To the second day of the TD Cowen 44th Healthcare Conference. For this next session, we are very pleased to have a fireside chat with EyePoint Pharma. From EyePoint Pharma, we have Jay Duker, President and CEO, as well as George Elston, Executive Vice President and Chief Financial Officer. Jay and George, thank you very much for being here.
Great to be here, Mark. Thanks for having us.
Of course. So before I get started, for those of you in the audience, if you guys have any questions, feel free to raise your hand or speak up during the fireside. Want to get you guys involved too. But Jay and George, you guys announced the appointment of a new Chief Medical Officer, so perhaps you could start by saying a few words about that.
Sure. So this was a process that actually took over a year. We found somebody, Ramiro Ribeiro, MD, PhD, who really met all our criteria. Up until Monday when Ramiro started, I was the only one in the company who was an ophthalmologist and a retina specialist, and now we have two of us. From a management perspective, I think it's very easy for me to sit in a meeting, being the only retina specialist and the CEO of the company, and not hear any pushback when I say things. And now I have somebody who's going to push back against me, which I really appreciate. We really also needed somebody who had experience running phase 3 retina trials, XUS, and Ramiro fit that as well. He has a PhD, which also I believe is going to help us substantially with our pipeline.
Our previous CMO, terrific guy, great employee, but he was hired at a time when the company was very different. We were a small spec pharma commercial company, and Dario was really brought in to assist with a company at that level. We've changed dramatically over the last couple of years, and especially since our good data. Adding people and adding expertise that's going to enable us to be successful in our future endeavors is really important. I think you'll see more of that.
Great. So, let's go ahead and get into EYP-1901, and we'll start with wet AMD. You guys reported great results with DAVIO 2 in December, but for those who are less familiar with it, maybe you could start with a high-level overview of those results.
Sure. And, again, starting with what EYP-1901 is, it's a small molecule vorolanib, which is a TKI, tyrosine kinase inhibitor, which has activity against all the VEGF receptors. We put it into Durasert E. Durasert is our proprietary drug delivery system, been in for FDA-approved products. Durasert E is the version that's now fully bioerodible. The DAVIO 2 trial was a Phase 2 trial that enrolled 160 patients. The trial consisted of three arms, two doses of our drug, approximately 3 milligrams and approximately 2 milligrams, dosed once, against an Eylea control. Eylea was dosed on label, which was monthly times three, and then every other month. The primary endpoint of the study was change in visual acuity, non-inferiority. That non-inferiority margin that was -4.5 letters, which is what it's been for all the last four FDA-approved wet AMD products.
One other thing that was a little bit different, the only other study that did this was port delivery. We enrolled previously treated patients, not treatment-naive patients, because we're going for a label of maintenance therapy. As you had mentioned, although the results really were beyond our expectations, we essentially tied Eylea, with change in visual acuity. Our 2 milligram dose was -0.3 letters worse at 8 months. Our 3 milligram dose was -0.4 letters worse than Eylea. And when you look at the p-values for difference between those values, it was 0.0009 or less, so essentially equivalent. And we did that with a very clean safety record. There were no ocular or systemic SAEs related to our drug. We had about 2/3 of the patients make it up to 6 months after our drug went in without any rescue.
The anatomic data was also very impressive in the sense that the OCTs, which measure the amount of fluid in the macula, were quite stable, and only a small amount less than a standard deviation, less than the Eylea arm. We hit all the secondary endpoints, that we sought to.
So when you say the inserts fully bioerodible, what does that mean? When does the insert disappear, and when does the drug stop being used?
Yeah. So, that's a good question. And first of all, the inserts that we're currently using and will be used in the pivotal trial are over 90% drug by weight and volume. On average, and we believe in humans, that they should last up to about nine months with therapeutic doses. So at nine months, the drug is completely eluted from the inserts. So they're over 90% gone from the perspective of weight and volume. The matrix that holds the inserts together was designed to last longer. And the reason is you want to control drug release till the end. If your matrix goes away first, you've got free drug in the eye, and you're no longer controlling the release. So they were designed this way.
We believe in humans, they should go away in about total 18 months after that last little nub in a matrix, but they're really hard to photograph. So when people ask me, "How long do they last in a human?" the answer is, "We haven't been able to really photograph them well enough to really know for sure." We certainly know in animals and in the laboratory that those timing of these inserts that we're using now, of how they go away. And it's variable. When we talk to the Phase 1 investigators, some of them can see them for, you know, even almost two years out, different percentage of matrix in those, however. And other ones after a year, they say, "I can't see them anymore." So it's variable. But we believe they should, in humans, be completely gone in a certainly 18 months.
If you actually take it out to 24 months and you do the calculation of if you go forward with 2 inserts as your dose, and you dose it every 6 months, which is what we expect to, you end up at a steady state, assuming 24 months for that last matrix to go away, of less than 3 inserts, volume and weight long term, which should be completely acceptable as we've dosed with 3 inserts at about 70 eyes so far.
Yeah. And maybe for a volume reference, each insert's about 1/5000th of the vitreous. So, you know, lots of, lots of capacity.
Yeah. These inserts are solid, and they're heavier than water, so they sink to the bottom of the vitreous cavity, and they basically stay there. They also stick together, which is an advantage, because when you inject new inserts into an eye that have these small nubbins of, you know, matrix left, they find each other and stick together. So there's not free floating. They're not migrating. They don't break into pieces. They don't emulsify. There's no P-PLGA in them. There's no hydrogel in them. There's no PEG in them. So some of the issues that have come up with other products in the retina, we don't have those same issues.
Again, when you dose with a new insert, the two or three other remaining ones. At that point, there will be a little less than there are.
Two-thirds gone. Well, at six months, call it two-thirds gone. At nine months, over 90% gone. And so, there should be a steady state that gets reached of just less than three inserts.
Okay. That's clear. Can you describe your current thinking on the Phase 3 trial design? I guess you called it Lugano. Now, I would love to hear a little background on that, as well as the dose you might be exploring.
Well, the background on the name?
Yeah.
Oh, that's easy. Well, you know where DAVIO came from?
I should, but no.
No. Well, again, it's Durasert and vorolanib had been ophthalmology.
Got it.
But it sounded like, for those of you living in Boston, there's a Davio's restaurant, Italian restaurant. So we went with the Italian theme, but it wasn't named after the restaurant. So we've named our trials after various Italian cities and/or lakes. Lugano and Lucia are the names of the pivotals. So nothing, nothing bigger than that. Anyway, the phase 3 trial design, if you look at our phase 2 trial design, the DAVIO 2 trial, the phase 3 trial design, as we plan it now, has three, I would say, differences, big differences. The first big difference is the primary endpoint is going to be at one year, not at eight months. That's mandated by the FDA. Also because we are going to load everybody with Eylea like we did in DAVIO 2, which pushes the endpoint out to a year.
Second change, we are re-planning on redosing every 6 months 'cause we'd like to get a label for redosing every 6 months. The third change has to do with the number of inserts. In the DAVIO 2 results, the 3-milligram dose, which was 3 inserts, and the 2-milligram dose, which was 2 inserts, performed almost exactly the same. There was no dose response. So we're going with 1 insert and 2 inserts as our two doses in the pivotal trials.
Interesting. I believe you're going to be discussing this with the FDA during an end-of-phase 2 meeting. Can you just talk about the timeline on that?
Sure.
When will we get an update?
Yeah. So our end-of-phase 2 meeting has a date, April 23rd. Briefing book has been submitted, and we remain confident that that trial design will be acceptable. And part of our confidence comes from the fact that we had a Type C meeting for a pivotal trial design with the FDA already in 2022, and they've already seen this basic trial design. In fact, the DAVIO 2 trial was based off what we had come to an agreement with the FDA on a potential pivotal trial, and so we're optimistic that the FDA, you know, has already seen the trial and said, "This looks fine," and that was before we had the phase 2 data. Now that we have phase 2 data showing safety and efficacy, we feel even better about how they'll approach our trial design.
We expect to, you know, have a public statement about the FDA meeting after the minutes come out, which will be approximately a month or end of May, approximately.
Okay. How large might this trial be, and how much might it cost, and how long do you think it'll take?
How large? Well, we've submitted a protocol to the FDA with some numbers. We'll see if they agree. Right now, I think the upper limit for size of the trial would be 400 patients, 200 in the high dose, 100 in the low dose, 100 control. From the statistics of DAVIO 2, we could go smaller than that and have a high likelihood of meeting the non-inferiority margin. However, we're not going to. The rate-limiting step for us is probably the safety numbers. The FDA insists for wet AMD that you have no less than 300 patients evaluable for safety at the go-to-market dose or higher. And you can't be less than that. If you're 299, out of luck. So we have to pad that.
So in the high dose, it's likely we're going to have 175, 180, maybe up to 200 patients in each trial to make sure we get that number. The lower dose in the Eylea arm probably don't need that many. And we'll probably have a 2:1:1 type randomization. But this is all, again, dependent on agreement with the FDA. Cost of the trial, George?
Depends on the end. But I think, when we, when we did our financing in December, you know, we, we had modeled out sort of worst-case scenarios. So with the cash we raised in our current cash guidance into 2026 assumes funding of both phase 3s. It's including, so the LUCIA trial, which is the OUS trial, includes funding for that program as well.
Okay. So if it's 400 patients, Jay, do you think that it'll take kind of around 3 years to get top-line data from start, like you saw with historical wet AMD trials, or?
Probably a little less. We think we can enroll the trial in under a year, certainly. We have hopes to do it significantly under a year. We enrolled DAVIO 2 with 160 patients in about eight months. We learned a lot from DAVIO 2 about enrollment. We're going to take those lessons and apply them to the phase 3. There's a lot of reasons why. Remember, I've talked about having that Type C meeting with the FDA. That was in anticipation of potentially going right from phase 1 to pivotal, which we elected not to do. We've learned a lot from the phase 2 and applying those lessons to de-risk the phase 3. In saying that, we had 70 sites in DAVIO 2. It doesn't mean we'll use all of them, but many of them will be in DAVIO 3 or Lugano.
It's a little bit easy to flip a site that already knows your drug, already had IRB approval, already seen essentially the same protocol. So we think we're going to have a jump start for the phase 3 to really get enrollment going. So we think, you know, our top-line data should be mid-2026, George.
Roughly. Yeah.
Okay. Just to again confirm, you said 1-2 inserts? So 1-2 milligram doses for the Phase 3?
Minimum. So we have tested three payloads so far with our inserts. We tested 440 micrograms. We've tested 1 milligram. And in our DME trial, we're testing 1.3 milligrams per insert. So we could, with two inserts, that comes out to about 2.7 milligrams as the high dose. 1.3 is the low dose, but that, again, we're not going to go below 1 milligram because 2 milligrams worked. We don't want to have our high dose be less than that. But we have some optionality. And that's, again, a little bit of a discussion with the FDA. But I think it's safe to say it'll be the insert payload will be between 1 and 1.3 milligrams of drug.
Okay. That's clear. As you know, a competitor is running up two single-dose treatment extensive priority trials that they believe will be sufficient for approval. So is this a design that you guys might still consider in your discussions with the FDA?
Well, we considered it. I, I think it was behooved us to consider all different, different trial designs. But in talking to our current FDA consultants, and looking at our phase 2 data, we concluded that a more traditional non-inferiority trial suits us much better. Remember that the last 4 approvals in wet AMD have been with non-inferiority trials, and our non-inferiority design is actually simpler in some ways than the last two because we're only testing one interval. We're not altering the intervals. We're going to ask for a 6-month interval, and that's what we're going to test. We're also going with 2 doses in our phase 3, which the FDA commented on in our Type C meeting.
The comment was, "Using a sham to use for your masking is acceptable because you're using two doses of your drug." The other reason to use two doses, by the way, is we didn't see a dose response in the Phase 2. 2 milligrams worked. We don't know what the lowest dose would be where we would start losing efficacy. But we'd like to test statistically the use of 1 insert, because obviously, from a COGS perspective, going to market with 1 insert would be preferable. So we expect to statistically be able to test that in our pivotal trials.
Okay.
Yeah. I think, Tyler, one more point on that. If you look at speed for recruitment, you know, there's about 12 previously treated patients for every newly diagnosed. And so much larger population to draw from, as we think about enrollment.
Okay. Very helpful. Any other wet AMD questions before we get to NPDR? All right. So NPDR, it's an indication that most people, or even those who do work on ophthalmology on Wall Street, aren't that familiar with. So maybe you could start by giving a brief overview of the disease, the history of development, and how these patients are treated.
So NPDR, non-proliferative diabetic retinopathy, is a disease that approximately a third of diabetics eventually get. It's changes in the retina that are due to these vascular problems. Early on, it's asymptomatic. As it gets more severe, patients can have mild symptoms from it, but it's a harbinger for the serious complications of diabetes in the retina, which are diabetic macular edema and proliferative diabetic retinopathy and the growth of new blood vessels. So it's very clear that if you can, slow down or halt the progression of non-proliferative diabetic retinopathy, you can have a beneficial effect on the number of patients who go onto these sight-threatening complications. There are two FDA-approved products, Lucentis and two milligram Eylea. The problem is no one's using them because the frequency of injections in this patient population is not something that doctors and patients in general want to take on.
These tend to be younger patients who are working class. They're diabetics. They have 1 million doctors' appointments. Most of them, the cadence of their eye visits are maybe once a year or once every six months. It's a big ask to have them come in when they're relatively asymptomatic, to have injections performed six times a year or so to prevent a disease that would be treated with 10 injections a year. We think this is, kind of a greenfield for sustained release, which is if we can offer a similar safety and even maybe slightly less efficacy, but a schedule that's every six months or every nine months or once a year, I think it's going to be much more acceptable to patients and doctors 'cause it's the cadence that they're being followed anyway.
Okay. And you guys are focusing on moderate and severe NPDR patients.
Yeah.
So how do you define moderate and severe?
Well, moderately severe to severe is all defined on this diabetic retinopathy scale, the DRSS. This has been around for a while. It's a way to gauge the severity of diabetes. The scale was developed many years ago. It's been shown that the more severe you are on the DRSS scale, the more likely you are to have visual loss or significant complications. So the definition is based on this scale. It's from a fundus photograph, which is a picture of the retina taken in the office. So unlike what AMD, where the primary endpoint is visual acuity, that's a functional endpoint, how the eye sees. This is a structural endpoint. What's the structure of the retina? So interestingly, it's not like you can get the results off a spreadsheet. The results are tabulated by an independent reading center.
We send the pictures taken in the office at day 1. Then they're taken every 3 months. At month 9, the reading center compares the pictures from day 1 to say, "Has the patient improved on that DRSS scale? And if so, how much?" Two steps on the scale is the approvable endpoint. Although just recently, the FDA has agreed for a program to go forward with failure to lose 3 steps on the scale. So that's another potentially approvable endpoint. The primary endpoint in our PAVIA phase 2 trial, though, is 2-step or more improvement.
Okay. Is failure to lose 3 a secondary, or is that something you'd have to consider moving forward?
That's a really good question 'cause I don't know offhand if it was a secondary in the study. But, but if we were to hit it, and we were to hit it in high statistics, and we thought that might be a quicker pathway to approval, it looks like the FDA is fine with it. So TBD.
What percentage of patients need to have a two-step improvement on that to be competitive?
Well, to be competitive, I'd say any 'cause there's no competition right now.
In line with.
That's the beauty. But what I would say is we let our KOLs answer that question. So we asked the retina community, "How many patients in order if you had a Q9 month drug that was safe, what is the percentage of patients that improve to or more steps for you to use it?" And the answer is about a third.
Okay.
So the bar is relatively low there if you are safe and you can do sustained release.
Okay. What about the incidence of DME in the trial? Is that going to be an important thing to look at, and what would you expect?
It is. It is. So, active DME was an exclusion from the trial. But some eyes will convert because remember the arm, that's the control arm, is essentially observation. So we would expect in the control arm that some of them will develop DME over the course of the trial, and it could vary. You know, if you look at the previous trials, some had as high as almost 40% require injections for either DME or PDR. So we will be looking at that as a secondary endpoint.
Okay. How big do you think NPDR can be, relative to what AMD, if you guys are able to develop it into a market?
Market size.
Yeah.
Yeah, boy. If I had a crystal ball like that, I'd probably be playing the lottery. I mean, seriously, it's really hard to know because the doctors who you speak to, many of them will say, "Sure. I'll use that." But that's theoretical. We think it could between let me put it this way. If we get approval for DME and NPDR, it's over $1 billion for diabetic diabetes. Our models are very conservative. And so in our models, we would say that, you know, most of that would be DME. But we think the conservative modeling, it could do much better than that.
Okay. You mentioned DME. It's a perfect segue to the phase 2 VERONA trial.
Mm-hmm.
Maybe you could talk about the design of that and the status of it and what we should expect from the top-line release early next year.
So actively enrolling, we would expect enrollment to be complete by the end of the Q2 of this year, and then results probably 7 months or so after that. So we're guiding towards Q1 of 2025 for the results. 25 patients randomized, 10 in each of the 2 milligram, I'm sorry, the 2.7 milligram arm and the 1.3 milligram arm against five patients in Eylea control. Everybody gets Eylea day one. Then they either get a second sham or they get one of our drugs. And then they're watched. And what they're watched for is worsening of DME or failure to improve. And the primary endpoint is time to first supplement. So it's not necessarily visual acuity here, although we're going to be measuring that. And we're also going to be measuring DRSS in this study.
It's to see if our inserts can control DME or improve DME better than a single shot of Eylea in this population previously treated but with retinal thickening. There's no dry patients being enrolled.
Okay. So you said, 1.3-2.7, those are the doses you're using. So that's the.
Yeah. One insert, two inserts.
Yep. Yeah.
And.
So that's the new formulation, right, that you guys are looking at?
Well, technically, it's not a new formulation. The FDA doesn't view it that way.
New dose.
They've, yes, payload. The FDA has looked at our data on all of these. Because the core matrix of the drug is the same in all these inserts, that they don't view it as a formulation change. But it's a payload change. Yes. So we're looking at that in Verona right now.
Interesting. So that data and DME could potentially inform what we might see in the wet AMD phase 3 in the future, right, in terms of the.
For payload.
Little different doses for payload.
Yeah. I would say from a timing perspective, we're probably not going to wait for all the data. We might see some of it. But as you go through the what if possibilities, there's not a lot of what ifs there that's going to change what we do with payload. The payload is going to be determined based on other things.
Yeah. Yeah. Tyler, I just want to make one point on the, the VERONA trial. Remember, there, we've gotten questions about TKIs and vorolanib bioavailability. 1901 is immediately bioavailable. We get a burst of drug after an, injection and then zero-order kinetics. So that's really critical as we look at the VERONA trial to time to first supplement because, you know, with our program, 1901, vorolanib is there, at therapeutic levels very quickly.
Yeah. I mean, in a way, it's almost like a superiority trial on that endpoint, right? Time to first supplement.
It essentially is. That's not the way the FDA might approve it in a pivotal trial. And we understood that from the start. This is a proof of concept. And then the pivotal would be designed around change in visual acuity as a primary endpoint.
What do you think we might see on time to first supplement with the different arms?
Boy, it's really hard for me to make a prediction. And I don't like to because it's actually not helpful. What I would say, though, in general, is every anti-VEGF that's worked in one indication has worked in all of them. But then you have to drill down and say, "What do you mean by worked? You know, how well?" Early on, diabetic retinopathy is mostly a VEGF-mediated disease. And therefore, we should do well in the patients who have it early on. One of the reasons corticosteroids works well in diabetic macular edema and doesn't work in wet AMD is that it's pretty clear that there are other cytokines in the inflammatory pathway later in the disease that influence it. And therefore, it's hard to predict in a relatively small study how early, how late will these patients be?
But even late in the disease, it's still VEGF anti-VEGFs will work. But if you look at the, you know, the data on, on how many injections before their vision improves or the fluid goes away, it can be quite a few. But what we believe we have this advantage is this microdosing, which is we're giving with zero-order kinetics a small dose every minute of every hour of every day for about nine months. And if you extrapolate back from our YUTIQ insert, which was approved for a very small indication called uveitis, YUTIQ insert had 0.18 milligram of drug, tiny. It lasts for three years. And it can control uveitis in 60% of the patients over three years. So that's the beauty of the microdosing. And so we think vorolanib might well work the same way.
Once you've got that receptor blocked, if you've got drug in that extracellular milieu that can cross and get to the receptor, you're going to keep it blocked. So sustained release should do better theoretically. But you don't, don't know till you see it. I think though, obviously, with PAVIA, there's going to be read-through. It's, it's the same disease at a different stage.
Okay. That's very helpful.
I'll tell you more in a couple months.
Sounds good. So going back to wet AMD, the Lugano phase 3 starting by the end of the year or in the second half, what's the, are you guys still considering a potential partnership for EYP-1901?
So that discussions have changed dramatically as of December 4th with our terrific data and our ability to raise and have enough money to get us out through the two pivotal trials in wet AMD. We don't need a clinical partner right now. And we're not seeking a clinical partner right now. Realistically, I think you know, we're a small company. And although we're growing, we're still small. The idea of us commercializing the drug on our own OUS is daunting. And if you ask me today, I think it's likely that when the time comes, we will get a partner for that. We believe we can commercialize this drug successfully within the United States. Remember, we were a commercial company not too long ago. The retina community is relatively small. We know them.
We kept most of the people who successfully commercialized YUTIQ in the company, to prepare for 1901. So that, that's the current thinking right now.
Okay. In the last minute, if you had to point investors in the audience or on the webcast to one of the other candidates in your pipeline, which one would it be?
Well, 2301, razuprotafib, Durasert E. Razuprotafib was formerly known as AKB-9778. Aerpio had the asset. We bought it from them, including all their IP. They studied it subcutaneously in diabetic retinopathy and NPDR. And they got really good anatomic data. However, because they were combining it with Lucentis at the time, they didn't get a robust visual acuity improvement. But remember, they were doing it once a day sub-Q. We're doing it with zero-order kinetics intravitreally. And we think and again, this program hasn't been fully, you know, worked out yet. But if we were to do it as a combination therapy with anti-VEGFs, and we could achieve superiority and change in visual acuity, I think doctors would use us in virtually every patient. So that's where the program sits right now. Again, we're [pretty confident in] the ability to have it go out 6 months.
We're just getting the final formulations ready for the IND and dosing studies.
Okay. Great. We're up on time. But to close, I'd like to ask you guys, the final question, which is, what aspect of the EyePoint Pharma story do you believe is underappreciated by investors?
I would still say that the ability for a sustained release to enable doctors and patients to control their visit cadence better is underappreciated.
George?
Yeah. I think maybe to add to that, you know, 1901 is not another anti-VEGF. We're bringing a new MOA to the all of these diseases, sustained delivery, you know, zero-order kinetic dosing for nine months. And I think it's really an opportunity to work not instead of but with the current therapies. And if you look at the DAVIO 2 data, you know, we can take two-thirds of patients on a six-month regimen.
Wonderful.