EyePoint, Inc. (EYPT) Earnings Call Transcript & Summary

Good morning, everyone. Thank you for joining the EyePoint Pharmaceuticals presentation on day 3 of the JPMorgan Healthcare Conference. I'm [ Won ] Heo from the JPMorgan Healthcare Investment Banking team, and it's my pleasure to be introducing you to Dr. Jay Duker, President and Chief Executive Officer of EyePoint Pharmaceuticals.

Thank you very much, and thanks all for coming and listening this morning. We are a publicly traded company. And as such, I'll be making some forward-looking statements. And if you need any more information on that, please feel free to visit our website. I point as in its heart is a drug delivery company to the back of the eye. Our lead product is called EYP-1901. It's a combination of vorolanib, which is a small molecule tyrosine kinase inhibitor that has activity against all the VEGF receptors, and it's placed into our patented Durasert E delivery system. E is for erodable. This form of Durasert is completely biorotable. We will have recently reported very positive top line Phase II data in wet AMD, which I'll be reviewing for you this morning. We plan on initiating the first wet AMD Phase III trial in the second half of this year. And we also have now 2 other Phase II trials ongoing. The first is called the PAVIA trial, which is nonproliferative diabetic retinopathy, and we are on target to release top line data from that trial in the second quarter of 2024. And just this morning, we announced dosing the first patient in our third Phase II trial, which is called the VERONA trial, and that's for diabetic macular edema, or DME. We would expect to have the top line data from that trial in about a year. In the past few months, we announced a new addition to our pipeline, EYP-2301. This is a combination of the small molecule razoprotafib, formerly known as AKB-9778, and we were able to place it in our Durasert E inserts for extended duration release. This is a type 2 agonist, which should help stabilize the retinal vasculature in serious retinal diseases. Durasert is our patented delivery system. Durasert in its nonerodible form has been FDA approved in 4 products. There's tens of thousands of patients around the world who have had Durasert placed. There's a very good safety and efficacy profile. In addition, the company now boasts a very strong balance sheet. We had a successful equity raise approximately a month ago, $230 million. And as of December 31, 2023, we have over $330 million in cash and investments. This gives us a cash runway through top line data from these 2 Phase III wet AMD trials that I mentioned. This is our pipeline. I mentioned it already. The first 3 targets are all VEGF mediated diseases, for which we're applying EYP-1901. Talked about EYO-2301, and we are also working on putting some GA, geographic atrophy, complement targets into our delivery system. Durasert is quite tiny. You can see one of our nonerodable forms on the left. The list of previous products that are FDA approved and have subsequently been licensed out include YUTIQ, ILUVIEN, RETISERT and VITRASERT. How we made this biorotable is actually quite simple. When you have a highly soluble drug, if you want extended release, you need to slow down the release rate, and we were able to do that by covering the matrix of the drug with a polyamide shell, which is completely insoluble. By removing the polyamide show, we're able to have the core drug matrix release more rapidly. And in the case of EYP-1901, we believe we'll get therapeutic levels of vorolanib for approximately 9 months in humans. The matrix of Durasert E take several more months to bioerode after the drug release is done. Vorolanib really met all our criteria for essentially the perfect tyrosine kinase inhibitor to put in Durasert. It is selective, meaning reduced off target. It fits very well into our Durasert insert. It has great patent protection out to 2037, which potentially could be extended to 2042. And in addition, vorolanib had previously been studied as an oral agent in wet AMD at the time and had the name X-82. And while it showed very promising efficacy through Phase II because of systemic side effects, the program is discontinued. I'd want to note, however, in those Phase I and Phase II programs, there were no ocular SAEs mentioned. Vorolanib should also block PDGF, which theoretically would lead to an antifibrotic benefit, and we recently reported a preclinical study in rodent, which suggested that vorolanib may be neuroprotective. Vorolanib therapeutic doses does not inhibit type-2, which should help with clinical results. EYP-1901, again, is vorolanib Durasert E. This is delivered intravitreally in a standard way that all the other anti-VEGF agents are local anesthesia in the physician's office. This comes in a preloaded injection system. And within the injection system, we can load up to 3 inserts, so we can deliver 3 inserts with a single injection. Our drug is immediately bioavailable in animals, we can detect to drug levels in the cord within minutes after intravitreal injection, and it reaches therapeutic levels within hours. And it features what's called zero-order kinetics, which means after reaching a steady state in approximately 3 weeks, you get a same release every day, every week, every month for approximately 9 months. This zero-order kinetics allows what's been called micro dosing, which is doses of the drug at what may seem to be less than therapeutic levels can be very effective. For example, our YUTIQ insert contains only 0.18 milligrams of flucinolone yet can control uveitis in 60% of patients for up to 3 years with a single injection. We have positive efficacy data in wet AMD now from both a Phase I and a Phase II trial, and ongoing safety data from 2 Phase IIs and that Phase I trial. And another differentiator for our drug is it can be shipped and stored at ambient temperatures. All the other anti-VEGFs need to be refrigerated, and you can imagine in a retina specialist office right now, the refrigerators have to be quite large because there's a lot of them out there. So the Phase II study was called the DAVIO 2. This is a noninferiority trial comparing 2 doses of our drug against an on-label aflibercept control. The design of the study dates back to a Type C meeting that we had with the FDA in 2022. At the time, we were contemplating going directly from Phase I to a pivotal trial. So we submitted a pivotal protocol and a statistics package to the FDA. There were comments, we made some changes. And at the end of the communication, the FDA had no further comments. We strategically decided to do a more typical Phase II trial, but we took the learnings from the Type C meeting and apply them to our Phase II. And so the Phase II trial design that we initiated as the DAVIO 2 trial is really a result of the Type C meeting. A couple of differences, which I'll go into. In all of these trials, because these are new drug entities, we need to allow for what's called rescue or supplementation of the patients. So if a certain criteria met so that the AMD appears to be out of control, the doctor is allowed to give a supplemental injection in this trial, that was aflibercept. You can see the supplement criteria right there. This is fairly standard. Although one thing we added compared to the Phase I trial was the top bullet, 5-letter loss with 75 microns of new fluid. This is a pretty tight control supplement. And while we wanted to do that to make sure the primary endpoint, which is noninferior visual acuity was met, theoretically, it may have led to an increased number of supplements. But as you'll see in the results, it didn't do that. So the primary outcome for this study was change in visual acuity from day 1 to a blended week 28 and week 32 endpoint. The reason we blend the endpoint is because the FDA asked us to. In the pivotal trial, it will be a blended endpoint as well. The noninferiority margin for this trial was minus 4.5 letters, and that's again what's been stated to us and publicly by the FDA for wet AMD noninferiority trials. And there are other secondary endpoints, which we look at as well. This is the study design. There's a couple of unique things about it. First of all, all the patients received a reload of aflibercept at the beginning of the study. So on day 1, week 4, week 8, all patients received an aflibercept injection and the 30 minutes later, the same visit, they either received their dose of EYP-1901 or sham. They were then watched monthly. In every other month, the eyes that were assigned to the aflibercept arm received another aflibercept while the EYP-1901 eyes were received to sham. All 3 groups were subjected to the same supplement criteria, and therefore, this ended up being the first trial that actually had supplements in the aflibercept arm. This is a top line baseline characteristics from the trial, reading across, again, the aflibercept at 2-milligram dose; EYP-1901 at 2 milligrams, which is 2 inserts; EYP-1901 at 3 milligrams, the high dose, that was 3 inserts. And as you read across, you can see is very well balanced. The bottom line shows you the main number of injections that the patients received in the year leading into the trial. And this was normalized. So for example, if a patient only had wet macular degeneration for 3 months, and received 3 injections, their normalized number of injections was 12. And the real world, on average, in the United States, doctors give about 6 injections a year. So these numbers, I would argue, possibly overinflate the previous treatment rate. But you can see this is a pretty heavily pretreated group, with, on average, 10 injections prior. This is a summary of the top line results. We achieved all endpoints and in particular, the primary endpoint of noninferior visual acuity was achieved. The 2-milligram dose was minus 0.3 letters worse than Eylea. The 3-milligram was minus 0.4 letters worse than Eylea. Remember, on the eye chart, there's 5 letters to align. So this is less than 0.5 a letter difference. And statistically, if you apply statistical analysis to these groups with paired statistics, the p-value for these being the same was 0.0009. In other words, the 3 visual acuity results were essentially statistically identical. There were no EYP-1901 related ocular or systemic SAEs and reduction in treatment burden in the first numbers mean how many injections to the patients get leading in versus how many patients -- how many injections the patients get after we gave them EYP-1901, you can see significant reduction in this historical measure. But we also measured it prospectively, meeting the EYP-1901 arms against the mandated Eylea and again, a very, very nice reduction in the treatment burden. The other secondary outcome was about 2/3 of the patients could go up to 6 months without a supplement. And the last anatomic control refers to measurements on optical coherence tomography, or OCT. If you're not familiar with that test, think of it as an ultrasound using light. We do this in the office. It's quick. It's noninvasive. And it's a measurement of the thickness of the center of the retina. Normally, it's up to about 280 microns, the standard deviation is about 10 microns. And if you get an excess of VEGF in the eye, you get additional thickening of the retina. It's a biomarker that we use to measure VEGF activity. Again, this is top line result, a reading across the top notice that all 3 arms gained about a letter of vision during the study. The reason primarily for that, as you'll see, was the load, the 3 injections that were given at the beginning. So one could argue that as a group, these eyes were probably slightly undertreated, but a change in one letter is noise. Patients would notice it and a doctor, a retinal physician, wouldn't make a big deal about 1 letter improvement. Difference from the aflibercept, I mentioned. This was 95% confidence interval that we achieved statistical significance. And the non-inferiority margin is minus 4.5 letters for wet AMD. What that means is our drug, compared to the control, there's confidence intervals around the division of our drug. And the lower limit of the confidence interval can't cross minus 4.5 letters. In this trial for both arms of our drug, the lower limit of the confidence interval didn't cross minus 3 letters. So we were nowhere near the cutoff for the confidence interval telling us a very tight statistical result. This is a graphic representation of the visual acuities. Notice in the first few months all of the groups improve a little bit. That's from the load. And then there's some bouncing around, which you're going to see random in nature, but really they all hover about one letter at the end. This is on a scale of 5 letters, and again, 5 letters to a line of vision. And so you can see at the end, there, all 3 of the colored dots are almost superimposed. From a safety perspective, we've seen no reports of EYP-1901-related ocular or systemic SAEs. As of the data cutoff, we had 4 ocular SAEs reported in the study eye, but none redeem related to our drug. Of the AEs, over 97% were mild grade 1 or grade 2, and most of them are the type of AE you would expect from an intravitreal injection, things like a subconjunctival hemorrhage, a bleed in the white of the eye, irritation to the eye or floaters. We had no insert migration into the anterior chamber. We wouldn't expect to see that. We don't see that with other forms of Durasert. But that's something that has been a problem for some of our competitors. So we make a note of that. And of course, in the retinal space right now, everybody is a little bit nervous about retinal occlusive vasculitis. We didn't see that either, nor should we. The matrix that holds our vorolanib together has been used in tens of thousands of eyes. There's nothing in that matrix, which has ever been recorded to cause inflammation. And vorolanib is a small molecule. It's not a biologic. And there's no reason to believe that these small molecules would lead to this type of problem. We also featured actually really low patient discontinuation rate. On average, in wet AMD trials, the discontinuation rate is about 10%. Ours was 4%, and none of them were related to our drug. So I wanted to go through the 4 ocular SAEs that were reported in the study eyes. The first was a retinal detachment that occurred 1 week into the trial, so 1 week after the initial aflibercept injection. So there was no drug. Our drug was not in that eye that was repaired surgically, and that eye exited the study. We had a case of bacterial endophthalmitis that occurred at week 32, so approximately 4 months after our drug went in. And this occurred 2 days after an anterior chamber paracentesis. So this is an office procedure where we stick a small needle into the front of the eye to take some fluid out to measure the PK of vorolanib. This patient happened to be on CPAP, and CPAP is a known risk factor for bacterial endophthalmitis after an invasive procedure. And this patient actually we cultured serratia out of the eye, which is a very unusual bug for endophthalmitis and one that would be associated with CPAP. Interestingly, the patient did very well and recovered vision fully. We had a case of noninfectious mild endophthalmitis, which occurred at week 29, again, 7 days after an aflibercept injection. That also cleared completely. There was a case of retinal tears in an eye that occurred at week 36, 4 weeks after aflibercept injection, and that was in an aflibercept arm. Just a table of the AEs, again, generally mild and self-limiting. Across the top, you can see on a percentage basis, there were more ocular AEs in the 2 treatment arms. However, when you drill down and say what was the reason, you can look at the top 3. First of all, worsening AMD. In the trial, about 1/3 of the eyes in the 1902 arms required supplement. That's worsening AMD. And that's something that the doctor decides it's either there or not there. Only 3 of the eyes in the aflibercept arm required rescue. That's the difference. Subconjunctival hemorrhage, I talked about. It's really something that occurs after any intravitreal injection, and it's of no consequence. And there were some cases of vitreous floaters more in the 1901 eyes than in the control and those generally occurred right after the inserts went in and generally were self-limiting. Just to level set here. This slide shows you the percentages of ocular AEs in Phase II from both VABYSMO and high-dose Eylea. So you can see ballpark, it's all about the same running between 40% to 50%. I will say in the HD Eylea study, they only reported treatment-emergent AEs. They didn't report all of them. So the way we reported it, those -- if they reported the same way, it might have been a little bit higher. In summary about the safety, we've treated now about 170 patients with our drug, and we have no ocular or systemic SAEs related to the drug. There were about another 150 patients treated with oral vorolanib during those studies, and again, none -- no ocular SAEs reported, there were certainly systemic SAEs from the oral treatment. This is the reduction in treatment burden figures, about 85% in the 3-milligram and 89% of the 2-milligram, and this is historical against what they received leading into the trial. This is the same type of calculation, but this is more prospective. All the patients in the Eylea arm were to receive 3 injections over 6 months. And so we measured against those 3 injections, you can see it's still about an 80% reduction in treatment burden. But notice the aflibercept arm didn't receive 3.0 injections, it was 3.3 because 3 eyes or 6% got rescued despite getting 3 injections. Also note that you can't be rescued on a day that you're getting a mandated injection. So the aflibercept eyes could only be rescued 3 times, the 1901 eyes could be rescued 6 times. So again, quite a difference here. So this really, again, I think, speaks to the ability of our drug to keep patients stable. About 2/3 of the eyes could go up to 6 months without a rescue. And if you allow and ask the question of how many received just 0 or 1? It's 80% of the eyes got either no rescue or just 1 rescue. This shows you at each visit percentage of eyes rescue free. A 2 takeaway messages from this slide. The first is it's been hypothesized by others that tyrosine kinase inhibitors are slow to work in the eye. We don't believe that's true of our drug. If it was slow to work, then you should see an excess of rescues at month 3 and month 4 til the drug got going, and you don't see that. The second thing is the curves are nice and smooth, which suggests that we still have drug working after 6 months. If the drug had run out in 4 months or 3 months, I think you'd see a sudden dip in the number of rescue-free eyes. These are the anatomic results. And remember, I said that normal is up to about 280 microns. This is a generally well-controlled group going in. Remember, it's -- these are eyes that were previously treated, so they were on pretty good control anatomically. And noticed there was some gain in central subfield thickness in all 3 arms. But the difference between the Eylea and the 1901 arms was under a standard deviation of the test, very, very small. And this shows it graphically. At the beginning, noticed all the eyes got a little bit dryer because of a load. And then between month 3 and month 4, they all gained a little bit of CST, not much. The difference here though is our eyes remained very stable, where we see that typical sawtooth pattern that we see from every-other-month Eylea. That was noted in [ view1, view2 ] and retinal specialists in general, ignored it because it didn't affect the visual acuity. But what most retinal specialists want is the ability to extend out the number -- lower the number of injections, extend out the intervals without sacrificing vision or anatomy, and that's what we showed in this trial. We did a subgroup analysis also, which I want to show you. There was a question that came up, well, perhaps did your eyes do so well because of the supplements? The supplements raised their vision. And it turns out the opposite was true. This slide shows you all the eyes that were unsupplemented at all 3 groups up to month 6. And the 2-milligram dose was 0.6 letters better than Eylea. The 3-milligram dose was 0.1 letter better than Eylea. So quite the contrary. The unsupplemented eyes actually did better than control. This is the OCT of the unsupplemented eyes, again, showing similar findings, very flat anatomy, also telling you, at the end of month 5, month 6 after our drug went in, it's not as if the drug stopped working and the eye started gaining fluid, they were perfectly stable. So to summarize, this slide shows you the results that I've gone over before. And on the right side, some of you may know that prior to the results, we had set some floors as to what we wanted to see from the results going forward, and we did much, much better than those floors. So I want to now shift to a brief overview of our plans for Phase III, and these are plans at this point. But remember, this is all based on the Type C meeting that we had with the FDA and subsequent comments that the FDA has made. Two or actually even 3, I would say, big differences between DAVIO 2 in the pivotal trials. The first is the length of the study. The pivotal trials will have a primary endpoint of approximately 1 year. It will be week 52 and week 56 combined, not 8 months as we did in DAVIO 2. The second big change is we want to get a label for reinjection. And therefore, we will be reinjecting in the pivotal trials, and we plan on doing it every 6 months. The third difference is the number of inserts that we injected in the high dose in DAVIO 2 dose 3 and the low dose was 2. But I think as you go through that data, you can agree that there really wasn't a dose response. And frankly, many people in the company didn't think there was going to be a dose response. It's once you're enough over the IC50 at 0-order kinetics, you've got the receptor blocked and it's going to stay blocked. What that tells us, though, is there's no reason to test 3 inserts in the pivotal trial. We're going to go forward with 2 inserts and 1 insert. The actual payloads have yet to be determined because we do have some flexibility around payload. We expect to start the first trial in the second half of this year. It will be U.S. and Canada. The second trial will be primarily OUS, and that should start several months later. This is an outline of the plan for that trial. It looks very similar to the outline that I just showed you for DAVIO 2 with the exceptions that I just mentioned. So the NPDR trial called PAVIO, again, readout is coming next quarter. PAVIA enrolled 77 patients with moderate-to-severe non-proliferative retinopathy. And the primary endpoint in this study is not visual, it's structural. A photograph is taken over the retina at day 1, and then photograph is taken at month 9. And a reading center independently evaluates the degree of retinopathy on a scale called the DRSS, and the standard is a 2-step reduction in the DRSS. So that's what we're looking for at the 9-month visit with secondary endpoints trying to prevent diabetic macular edema and proliferative diabetic renopathy. So we should have that, again, results in the second quarter. However, from a safety perspective, just to update you, again, of the 77 patients, probably approximately 50 got our drug and that there had been no ocular or systemic SAEs related to the drug. We had 2 ocular SAEs, a hemorrhagic posterior vitreous separation, which is a common occurrence in diabetes and eye getting an injection. That was 8 weeks after the study eye. And again, we don't know whether that eye received the drug or not, and we had a fellow eye get macular edema. The VERONA trial is the name of our diabetic macular edema trial. And actually, we're proud to say we dosed the first patient yesterday. And so that trial is relatively small, plan to be 25 patients, 10 in each dose of EYP-1901 in a 5-patient control group, getting aflibercept. Everybody gets an aflibercept shot on day 1, and then they either get a sham or they get our drug, and these are diabetic macular edema patients who have macular edema. So they will -- patients completely drive would not be in this study. The primary endpoint is the time to first supplement. And so what we're looking for is a difference between the 1901 eyes and the aflibercept eyes for time to supplement. And this is the supplement criteria that we're using. Starting at the first month you can see that's actually pretty similar to what we use for wet AMD. There's one additional supplement criteria here, which is after 3 months, if the eyes haven't started to show anatomic improvement, we're going to supplement them also. So the last topic I want to cover is razuprotafib in Durasert E or EYP-2301. For those of you who have may have been following this Aerpio was the name of the company that was developing this. They called it AKB-9778, and they were developing it in diabetes, diabetic retinopathy, diabetic macular edema, but were delivering it subcutaneously. It's a tie 2 agonist, so it should stabilize the retinal vasculature, work well in conjunction with anti-VEGF. They show really terrific anatomic data, but the visual acuity data was not as good, and therefore, they didn't continue the program. Our scientists were able to reformulate razuprotafib to put it into Durasert E, and we believe we can get therapeutic levels for 6 months or longer with a single injection. Our balance sheet looks great. Equity financing successfully about a month ago. And again, we have about $330 million right now in cash and investments, which will take us through multiple key data inflection points and should take us through top line results of both wet AMD pivotal trials. One of the things that I'm really proud of of this company is we've really executed well. The first patient ever dosed with our drug EYP-1901 was dosed in January of 2021, and we have robust top line Phase II data in December 2023. You can see all the green checks. There are things we said we were going to do and we did. We have a few more boxes to check in 2024, as you can see, but I'm optimistic that the team will check the boxes. Thank you very much, and happy to entertain any questions. I just also would like to introduce our CFO and EVP, George Elston, who is right next to me.

Thanks again, Jay, for such a wonderful presentation. And we just want to thank this -- take this time to thank the broader EyePoint team here sitting with us, Nancy, George, Michael and Aaron, just for your commitment and innovative and sustainable ophthalmology. So thank you so much. Now moving on to Q&A. Please be reminded, for those of you online, you can feel free to submit those. I can read them off to you. And then for those of you in the audience, please feel free to raise your hand.

Was it that good?

Very good, but I'll kick it off with just one question from my side. Again, I think the DAVIO 2 results were definitely phenomenal. I think we can all universally agree on that. As you get ready for the next sort of phase of development, 2024 seems like a very catalyst-rich year as well as 2025. Are you thinking about sort of -- in terms of BT, are you thinking about sort of partnership in any way as you gear up for commercialization in the future? If so, is it regional? Or is it sort of more of a platform play? And what is your selection criteria in choosing the right partner for EyePoint?

Terrific questions. And I could say, our strategy around that has changed markedly since the data and the fund raise. I think if you'd asked the question a few months ago, I think my answer would have been, given the macroeconomic environment and given the uncertainty around our data, we probably would seek a partnership for ex U.S., both clinical and commercial to help fund the studies. We don't need that anymore. We can fund the studies ourselves. So that at this point, we're not really seeking a clinical partnership. However, we're a small company. And while I am quite confident in the company's ability to commercialize and market the drug successfully in the United States, we were a commercial company only 7 months ago, and much of that commercial team is still with us. The idea of commercializing the anti-VEGF globally, ex U.S. is daunting. And I think at some point, when the time is right, we would seek a global strategic partner to help us with the commercialization of EYP-1901.

Maybe one more question on perhaps sort of in-house manufacturing and so on. What is your plan for that?

So, again, excellent question. So we manufacture EYP-1901 ourselves. There's quite a bit of technological know-how that goes into this. And we do that right now at our headquarters in Watertown, Massachusetts. We have the capacity at Watertown to make EYP-1901 for the first 2 pivotal trials and probably even the pivotals beyond that. However, we recognize that, that capacity is going to be limited. And as a result, we initiated the start of a manufacturing facility almost 2 years ago, that process started. So we have a manufacturing facility under construction about 45 minutes west of our headquarters in Northbridge, Massachusetts. It's a 40,000 square foot facility. And when it's online, it should satisfy all the commercial EYP-1901 needs, along with we're still a supplier for YUTIQ for our partner Alimera in the United States, and our partner Ocumension in China, and that would be easily to take care of that as well. The building is up. It's almost completely enclosed. We probably have 6 more months of construction. And then, of course, we needed to get cleared and approved by the FDA, but we are on target and under budget with that facility, and we are confident that we'll be able to the commercial needs there. One more thing that I'd like to point out and tip my hat to George is we have to -- I got to wear a lot of hats and I got to keep tipping to George. He's done such a great job. But we have a really, really good lease, we don't pay rent until 4 months after we take occupancy. Our landlords wanted to really try to build a biotech hub in Northbridge, and we're the first tenants. And so we got a terrific deal. They are -- all costs are -- they're paying for upfront, but it's to our complete specifications.

George, one last -- sorry, thanks again, Jay. One last question for George. I think we touched on a little bit about the strength of your balance sheet as you get ready for, again, catalyst for its 2024 and 2025. Can you just elaborate a little bit more on where your balance sheet stands and how you're funded?

Yes. So we updated this morning in an 8-K that we have $330 million of cash and investments at 12/31. Very happy with our balance sheet. We paid all of our debt off last year. Our cash guidance is through top line data in both of the wet AMD Phase III trials, which is into 2026. And it also includes covering all of our ongoing Phase IIs as well. So we're very well capitalized through these key catalysts.