Should we get started? Still running. All right, good morning, everyone. Thanks for joining us at TD Cowen's 45th Annual Healthcare conference. I'm Tara Bancroft, one of the Senior Biotech Analysts here. And for our next session, we have a fireside chat with Ocular Therapeutix. And with us from Ocular, we have the CEO, Pravin Dugel. Thank you, Pravin, for being here. It's always a privilege to have you. And so, actually, before I get started, I just want to note to the audience that if you have a question, raise your hand, call out, and we'll make sure you get your question heard. And so, Pravin, to start the discussion, maybe you can begin with just a general update, overview, maybe even focusing on what you talked about on your earnings call yesterday, just to start us going here. Great. First of all, Tara, thank you for having us here, and thanks to everybody for being here. We had a very important update yesterday, and I think the bottom line of the update is what we have done is announced a path very clearly to get to the finish line faster and in a much more cost-effective manner, so a lot faster and a lot cheaper without changing the primary endpoint at all. Let me explain. As you know, we have two trials that are ongoing, two registrational studies. One is called SOL-1, which is a superiority study, and the second is called SOL-R, which is a non-inferiority study. The superiority study is designed really for the first year, which is efficacy, the primary endpoint being at month nine. The second study, which is SOL-R, the non-inferiority study, had 825 patients originally to be enrolled. And the reason for that is primarily for safety. As you know, the FDA requires that 200 patients or so be followed for at least two years in the target that you're applying for or a different target at the same dose or a higher dose than what you intend to apply for. So to satisfy those requirements, SOL-R had to be fairly large with 825 patients. What we noticed was, this is under masking with SOL-1, which is completely randomized, as you know, what we noticed were two things. First of all, we noticed that the retention, the patient retention, was absolutely fantastic. We were thrilled, still are, absolutely thrilled with the patient retention in SOL-1. The second thing was the primary investigators, the PIs, came to us and said, "Look, we want to continue this study. We want it to go on to year two. The patients want to continue on this study. We don't want to stop." So with those two things in mind, we looked at this and said, "Look, we can actually make SOL-1 a lot more efficient. The first year is all the FDA cares about for efficacy. So here we give one injection. We'll have information for month nine, but if we can extend the study for two years, we'll also have information for month 12 regarding efficacy with that single injection. We really believe that this drug will last at least till 12 months, and we'll have that information. For the second year, what we can do is to dose every six months, I'm sorry, so that we can maximize the exposure of the drug to patients so that will qualify for the safety review for the FDA." So we wrote an amendment to our SPA. We proposed to the FDA, and the FDA said, "Absolutely, that's perfectly fine to do." We also decided at that point that because we could do that, SOL-R didn't need 825 patients. Now we could reduce the size of SOL-R by about a third to 555 patients, and together with the pooled data, that would satisfy the FDA's requirements for safety and allow us to decrease the delta between the time delta between SOL-1 and SOL-R readout, and as I said earlier on, get to the finish line a heck of a lot faster and a heck of a lot cheaper, but, and this is really important, without changing the primary endpoint of SOL-1 at all. It deserves emphasis that the primary endpoint for SOL-1 remains exactly as it is at month nine. Nothing changes. The statistics don't change. The primary endpoint changes has absolutely no changes whatsoever. So all of this that we announced yesterday was an absolute bonus without having to give up anything whatsoever. I see. So in adding that redosing, actually, could you just elaborate a little bit more on why you wanted redosing specifically in the SOL-1 trial? So remember, it's a bit of a schizophrenic view. And SOL-1 is about as efficient as any study gets. The first year is entirely for efficacy. That's what the FDA cares about. The second year really is entirely for safety. So given that, what you want in your first year is to be able to show durability and flexibility with a single dose, which is exactly what we have now. And in that first year with this trial design, we'll be able to get information in the primary endpoint, which doesn't change at month nine, as well as month 12. In the second year, what we want to do is to maximize patient exposure to get the information that we need to satisfy the FDA for safety. So what will happen now is that at week 52 and week 76, everybody will get redosed. Okay? So week 52 and week 76, everybody will be redosed. And they will be redosed whether they were rescued or not with the original drug to which they were randomized. Okay? Let me make that very, very clear. That's how the year one and year two are divided. So if you look at the patients in year two, the exposure they'll have is every six months to AXPAXLI. Now, if you look at SOL-R, the exposure that they have is also every six months to AXPAXLI. So pooled together, those two data sets will provide enough information to satisfy the FDA safety requirements. Okay. Yeah, that makes sense. So I think one thing that people are confused about is if this doesn't impact the primary endpoint, why is the data coming later? And down the line, does that impact filing timelines at all? The answer is yes, it impacts filing timelines in a very, very positive way. It allows us to file a whole bunch faster because SOL-R is so much smaller now. The recruiting will be only for 555 patients as opposed to 825 patients. So what was the first part of the question again, Tara? Mostly what people were confused about yesterday, I think, is if the primary endpoint didn't change, why is the data later? Yeah. So because we want to make sure that we have information regarding month 12, patients will have to remain masked until month 12, right? And because they have to remain masked until month 12, that's why instead of being able to turn the card and unmask the patients and have the top line data in the fourth quarter of 2025, we'll only be able to do that in the first quarter of 2026 because of the masking. They have to remain masked. Now, I realize that there's a slight delay in terms of the card turn for SOL-1, but taken together, because SOL-R can be reduced significantly, the timeline to submission is much, much faster in general. Remember, we will be submitting to the FDA with a positive result of SOL-R at week 56. That's the primary endpoint for SOL-R. So with a positive result at week 56, we'll be submitting to the FDA. Now that week 56 data is going to come a lot faster because SOL-R does not require 825 patients. Instead, a third of that, only 555. If you had to quantify it, you know we love asking these questions, how much faster? A year, six months? Yeah, it's a great question. It's an appropriate question, and you know when the card turn will be in SOL-1. We have not guided you yet as to when we believe the card turn will be for SOL-R, other than to say that it'll be a lot faster for the reasons that I've mentioned, and when we get enough data, we will. Look, we will always inform you when appropriate. We want to make sure we have enough data to inform you in a way that's as accurate as possible, but let me go back to SOL-1. I think the read-through of everything that I've mentioned to you should be the following. First of all, it should be that the trial conduct of SOL-1 is going extremely well. I mean, and just think about this. If we were not absolutely confident in the patient retention and the trial conduct of SOL-1, we would never be doing this, regardless of what the FDA said, right? The second take-home point is that our FDA is extraordinarily collaborative and extraordinarily supportive, and nothing has changed in terms of their thinking since Dr. Chambers has left. It's the same people other than him. It's the same conclusions they come to, the same guidelines. They're absolutely consistent, and our collaboration is constant and is very, very good. Okay, and speaking of accelerated timelines, the enrollment rate in SOL-1 was remarkably fast. And so I guess I'm curious to get your thoughts on what exactly went into that and how and why you could complete it. What was it? At least a quarter ahead of schedule. So the reason is because it's retina. I mean, look, we're a one-trick pony. I mean, we know retina, and we know retina. We believe better than anybody else. If you look at our, without sounding too arrogant, if you look at the team that we've recruited, between us, we probably train 25% of the people that are out there and pretty much know everybody. So we know exactly the sites to go to around the world. We know the network. We know the people better than any other company. I mean, that's the one thing, and really the only thing we know how to do. And we've demonstrated that. And the question was always, can this group of all-stars actually execute or not? I mean, are we going to be the old Brooklyn Nets or are we going to be the LA Lakers, right? And we became the LA Lakers because we've showed that we can execute, and we're doing that again. And ironically, the biggest concern, I can tell you this now because it's off the table, that we had was these sites are so good and they're so motivated to recruit. What we kept on hearing back was, "Look, with SOL-1, we don't want to stop. We want to go to year two. Give us a way to go to year two because we don't want to have that conversation with the patients saying, 'Yes, it's positive. The drug works, but now you can't have it anymore.'" That was our biggest concern. But now with this amendment, that conversation will never have to happen. They can certainly go to year two, which is a huge relief. And the PIs absolutely love this extension of SOL-1. Okay. By the way, at TD, we're Celtics fans. Yeah. There you go. Okay. The Celtics. So, all right, that aside, I guess let's look forward to the top line data that are now in the first quarter of 2026. We're trying to get some additional context to help understand the probability of success. I know you guys put a lot of thought into who you're enrolling in this trial. And so maybe try and help us set expectations for that data, especially based on the types of patients that you enrolled in this trial. Yeah. First of all, I mean, I've said this before, and I take no credit for this. This is not me. These are the people that are around me that have designed these trials. And I really believe the trial designs and the complementary nature of these trial designs are absolutely revolutionary and historic. I bet everybody expected us to have a SOL-1 and a SOL-2 because that's what everybody does. It's a knee-jerk reaction of having the same trial twice. I mean, that's everywhere, right? But when you think about it, that's terribly inefficient because the second trial may validate the first trial and can get to approval, but it adds no further information whatsoever. In our case, that's exactly what we did not do. We have two trials that are designed completely differently. One is a superiority study. The other is a non-inferiority study. Together, they provide a huge advantage in terms of the logistics, in terms of the commercial leverage, as well as in terms of the regulatory outcomes. We can discuss any and all of them. As far as we're concerned, the way that we've designed this is strategically from a commercial point of view to give us the very best label that we can possibly get, right? We're thinking about that right now. If you think about what we've done and with the recent amendments, our goal is to have the first and only superiority label that has the flexibility of treatment from every six months based on SOL-R to every 12 months based on the amendment with SOL-1 and repeatability. It will potentially be the best label the field has ever seen, the only superiority label. Now, as far as de-risking is concerned, Tara, which is what you're asking about, we have spent a huge amount of time and thought in terms of patient selection and selecting patients that are absolutely de-risked for the clinical trials in a bespoke manner. And I guarantee you nobody else has put in as much thought as we have. And this is the most important thing in terms of de-risking. If you look at the superiority study, what we've done is to specifically select patients that have good vision, in other words, a large, juicy neovascular membrane with the most amount of anti-VEGF receptors, good vision. We've tested those anti-VEGF receptors to make sure that they function by requiring them to gain at least 10 letters of vision or get to 20/20. And we're loading them up and letting them fail once. It's the perfect patient selection. Now, for the other study, SOL-R, it's entirely different. Here, we have a non-inferiority study. So that's not the patient selection we want. It's actually very different. Here, stability is king. We want patients to be absolutely stable because it's a non-inferiority study. So by definition, stability always wins. So we've got three loading doses, and then we've put in something that no one else has, which is two periods, two opportunities to observe the patient. And what are we observing for? We're observing for instability. We're observing for fluctuations. We get a chance to weed out those patients that are unstable, two opportunities, and then two more loading doses, and then we randomize. Again, this is a huge amount of work and thought that has gone into making sure that patient selection is absolutely perfect to de-risk the trials in a bespoke manner separately for each trial. Okay. And so the changes that you announced yesterday, do they change your expectations for what the data are going to look like? I know you said it's not going to impact the primary endpoint, but as far as the data that you're going to be presenting in Q1, how does that change? And then what is it powered to show, and what are your expectations? So the powering doesn't change. The primary endpoints don't change at all. That's really important. What does change is that we'll have a lot more information. Now we'll have 12-month data information as well. If we left it as it is, we would hit the primary endpoint at month nine. And my colleagues would wonder, "Look, how long does this drug actually last?" We're absolutely confident that it lasts for 12 months in a large number of patients. We have the opportunity to show that, right, in a scientific study that is very, very well scrutinized. Why would we miss that opportunity? So nothing changes in regards to the statistics or the primary endpoint. We will just have a lot more information, which will allow us to have a much better label, to have a huge commercial advantage, and we can get all of that much faster and much cheaper. So there's nothing that we've given up here. It's a win-win-win for us. It's all bonuses. Okay. And as we wait for the data now over this next year, what can you share now that you know based on or in terms of masked rescues, both in terms of quantity and timing? Yeah. It's a great question. So as you said, we're masked, but the number one priority in our company, the number one priority is trial conduct. And that's the trial conduct right now of both studies, but particularly SOL-1, which, as you know, is fully randomized. It's extraordinarily important for us to make sure the conduct of that trial is as good as possible. So the way we judge conduct under masking is to look at rescues. And what we look at is we look at the number of rescues, we look at the cadence of the rescues, and we look to see whether the rescues are on protocol or not. And what I can tell you, again, under masking, is that we're absolutely thrilled, absolutely thrilled with what we're seeing, absolutely thrilled. I can also tell you that the vast, vast, vast majority of rescues, and I've repeated that three times, the vast, vast, vast majority of rescues are per protocol, which is exactly what we want to see. Okay, and maybe some additional context because just in case there are people listening that might not even understand why I'm asking that, I think it would be helpful for you to explain how do rescues change the integrity of the data? Yeah. So just to give you the powering and the statistics with SOL-1, remember the primary endpoint is at nine months, and the primary endpoint is the percentage of patients who maintain vision. Effectively, after loading, the randomization is one injection of Eylea versus one injection of AXPAXLI. We have a lot of data in terms of what those expectations are, right? And we believe that with Eylea, at most, that number is going to be 20%. We believe that with AXPAXLI, based on studies that we've done in the US, that number is going to be at least 70%. So we have a 50% delta. And what we need is a 15% delta, 1/5, to be statistically significant. So we're very, very well powered for that. The other thing to note is that in the SPA, the FDA has stated that nobody will be censored. Every single rescue must be counted as a failure, not censored, in the primary analysis. Every single rescue. That means whether the rescue is on protocol or not on protocol. And that's really important. Nobody gets censored. There'll clearly be many more of those in the Eylea group than in the AXPAXLI group. So that's very much in our favor. So not only is the powering in our favor, but the statistical analysis is also very much in our favor. And that's true even if a patient has very many rescues? So here, it's a different rescue question, and we can answer the second one too, in a superiority study versus a non-inferiority study, right? In a superiority study, the first rescue is counted as a failure. And at that point, that patient is counted, and that patient from that point is treated with Eylea per the doctor's discretion. So there's only one rescue that's there. The protocol is 15 letters or greater in the superiority study. That is very, very different than the non-inferiority study rescues, and it's not the same at all. And we can certainly talk about the non. Didn't we talk about the non-inferiority study rescues? I'm happy to do that too. Yeah, absolutely. In the non-inferiority study, there are rescues that are allowed. And that question came up at a certain point of saying, like, how many rescues are actually allowed? And there was some confusion as to whether it was sort of limitless or not, which made no sense, right? You can't just have limitless rescues. So that question was asked openly to the FDA, to Dr. Boyd, and his response was the same as it's been for the last 10 years, which is that nothing has changed. The guidelines for rescues are exactly the same for everybody, not just for us, but for every non-inferiority study. And the guidelines are that the first rescue, the first rescue is free. It's allowed as long as it doesn't affect the primary endpoint. The rule of thumb is that as long as it's not within about three months of the primary endpoint, anything beyond the first rescue is subject to review. And that's the way it's been for decades, and that's the way it is for everybody. Now, we have a huge advantage here. And the advantage is that our non-inferiority study, in fact, neither of our studies will ever be looked at in isolation. They'll always be looked at in the context of the other study. So imagine a situation where, let's just say there are three rescues, and say there's a sponsor that has exactly the same studies twice as everybody else does. They look at the three rescues, and they go, "Okay, there are three rescues, and this may be subject to review," right? As I said. But now imagine there are three rescues in SOL-R in the AXPAXLI group. If there are three rescues, they will have the additional information from SOL-1 being positive that this drug lasts for nine to 12 months. It's an entirely different context altogether. And that's the beauty of these two complementary designs. They'll never be looked at in isolation. One will always influence the other, and one will always contextualize the other. So we have a huge advantage in that sense in having information showing that this drug lasts for nine to 12 months as one looks at those rescues. Okay. Thanks. Oh, question from the audience? Sir, quick question. Yeah. At nine months, you think you'll get 70% patients not needing rescues. Do you have that kind of number for 12 months? And the second question is, can you answer the first question? I'll see if I can answer. Trust me, I've had those moments many, many times, especially today. So it's a great question. The answer is yes, we do. And we have a study in the U.S. that showed that at 12 months, 60% of patients were rescue-free. And realize that that was in a non-super-selected patient population. We believe that number is going to be even higher in a super-selected patient population. But the truth is, we don't need a whole bunch of patients to get that on the label. If you look at the Vabysmo experience, right, they've got the four months on the label, but there are actually very few patients that actually went that far. So history shows that it's not like you need a whole bunch of patients to cross that threshold to get that on the label. But the answer to your question is yes, we do have that data. That data in a non-selected patient population shows that 60% of patients are rescue-free at month 12. Just the second question was, I know you're obviously not up to labeling conversations. One is a non-inferiority, one is a superiority. How confident are you that if they both hit that you'll have a label that has superiority in it? Or what gives you that confidence? Yeah. So I mean, look, we haven't had, you're right, we haven't had labeling discussions, although we have had numerous discussions and continue to with the FDA. There's never been an opportunity for the FDA to really see a superiority study in terms of labeling. This is their opportunity. And given our discussions and given our data, we're actually very confident that with positive data set, we will have superiority on the label, which will be a big deal. Importantly, we'll also have flexibility, as I said, from six months to 12 months with both studies being positive, and we'll have repeatability. Okay. We have about four minutes left, but I do think one of the most important things that we really need to do work on this year as an industry as a whole is what can we take away from the SOL-1 data that can read through to SOL-R to help us predict the probability of success there? Yeah, it's a great question. I think what you can read through there is the drug itself, not the patient population. As I mentioned, the patient populations are necessarily different and selected differently in a bespoke manner. But what you can really read through there is, look, with a positive SOL-1, that means that this drug in the first year of the study is going to last for nine to 12 months, right? There's no reason that a drug that lasts for nine to 12 months isn't going to last for six months, which is the non-inferiority study, especially when the patients are super-selected like that. There's a lot more internal analysis that we've done. We are very, very confident that with a positive SOL-1, the chance of SOL-R being positive is extremely high. At some point, we'll go ahead and share some of that data with you as we can. But for now, I think the logical and easy answer really is why would a drug not last for six months when we know the drug is going to last for nine to 12 months, especially in a patient population that we've super-selected to be stable? Okay, great. So I guess we do have three minutes left. I want to get to the topic of NPDR. So can you update us on what you're doing there and expectations, and what gives you confidence that even given that Eylea is not really used in this population, how could AXPAXLI? Well, first of all, the population is humongous. It's much larger than even the wet AMD patient population. Anti-VEGFs work extremely well. The problem is, as you know, they have to be given every month or every other month, and these patients are young, they're working age, they're asymptomatic, they're not going to come in. What was absolutely remarkable about the HELIOS data set is that when you look at the most important condition, which is prevention of blindness, what we call vision-threatening complications, in the control arm at week 48, it was 37.5%, which is in line with what you see in natural history. Year upon year, there's a 30%-40% chance of these patients untreated developing some kind of a blinding complication. That's really high, right? And there are 6.5 million of these patients in the U.S., and less than 1% are being treated. So the control arm was exactly in line with the natural history. In the treatment arm, with a single injection of AXPAXLI, one injection, after 48 weeks, that vision-threatening complication rate was literally zero, literally zero. And that is remarkable. And every patient who had diabetic macular edema improved, every single patient. What we've done is to show you absolutely every eye of every patient in HELIOS. Clearly, we have no competition here because our competition is already thrown in the towel. So the field is completely open for us, and we will go ahead and target non-proliferative diabetic retinopathy and diabetic macular edema. We said that we will meet with the FDA in the first half of this year. That hasn't changed. And after that meeting, we will have more information for you regarding our trial design and other details as well. Okay. And we have one minute left, so I'd love to be able to get the chance to ask you guys this. But what do you think is the most underappreciated aspect of Ocular? You know, I think the part that's well known is the people that we have, the people that we've recruited, how we've managed our capital, so on and so forth. The part that's really underappreciated, I think, is exactly what I spent time talking about today, which is everything that we have done painstakingly and thoughtfully to de-risk these clinical trials in a bespoke manner. And I always tell people, if I was on the other side, the one question that I would ask me that not a single person has ever asked, not only in this company, but the company before, is very simple. It's to say, what are you doing to de-risk your patient population and increase the chance of success of this clinical trial? It's really all about patient selection. It really is. And people don't realize that this disease is extraordinarily variable. I mean, we have patients that we have to treat every two weeks. We have patients that we can get away with treating every six months. They all look the same. We can't tell. We're 75 years behind oncology. We have no genetic markers. We have no anatomic markers. And think about this. If somebody came to you and said, "Look, I want you to invest in a drug for bladder cancer or colon cancer," I mean, you wouldn't just say, "Okay," or "No." You'd say, "What do you mean? What kind of bladder cancer? What stage? What grade? What kind of colon cancer?" It's no different with us, except that we just don't have that type of granularity. When we design studies, the most important thing that we can do is to make sure the patient selection has been done as carefully as possible to increase the chance of success. I think that's the single most important thing that anybody can be asked. That's a very great point. But unfortunately, with that, we are out of time. But I thank you for your time. A great discussion, Pravin, and thanks everyone for listening. Tara, thank you. And thanks, everybody.