Hello, and thank you for joining us to discuss Belite Bio's third quarter 2023 financial results. Joining the call is Dr. Tom Lin, Chairman and CEO, Dr. Nathan Mata, Chief Scientific Officer, and Hao-Yuan Chuang, Chief Financial Officer of Belite Bio. Before we begin, let me point out that we will be making forward-looking statements that are based on current expectations and beliefs. These statements are subject to certain risks and uncertainties, and actual results may differ materially. We encourage you to consult the risk factors discussed in our SEC filings for additional detail. Now, I'll turn the call over to Dr. Lin. Thank you. Thank you for joining our reporting for the third quarter. I'm Tom Lin, CEO of Belite Bio. Joining me is our CSO, Nathan, and CFO, Hao. I'd like to start off by giving an overview. So, Tinlarebant is a novel once-a-day oral tablet designed to bind to serum retinol-binding protein, or known as RBP4, as a means to specifically reduce retinol delivery to the eye. This approach is intended to slow or stop the formation of the toxic retinol- derived by products, which are generated in the visual cycle and are implicated in the progression of Stargardt disease and geographic atrophy, secondary to dry AMD. We believe that early intervention directed at emerging retinal pathology, which is not mediated, mediated by inflammation, will be the best approach to potentially slow disease progression in Stargardt disease and geographic atrophy. There's still a significant unmet need for both indications, as currently there is no approved treatments for Stargardt disease and there are currently no approved oral treatments for GA, and we are already in global Phase III trials for both indications. So far, we have been granted Fast Track designation, Rare Pediatric Disease designation, and Orphan Drug designation. We have several patent families and with composition of matter patents lasting until 2040, and with patent term extension and new patents to be filed, which, we will have patent protection way past the 2040s. So we still have a very long patent life on this drug, and we are already in late-stage development. For Stargardt indication, the Phase III is already fully enrolled, with estimated interim readouts by second half of 2024. We've also just recently presented positive 24-month treatment results from our Phase II, which Nathan will be presenting the results later on. For GA in dry AMD indication, the Phase III is already enrolling subjects. And with this, I'd like to pass this on to our CSO to give an update on our clinical trials. Nathan? Thank you, Tom. So I'd like to start by giving you an overview of our clinical trial designs in Stargardt disease. I'd like to first start, however, just to orient everyone, that it's important to note that the reduction of atrophic lesion growth rate, as measured by retinal imaging, is the FDA-accepted primary endpoint for both Stargardt disease and geographic atrophy. In Stargardt disease, the atrophic lesion is called a definitely decreased autofluorescence lesion. Just keep that in mind. Whereas in geographic atrophy, it's simply referred to as atrophic lesion growth, but it is essentially the same thing. Back to the Stargardt trial design. There are two trials that one we've recently completed and one that's ongoing, as Tom mentioned. The one shown on the left-hand side is our open label phase II Stargardt trial, enrolling 13 adolescent subjects aged 12 to 18 years of age from Australia and Taiwan. This is a two-year study looking primarily at safety and tolerability. We had identified the optimal dose in a previous phase Ib, that optimal dose being 5 mg daily, and I'll show you some of the pharmacodynamic data from that dose. We're looking primarily in terms of efficacy at the atrophic lesion growth, but in Stargardt disease, we're also looking at a predecessor of the atrophic lesion growth, which is autofluorescent lesions that are referred to as questionably decreased autofluorescence, or QDAF. In this phase II study, the 13 subjects enrolling only started with this early lesion type. They did not have atrophic lesions. So we want to measure two things in this study. One, the conversion time from the QDAF lesion to the atrophic lesion, the DDAF lesion, and then also, once the DDAF lesion is formed, we want to measure the growth rate of that incident DDAF lesion. I'll share some of that data with you in a moment. You can see at the bottom there, the key inclusion criteria, 12-18 years of age. These, all of these subjects have been both clinically and molecularly confirmed as having Stargardt disease. The other study that we're doing in Stargardt disease is our pivotal phase III study called DRAGON. This study, as Tom mentioned, has just recently completed enrollment at 104 subjects. It's important to note that all of these subjects in the phase III trial will have atrophic lesions at baseline, because that's the only way that you can really measure the endpoint that is growth of atrophic lesion growth when you have it at baseline. That's what we'll be doing in phase III. This is a global study. That's the only way we could get 104 subjects. In fact, this is the largest study in Stargardt disease in adolescent patients ever conducted. You can see the randomization there is 2-1 favoring Tinlarebant. It is a two-year study with the same efficacy endpoints that we're looking at in the phase II. We're also looking obviously at vision. We're looking at anatomic markers by spectral domain optical coherence tomography, and we're looking at retinal sensitivity by microperimetry. At the very bottom there, you see the key inclusion criteria, very similar to the open-label phase II, except we've expanded the age range another two years to go from 12 to 20 years of age. And we've also defined the upper size of the lesion at 7.62. So all of these subjects will have lesions that size or smaller, and you'll see there the BCVA vision requirement of 20/200 or better. With that, I'd like to next go to the next slide and talk to you about some of the pharmacodynamic data I just mentioned earlier. What you're seeing here is a pharmacokinetic and pharmacodynamic profile of 5 mg Tinlarebant in those adolescent Stargardt subjects that were participating in the open label phase II. The blue line shows the level of Tinlarebant in blood, and the red line shows our pharmacodynamic biomarker, retinol- binding protein 4. This is what the drug is hitting. And you can see here that there's a very nice correlation between the increase of Tinlarebant in blood and the decrease of retinol- binding protein 4 in blood, until we withdraw the drug at month 24. You'll see a very rapid reversibility over about 28 days of drug cessation. So we get about 87%, return, that is, of the baseline value at the end of study. But during study, we have approximately 80% reduction of retinol- binding protein 4 in the blood. This is important to note because in a prior clinical study, not conducted with Tinlarebant, but a different RBP4 antagonist, we found that a reduction of 70% or more of retinol- binding protein 4 was effective to produce a slowing of lesion growth in patients with geographic atrophy. So this has become our target threshold for RBP4 reduction. We want to achieve at least a 70% reduction or more, and here, with a 5 mg daily dose, we're achieving a mean 80% reduction in all subjects that have been dosed. Next slide, please. I mentioned to you that in this open-label phase II, of which you're looking at the 24-month data at right now, these adolescent Stargardt subjects did not have atrophic lesions at baseline, only autofluorescent lesions. We wanted to understand if we're actually having a treatment effect over 24 months of treatment. So what we did is we did a comparator study. We compared our data to data from the largest natural history study of Stargardt's conducted to date. That study is known as ProgStar. It was conducted over several years, involving 400-500 patients in these studies. Most of those subjects were adult Stargardt subjects. However, there was a subpopulation within that larger group of about 50 subjects that had the exact same baseline characteristics as our subjects in the open label phase II. That is, they were 18 years or younger, and they had no atrophic lesions at baseline. So this subgroup served as a nice apples to apples comparison for our analyses, and you're looking at them here on the left-hand side. What you're looking at is the growth of incident atrophic retinal lesions over the course of 24 months in each group. The ProgStar group is shown in blue, and our group, in Tinlarebant treatment, is shown in red. So neither group had atrophic lesions, so we're measuring the time to the incident lesion growth, and then we're measuring the growth of the incident atrophic lesion over time. You can see that at every time point, from six to 24 months, we consistently have a lower reduction, or sorry, a lower DDAF lesion growth compared to ProgStar, such that by the end of this trial, at 24 months, we have a highly statistically difference between ProgStar growth and our growth. That is, our lesion growth being about 50% lower than what was published in the ProgStar studies. This is phenomenal. In fact, it's never been reported before. The other important aspect of this study is among the subjects that we had in study, five of them did not convert to any atrophic lesion, and that, again, is another important finding because it suggests that our drug is slowing the conversion of the early lesion type, the autofluorescent lesion type, to the atrophic lesion type. That's the first observation. The second observation being, once those incident lesions are formed, we have a slowing of the growth rate of those lesions compared to natural history. So it's important to note that this phenomenon, this, what we're seeing, this slowing of lesion growth, is statistically significant at every time point that we've measured, and in fact, as I said before, has really never been demonstrated in another interventional treatment trial of adolescent Stargardt patients. Next slide, please. So now I'd like to show you our 24-month visual acuity data. What you're looking at is the visual acuity data from both the study eye, which is designated at baseline, and the fellow eye. This disease is a bilateral disease, so the disease will affect both eyes equally, and our drug, of course, is systemic, so it will affect both eyes equally. But what we're showing you here is both eyes, just to show you over time, we're getting stabilization in each eye of each patient over 24 months. This is significant because although this study was not powered for treatment efficacy against vision, it is important to see stabilization over 24 months because these subjects will in fact lose vision annually over time. So the fact that we're slowing lesion growth and also stabilizing vision is a very important finding. The other point I want to make is that visual acuity losses or gain within 10 letters is not considered clinically significant. So what we have here really is potential test-retest variability affecting these data. So until you get outside of 10 letters, there really is nothing clinically significant to say about the data. But it's important to note that there is stabilization while we are in fact slowing lesion growth, as I've showed you in the previous slide. Next slide, please. Now I'd like to show you the drug-related adverse events. It's important to note that although this is a systemic drug, there have been no clinically significant findings in relation to vital signs, physical exams, cardiac health, or organ function. What we see are two anticipated ocular drug-related AEs that we would like to see because they're telling us we're having the intended biological effect on the retina. The first is a form of chromatopsia called xanthopsia. This is mediated by cone photoreceptors. This is a photoreceptor subtype in your retina, which confers bright light and color vision. So when subjects under our treatment transition suddenly from a very darkened environment to a bright environment, this activates cone photoreceptors. Cone photoreceptors will demand chromophore. Under our treatment regimen, that chromophore will only be slowly supplied to the cone photoreceptors, so there'll be a delay in their timing to attain maximum bright light sensitivity. During that time, there will be an artificial and electrical produced hue of color in the visual field, in this case, yellow. It can last seconds to sometimes a few minutes. It's important to note that these subjects are reporting it as mild, and of course, it's transient. The majority of subjects have encountered this AE, 10 of 13, but no one has left study because of them. The other manifestation is called delayed dark adaptation, and this is mediated by another photoreceptor cell type called the rod photoreceptor, which confers dim light vision. So when subjects under our treatment transition suddenly from a very bright light to a very darkened environment, that activates rod photoreceptors. They will also demand chromophore to maintain dim light sensitivity. Under our treatment regimen, that chromophore will be slowly supplied. So there will be a period during which subjects will not have maximum dim light sensitivity. They will eventually gain it. It's somewhere between 8-15 minutes delay, but it will be attained. This is not night blindness. And once again, you can see a high number of subjects having this, this AE, but nobody leaving the study because of it, because, again, it is mild, it is transient, and of course, it's fully reversible. So we believe this drug, 5 mg daily, is effective to achieve what we want in the retina with these very mild and well-tolerated AEs. It's also important to note that, in our treatment overall, this has been very well tolerated. Again, no systemic AEs and only ocular drug-related AEs, which we need to see as they are intended, to have this effect on the retina. Next slide. I'll now show you our trial design for geographic atrophy. This is a Phase III trial design. It's going to look very similar to our Phase III trial in Stargardt's disease, with two important exceptions: the indication, geographic atrophy instead of Stargardt's disease, and the number of patients, 430, to reflect the higher prevalence of GA in the population. Of course, it'll be a global study, same randomization scheme as in the Phase III Stargardt trial, two to one favoring Tinlarebant, two years in duration. We're going to be looking at the same efficacy measures. Of course, here we're looking at atrophic lesion growth. It is the same as DDAF, but just named differently, but it is the same phenomenon. And of course, we're looking at visual acuity outcomes, anatomical outcomes by SD-OCT, and retinal sensitivity by microperimetry. And there will be an interim analysis at one year, as there was in the phase III—as there will be in the phase III Stargardt trial. Next slide. With that, I, I'd now like to throw it back to Hao-Yuan so we can discuss the Q3 2023 financial results. Thank you. Thank you, Nathan. So in Q3, we had $8.7 million R&D expenses, increased from $1.2 million for the same period last year. The increase resulted primarily from increase in expenses for conducting the DRAGON and the PHOENIX study, and also the wages and salaries due to share-based compensation granted to our R&D team in Q3. We had higher expenses this quarter because we met several development milestones on both DRAGON and PHOENIX studies, such as the completion of the enrollment for DRAGON and the first patient in for PHOENIX. On the G&A expenses, we had $2.2 million G&A expenses, compared to $1.4 million for the same period last year. The increase, again, is primarily from an increase in the share-based compensation granted in Q3. On net loss, we had a net loss of $10.9 million, compared to $2.4 million for the same period last year. In terms of cash, we received $6.5 million cash inflow from warrant exercises, from ATM offering and R&D refund. With approximately $9.4 million cash outflow, our net cash outflow in Q3 is around $2.9 million, leaving us with $54.5 million cash by end of Q3. Thank you. Back to you, Tom. Thanks. I would like to conclude with the key milestones that we have achieved this year. In Q1, we've initiated our phase III study in geographic atrophy, secondary to dry AMD. In Q3, we completed the enrollment for the phase III study in Stargardt's disease. We also just recently completed our phase II study in Stargardt's disease and presented very promising and positive results at AAO last week. And finally, we are expecting interim results from our phase III Stargardt's trial in the second half of 2024. Thank you. Right. So we will start to take questions. For anyone who has questions, please raise your hand. I got Jennifer Kim from Cantor. The operator, please unmute Jennifer. Hi, can you hear me? Yes. Yes, we can hear. Thank you. Okay, great. Thanks for taking my questions, and congrats on the progress this quarter. I have a few questions. The first is: Is there anything you can say in terms of the number of patients you've enrolled so far in PHOENIX and how many clinical sites are up and running? And then my second question is more of an OpEx question. The higher R&D expenses, how much of that was due to the development milestones and aren't expected to repeat in future quarters? And, similar on the GA side, is the $2.2 million a fair way to think about as, as a base going forward? Thanks. ... Nathan, you want to take the first question? Yeah, I can handle the clinical trial question. So Jennifer, in the phase III Stargardt study, we've enrolled 104 subjects. We have some baseline demographic data from those subjects that shows actually very favorable balance in terms of the prognostic factors that would be predictive of disease. So for instance, a foveal involvement, 98% of our subjects are foveally involved. BCVA, all the subjects have similar BCVA. They all have similar lesion size, which is a smaller lesion size. It's exactly what we are going for because it gives us an opportunity to interrogate this whole early intervention differentiator, which is what we believe our drug will be able to do. So with these smaller lesions, fovea involvement, these two things alone allow us to test this idea about early intervention as well as potential visual acuity benefit, because these lesions are encroaching into the fovea. These kids will be losing vision over two years. So I think everything's turning out very well in terms of the baseline value in terms of the baseline demographics for this phase III Stargardt study. Your other two questions were largely financial, so I'll throw those to Hao-Yuan. Oh, actually, I think I was asking, the PHOENIX trial on the- Oh. - GA side, how enrollment is going there? Oh, sorry. Yeah, so we've just begun enrollment. I think we've got up to six, maybe seven subjects now. We've only been at it a couple of months, so we're really just starting to ramp up that recruitment effort. Again, we're going after 430 subjects, so we predict probably another year to maybe 14 months to complete that enrollment. And again, we're really just beginning in the early phases of that enrollment effort. Hey, yeah, so about the R&D questions. So for this quarter, because of the milestone was hit, I think we paid about $2 million for the DRAGON study. And also for the PHOENIX, I think about $1.9 million for the first patient screen and dosed. So we don't really expect this high expenses moving forward. As you guys may recall, we had much lower Q2 expenses because we have over signed up on the DRAGON study. We wanted to enroll only 90, but before we closed the enrollment, there was already 104 subjects signed up. And we had to wait for the last one to be scheduled for dosing. That's why we had to delay the milestone for being fully enrolled from Q2 to Q3. That's why the Q2 expenses was much lower than what we expected because you didn't have the milestone, and for the Q3, it does. And the Q3, you have the DRAGON fully enrollment milestone hit, you got progress of the DRAGON study milestone hit as well, and also you have the PHOENIX first patient dose milestone hit as well. So that's why you see a much higher Q3 expenses. But moving forward, it probably will goes down with, you know, to probably less than 2 million-I mean, have $2 million lower, given we don't have those milestone for the DRAGON to continue in the following quarters. And then on the GA side? On the GA side, we don't really know at the moment to give a very accurate expenses forecast, because it will highly depends on the progress of the sites and the screening. And because, you know, the cost will also highly depends on the progress on that. So we kind of can have a like a full year idea, but for quarter by quarter, we probably will have to wait and see the sites to start up and running for a bit, to be able to know like how fast the IRB approve the contract the study, and also how fast the PI can call back their patients. For now, what we have is like how many patients they have, but until they... You know, not until they actually call the patient to schedule a visit, we wouldn't really know how fast those patients will be able to come in. Sometimes you have many patients, they all want to come in the same week, but there's just not enough manpower to handle so many patients, and they will have to reschedule those patients. I think I will have a better idea, probably, in Q1 next year, once we have more sites up and running. Thanks, guys. Thank you. I have next questions from Mark from Leerink. Hi, can you hear me? Yes. Hi, this is Basma. I'm for Mark. We have a questions about whether you guys gonna release more analysis or detailed analysis of the phase II data with related to the genotype and the phenotype of the patients. What I mean is unifocal, multifocal, and the different mutations of the patient that were enrolled in the phase II. And how are you gonna regarding the phase III, which is DRAGON, are you trying to match the baseline characteristics when it comes to phenotype and genotype to the phase II, or is it a little bit different? ... Yeah, so, Padma, let me take that question. So regarding the phase II data, will we be providing additional data analysis? The answer is yes. So I'm actually presently preparing an abstract for ARVO 2024, that meets in Seattle in May, to talk exactly about that, to provide those genotypic data, because we were very surprised to see no DDAF lesion growth in five of the 12 subjects throughout two years of treatment. And we thought maybe it's possible that these kids have more mild mutations. In fact, that's not the case. They all have very severe mutations. So the finding that they're not converting from the early autofluorescent lesion type to the atrophic lesion type is actually very profound. So we're providing both the genotypic and phenotypic data and a better breakdown of the lesion growth on a subject-level basis. So that will be provided at ARVO. In terms of paralleling the phase II and phase III trial designs, the study designs themselves are essentially identical. That is, it's a two-year study looking at the same endpoint with the same dose, 5 mg daily, and of course, it's the same patient population. We've increased the age range from eighteen to twenty, so in the phase II, it was twelve to eighteen, in the phase III, it'll be twelve to twenty, and that was at the behest of principal investigators who were wanting to get their patients in. But other than that, there's no, no real difference other than this name difference, which is in the phase III, these kids will have to have atrophic lesions at baseline. That's necessary in order to meet the requirement for the endpoint, which is slowing the growth of atrophic lesions. That is what the FDA wants. So you have to start with some measure of atrophic lesions at baseline, whereas in the phase II, the open label study, all of these children, these kids had earlier lesions, so they only had the autofluorescent lesion types. And so that's really the big difference between the phase II and the phase III. And of course, we're not trying to match the genotypic profiling that we saw in phase II with what we're seeing in phase III. We're basically taking all comers as long as they fit our inclusion, exclusion criteria without respect to genotype. Got it. Thank you so much, Nathan. Thank you for the question. Hey, next questions we got, Yi Chen, from H.C. Wainwright. Thank you for taking my questions. My first question is, is this so that, in the DRAGON trial, the lesion size at baseline is smaller than the baseline in the phase II? And how is that going to impact the data readout? Yeah. So it's important to note, as I mentioned earlier, in phase II, we don't have any atrophic lesions at baseline, so these patients only had autofluorescent lesions, and they were varied sizes. One patient has a lesion as small as 1 mm. Another patient has a autofluorescent lesion as big as 11 mm. So there's a very big size range in the autofluorescent lesion types, and of course, we'll be presenting that as part of the ARVO abstract and presentation that I spoke earlier about to Padma. In terms of the balancing in phase III, the lesion type, as I mentioned earlier, is actually quite small, which is what we want. So we look at the mean lesion type at baseline in the phase III trial, and we see an atrophic lesion, mean size of 2 mm square, which is great. So when you're asking, are the lesion, sizes different between two studies? Yes, because the lesion types are different between the two studies. But in this phase III trial, we have exactly what we are shooting for. We want to start with smaller lesions at baseline, and quite honestly, the smallest detectable lesion is about 0.05 mm. We're at 2 mm, and we believe, based upon some other trial results from other companies, that these smaller lesion types will respond to this type of pharmacotherapy. So that was sort of a long-winded way of answering your question, but I hope I got to the answer for you. Thank you. But, the lesion size at baseline in the DRAGON trial, is that representative of the overall patient population in the real world? Very tough to say. So remember, we have 104 adolescent Stargardt subjects from all over the globe: China, Europe, North America, and other countries. So this is a good representative sampling of all the genotypes and phenotypes that are manifest in adolescent Stargardt disease. This is the largest study of its type. It's never been done before. So if there's any true sort of homogeneity of the demographics, this study would tell you what it is. There's no other study that we can look to and say: Well, we want to match that demographic because that fits best our patient population. No one's ever done a long-term epidemiological study or natural history study in just Stargardt's disease, with the exception of ProgStar, where they have a subset of patients that are of the same baseline, that is, the same age and lesion characteristics as our patients in the phase II. But in terms of an atrophic lesion type, that it would be representative of the Stargardt population, I would say we have it, because, again, this is the largest collection of adolescent Stargardt subjects in any investigational trial that I'm aware of. So, Yi, may I add that in fact all studies right now to date, most of those patients came from North America and Europe. Same goes for observational studies such as ProgStar. They are mainly patients from North America and Europe. So, what's being reported out there and published out there, it's all similar genotypes, I believe. Thank you. And, how much slowing of the lesion growth rate in the DRAGON trial will allow the company to ask for accelerated approval from the FDA? I can take that. So- Oh, so you want to take it, Nathan? No, no, Tom, please. Oh, yeah. So we had discussions with FDA regarding this. I think at this moment, the FDA don't... As Nathan mentioned, this is the first-ever study, especially in pediatric or adolescent Stargardt disease. So the FDA right now just asked us to show them the data. So right now, they, they have not given too much away. So I guess that's the standard response from the FDA. So they want to see the data first. Got it. My last question is, if the PHOENIX trial has data similar to what has been observed from the phase II Stargardt trial or the later DRAGON trial, how competitive do you think Tinlarebant is compared to drugs already approved for dry AMD and GA? Nathan, you want to take this? I think you're- Absolutely. Yeah, absolutely. I think it's gonna be a game changer, Yi. I mean, so if we look at the approved drugs for GA right now, you're not getting anything better than about a 20% treatment effect on slowing lesion growth, and you're not getting much of a visual acuity benefit, and you have a significant safety risk. With an oral therapeutic that achieves the same level of efficacy, that is a 20% slowing of lesion growth, with a very clean systemic safety profile and no potential for adverse ocular events that would be permanent, such as the occlusive retinal vasculitis that's been documented in rare cases with the Apellis drug. When you have that, and you're reducing the treatment burden so significantly with an oral once a day, I think patients will flock to this oral once-a-day treatment, because it's going to be superior in terms of benefit to the patient, in terms of reducing treatment burden and reducing any potential risk to their ocular health and safety. So that's the one thing. The other thing is that, you know, these drugs, the complement inhibitors that have been approved, the Apellis drug, the Iveric drug, these are anti-inflammatory agents, so they, they address late-stage disease, which is driven by inflammation. They will not. Those drugs would not be effective in early-stage disease, which is one of our differentiators, because there's no inflammation in early-stage GA or early-stage Stargardt. So we see there another potential market opportunity for us because they cannot use their drug effectively in our subjects. However, our drug can be used effectively both in the younger patient population as well as the late onset, as sort of a maintenance therapy. So I see more of a synergy than I do a competitiveness. But in terms of, you know, what would be preferred, I think if we achieve the same level of treatment effect with this very clean safety profile, our drug would certainly be preferred by most GA subjects, patients. Thank you very much. And I think there's a survey done with 3,000 ophthalmologists, and I think over 80%, they, they say that, that they wouldn't prescribe the attack complement injections for their patients. So it's very overwhelming and surprising that even the drugs approved for GA, most ophthalmologists still wouldn't prescribe it. Got it. Thank you. Thank you. Okay, we also have Bruce from Benchmark. Hi, good afternoon, and thank you for the informative presentation. Just one quick financial question, if I may. The non-cash expenses for the quarter, do you have an estimate of the D&A and the stock comp? Oh, sorry, say again? The non- Yeah, the non-cash expenses for the quarter. Right. That's mainly for the ESOP that we grant in Q3. Okay. All right, well, that's it for me. Thank you. Thank you. Well, Bruce, thank you for asking that question. Yeah, we actually even if you look at the income statement, it looks like we have much higher expenses this quarter. But cash-wise, it's about, I think, $2 million, is about the non-cash component coming from the ESOP that we grant in that quarter. And also we have been seeing more and more people exercising their warrants from the follow-on. So we did raise the $5 million from the warrants in this quarter. So in fact, our net cash outflow in Q3 was only, I think, close to $3 million. So it's not that significant. Financially, cash-wise, we're doing pretty well. Okay, if anyone has any questions, please raise your hand. I don't have any written question on the platform here. Okay, so, if no other questions, we'll conclude today's presentation. Thank you for your time.