expand?
Yeah, that's a good question. We have been really focused on Stargardt for the last few years, and we know how big the potential is, how strong the unmet need is. The team has been very devoted on the development of both Stargardt and GA. Once we got some revenue coming in from Stargardt, we definitely are open to look into other expansion. One of the area that you could potentially think about will be expanding to more IRD disease. I think currently the market does not have a company specific focus on IRD, with some success on Stargardt, that could be a good foundation for us to continue to look for treatment for other IRD disease.
Great