Hi. Good morning, everybody. Welcome to the next session at the UBS Global Biopharma Conference in London. I'm Alice Nettleton, one of the U.S. mid biotech analysts at UBS. Happy to introduce Ocular Therapeutix. With us today is Pravin Dugel, CEO. Thank you so much, Pravin, for joining us. Good to see you.
Alice, thanks for having me. I appreciate your invitation to be here.
Absolutely. Maybe just to level set for the audience, for people who are less familiar with the Ocular story, could you give a quick overview, and then we can move to some Q&A?
Sure. The company is called Ocular Therapeutix. We have a drug that has successfully completed a phase III study. We have been able to show superiority versus an anti-VEGF for the first time ever. We have a result that has a very robust p-value of less than 0.0006, and we're getting ready to submit for approval this quarter. We're on track to do that. The drug is called AXPAXLI. It's a TKI. We believe that we'll be the first and only proper maintenance drug. We have shown in this trial that this drug will last for a year in most patients. Unlike the other drugs that you know about that are bolus injections, and there's a lot of them, whether they be generic or premium drugs, we believe that our drug will be the first and drug that will be approved as a true maintenance drug.
Great. Thank you for the overview. Following the recent competitor phase III miss in wet AMD, there has been some debate among investors around the potential read-through to AXPAXLI, given these drugs are both TKIs. Could you talk about the important differences and why you are still confident in AXPAXLI?
Yeah. The first thing I would say is, look, in anything, I think to classify a class effect for something that fails would mean that we would have no other drugs. We would not have an anti-VEGF, for instance, right now, and we obviously know the anti-VEGFs work extremely well. Remember, one of the first anti-VEGFs to come out there was Macugen, and you know that LUCENTIS is a lot better than Macugen. So if we had a class effect based on Macugen, we would not have any of the other great drugs that we have now. So, that is the first thing. The second thing I would tell you is that, I do not necessarily enjoy speaking about other people or failures, but we absolutely predicted that this was going to happen two and a half years ago. We told you exactly what was going to happen with EyePoint.
We said that their TKI was an inferior TKI as well as VEGF suppression is concerned. We know that from a lot of studies that have already been done that we have cited. We said that their delivery system was inferior. Very, very importantly, and this is extremely important, we said that they had really put no thought into patient selection, and patient selection was absolutely key. We have spent a great deal of time, a great deal of thought selecting the proper patients in a bespoke manner for each study, which they did not do, and we can talk about that in detail. So everything that happened, we absolutely predicted.
We predicted that that drug would have a signal, and it does, and we predicted that it would miss the primary endpoint, which it did. It absolutely has no effect on what we are doing. We hit a superiority endpoint against a very, very good drug in EYLEA with a robust p-value. We have shown everything that we possibly can in our SOL-1 study, and has nothing to do with the class effect whatsoever. Their failure has everything to do with selecting a weaker TKI in a worse format with no thought to patient selection whatsoever.
Do you want to expand on that point then about patient selection and what is different about your trial?
Yeah, I will be happy to. So, I think this is, in my field in retina, I would say that this is probably the most underappreciated thing there is, and let me explain. We are at least 50 to 75 years behind oncology or rheumatology or immunology. We have absolutely no disease stratification whatsoever. We have no biomarker of disease response, whether that biomarker be genetic or anatomic. So, when people come in, they look exactly the same. They respond in a very diverse manner, but they look exactly the same. We have no way of differentiating that. I will give you an example. If I were to come to you and say, "Alice, I have a drug for colon cancer," you would just laugh. You would say, "What kind of colon cancer? What stage? What grade?" Et cetera, et cetera. We have no way of differentiating that.
However, we can still come to you and say, "Look, we have a drug for wet macular degeneration," yet we know that there are some patients with wet macular degeneration, even with regular LUCENTIS, that require treatment maybe once every six months. Others may require treatment once every two weeks. We have no idea how to differentiate them, so we do something called treat and extend, which is essentially trial and error. We go ahead and inject, we see how they do, and then we figure out little by little what cadence to inject them in. You can imagine that when you do huge studies like we used to do for ANCHOR and MARINA or VIEW 1 and VIEW 2, where we had 2,000, 3,000 patients, these kind of imbalances may work themselves out just because of the large sample size.
But when we are doing smaller studies like we are doing now, these imbalances can go ahead and completely upend the study. So that is why patient selection is extraordinarily important. What we did in SOL-1, for instance, was say we needed patients who were going to lose vision, because the delta was based on maintenance of vision. So, we specifically selected patients who were anti-VEGF dependent. We gave two injections of EYLEA. We only selected those patients that either became 20/20 or improved by 10 letters or more, and then we stopped injecting them with the anti-VEGF. We proved they were anti-VEGF dependent, we took away the anti-VEGF, and we allowed them to lose vision once to create that delta, and we succeeded. And that is the opposite of what we are doing in our second study called SOL-R. Here, we want absolute stability.
What we're doing there is giving three loading injections of EYLEA, and then we're waiting to observe the patient for two months, and that's never been done before. We're simply observing the patients for any fluctuations in the OCT, and if they're unstable, we're weeding them out, and then after that, two more injections, and only then are we randomizing them. We're trying to do our best to de-risk this patient population by ensuring absolute stability. That much thought has been put into proper patient selection in a bespoke manner for each study, and I don't know of any other clinical trial that's doing that.
Moving on to the regulatory side, you've obviously had a successful SOL-1 trial, and you're hoping to submit your NDA soon, or perhaps that's already in process. Can you provide an update on that?
Yeah. What the FDA has said, in a very consistent manner, is that there is an option for single trial submission that has been there for decades, and their criteria for that, the evidentiary standards, as they say, for that, has not changed at all. Has not changed at all. There are two evidentiary standards prior to the card turn, and there are two evidentiary standards after the card turn. Prior to the card turn, first of all, the study has to be well-masked, and that's really important. This is where pretty much everybody other than us doesn't qualify, because well-masked means that there can't be a sham arm. The FDA does not consider sham to be proper masking. Sham is when you touch the eye with the hub of a syringe, and patients know when the drug is coming in or not.
Because it's a subjective endpoint, which is visual acuity, the FDA feels that that's not proper masking. That's one evidentiary standard. The second is that it has to be properly powered. We've met both of those because we have a SPA, or a Special Protocol Assessment, with the FDA based on the SOL-1 study. After the card turn, there are two other evidentiary standards. The first one is that it has to have a p-value of less than 0.001. Our study has a p-value of less than 0.0006. The second is that you have to be able to supply at least 300 patients who've been exposed to the drug for safety purposes at the time of submission.
Immediately following our card turn, knowing that we had a p-value that checked the box, we had a Type C meeting, immediate meeting in May. The card turn was the middle of February. We have a clear path forward for single trial submission officially. We said that we're going to have our pre-NDA meeting in the third quarter of this year. We did. It was a very successful pre-NDA meeting. We said that we would be submitting the NDA in the fourth quarter of this year, and we're on track to do that.
Great. If you could help us understand the timelines then. Once you submit the NDA, which sounds like it's on track, what are the next updates that investors will receive from you about how that's going? If it's approved, when could AXPAXLI launch, by your best guess, in the U.S.?
Yeah. We haven't given timelines for those things. I'm sure that we'll be giving you updates as appropriate. What I can tell you is typically, not just for us, but typically, when an NDA submission is done, the FDA gives you about 60 days, and then gives you a response as to whether they'll accept that or not. We expect that to be the same for us as well. As far as the track of the filing is concerned, realize that we're in accelerated track, right?
We officially have a 505(b)(2), meaning that both components of the drugs have been previously FDA approved independently. So, that saves us at least two months on the front end. We expect this filing to be very efficient also on the back end because we have a SPA agreement, right? So, we expect the filing to be very efficient. We just haven't given you timelines as yet, but we are an accelerated pathway.
Okay. Now, we've seen some changes in the wider FDA leadership, including a new ophthalmology division head. Does this present any type of risk to your regulatory process, in your view?
We don't think so at all. Look, members of this team have worked with the FDA for almost 30 years, as I have, on the other side as a physician. We know the FDA extremely well. We know everybody there. The lineage goes from Dr. Wiley Chambers, who's retired, to Dr. Bill Boyd, who's announced his retirement, I think effective at the end of this calendar year, to Dr. Lloyd, who is going to be the new appointed chief of our division. She has been in every one of our meetings. We know her very well. There is really no inconsistency in that lineage whatsoever. To really guarantee that, we have a SPA. We have no issues whatsoever with the consistency that we have in being aligned with the FDA.
Okay. For ex-U.S., what could the timelines look like there? Do you have to wait for SOL-R to read out before you can think about filing for approval ex-U.S.?
Well, what we've said is that, look, we believe that SOL-R is not going to be necessary for FDA approval in the U.S., and we are more confident than ever in the U.S. that we'll be approved with SOL-1 alone. However, we have said that SOL-R will likely be important in some regulatory agencies outside the U.S., and that's why it's absolutely essential for us to protect the integrity of SOL-R, as we have done, as you've seen with some of the changes that we've made, which is to really protect the integrity of SOL-R for its purpose outside the U.S.
Okay. Maybe we can talk a little bit about those changes then to SOL-R and SOL-X that you recently announced. For SOL-R, the change was you moved the primary endpoint up to week 40 from week 56, and the SOL-X update was you extended the treatment intervals for this open-label study, I think from 24 weeks to 48 weeks. Could you talk a little bit about those two updates and the rationale for those?
Absolutely. I think the other thing to add, because they're really part of the triad, is to also say that we extended the card turn for SOL-R to week 96. Chronologically, let me go over how that happened. As soon as we had that Type C meeting in early May, and we were confident and are confident that we will be approved based on a single trial alone, our attention was then focused on SOL-R. The first thing that we wanted to do was to make sure that with the card turn that was planned in SOL-R, we would not get a major amendment. Remember, originally, the card turn in SOL-R was planned for the first quarter of next year, which would be in the middle of the review cycle, and that would risk getting a major amendment, in which case we would lose all of our advantage altogether.
For that reason, we extended that card turn to week 96, which also allowed us to study superiority versus high-dose EYLEA, which would be very, very useful, as you can imagine, from a commercial leverage point of view. Now that we've removed the risk of a major amendment from the table, the next question was, what about the integrity of SOL-R? Because remember, we're going to have to borrow approximately 130 patients' safety data from SOL-R to complete our package for the full submission for Sorry, I'm just moving around a little bit. I'll just hold this. For SOL-1.
In order to preserve the integrity of SOL-R, what we did with multiple discussions with the regulatory agencies is to say, "The way we can make sure that we are completely protected against any operational bias is to go ahead and choose a primary endpoint where everybody would have crossed the primary endpoint already." So there'd be no risk of any kind of operational bias whatsoever. If you remember, the U.S. FDA recognizes anything past week 56 as a complete study for a year, as long as you have treatment-naive patients. We do have treatment-naive patients, so we certainly qualify for anything after week 36.
At week 40, everybody would have crossed the mark anyway, so there would be absolutely no chance of us introducing an operational bias as we look at the safety data for the 130 patients, and that is why we chose week 40 as the primary endpoint to protect the integrity of SOL-R. We did that despite the fact that we had said earlier on that week 56 would be ideal for us based on the peaks and troughs of the drugs. We did that because we saw the results of SOL-1. We were so confident in the results of SOL-1 that we did not think that it was necessary any longer to look at the peaks and troughs of these drugs at all. We did not mind when the primary endpoint was, because clearly this was a drug that easily lasted in the most difficult patients for a year.
There was no reason that it would not last for six months. We did that from a position of absolute confidence in what we saw regarding this drug in SOL-1. It was the same confidence that allowed us then to say, "Look, why are we duplicating the data for every six months in SOL-X?" Remember, we will have six-month data in SOL-R already. We are absolutely confident that this drug lasts for a year, so why not show that this drug lasts for a year? That was the opportunity that SOL-X presented. For that reason, we said, "Look, there is no reason to do an every six-month arm here, because we already have every six months in SOL-R. So why not just show that this drug lasts for a year, as we know it does?" That is why we extended SOL-X to every 48 weeks.
On the timing of this decision, why did you decide, or why did you disclose it now rather than previously?
Yeah, it is a great question, Alice. Look, we said that we were in these discussions once we had the Type C meeting in May. The question is, well, why did not we disclose it in May? Remember, when you do a major protocol amendment like this, it is not just a matter of disclosing when you think of it, right? You have to go through the regulatory agencies, make sure that they are aligned, and then, and this is really a time-limiting step, you have to go through all the IRBs.
It is not a centralized IRB. We are in multiple countries. In the U.S., there are multiple IRBs. We are also in Argentina, we are also in India and Australia. All these IRBs meet only once in a while, so it was important to make sure that they were all aligned, then go back to the regulatory agencies to submit that and show that they're aligned, then make the announcement, and that's why it took this long.
Okay. I think investors are also curious to know what you will know internally at the 40-week primary endpoint, if you're submitting some data to the FDA from that trial.
Yeah, it's a fair question. The answer is we'll know nothing. Remember, we're completely masked. We'll remain masked. That data will be not in our hands at all. It will be in the hands of a third party who will then submit that to the FDA. So, we will not know anything at all. It'll be part of the full package, but we'll have no access to that data whatsoever, and the trial will remain completely masked until week 96.
Okay. What implications are there, do you think, from moving the primary endpoint from week 56 to week 40 in terms of the outcome of the primary endpoint? It sounded like you're very confident, but how are you thinking about that?
Yeah. So, Alice, I can answer that based on what we have seen in SOL-1. We really obsessed about week 56 prior to the results in SOL-1 because the peaks and troughs seemed to be perfect for us. Once we saw the results in SOL-1, we really did not think that was important anymore. Remember, in SOL-1, as I described earlier on, we had the most difficult patients. These were patients who were specifically chosen to lose vision, and despite that, with a single AXPAXLI injection, more than two-thirds of patients were rescue-free at year one. Given that patient population and given those results with superb disease control, as we have shown over and over again based on the OCT, we have absolute confidence that in the most stable patient population. Remember, in SOL-R, it is really the opposite selection than SOL-1.
We have chosen patients who are going to be absolutely stable based on a six-month ramp. So, where in thereafter in the six months we choose the primary endpoint really does not matter at all. We are absolutely confident that this drug will succeed based on the SOL-1 results. So, our confidence has not changed. We do not think that anything is going to change with the change in the primary endpoint, other than the fact that now we are absolutely certain that the integrity of SOL-R will be preserved.
Okay. You mentioned SOL-R could be a more important trial commercially. Can you just explain why you think that?
Well, so SOL-R is going to be important from a regulatory point of view. I just want to reemphasize this, outside the U.S., and that is why it was so important to preserve the integrity of SOL-R. In the U.S., SOL-R, we believe, is going to be very important from a commercial point of view because we do have an opportunity there to go up against high-dose EYLEA. We believe that we will be able to show superiority to high-dose EYLEA based on several parameters. The parameters that we are examining is not just visual acuity, but also treatment burden, OCT, fibrosis, atrophy, et cetera. Remember that we have not talked about this much, but remember that the purpose of this drug is twofold.
One is sustainability, as the only maintenance drug that we believe will be approved, but also as a drug that has better long-term outcomes. Better long-term outcomes because we have constant suppression of VEGF as opposed to pulsatile episodic suppression, and that's what leads to atrophy and fibrosis. We believe that based on SOL-R and also based on SOL-X, which is our extension study, we'll be able to show that there is a decrease in atrophy and fibrosis, which will lead to a better long-term visual acuity result as well.
How do you see the initial addressable population then with AXPAXLI? Because it would be approved based on SOL-R. Oh, sorry, SOL-1, and then later on, we might have SOL-R, which could be more commercially attractive. So, how are you thinking about this initial addressable population versus how that could expand later on?
Yeah. It's a great question. Look, we've given the roadmap of how to use this drug and where this drug would work. What we have shown in the academy recently and what we have shown more recently in other meetings is that if you use two bolus injections, and probably the bolus injection doesn't matter, and the patient is stable after those two bolus injections, you can go ahead and use AXPAXLI, one injection, and confidently predict that the patient will be rescue-free and in an excellent disease control for at least nine months. Now, what do I mean by excellent disease control?
Excellent disease control means that they will be within 10 nanoliters of volume of fluctuation in the OCT, and that's unheard of. That is the strictest, most stringent definition of dryness that anybody's ever seen, right? This drug has superb disease control, and we believe that that's the roadmap of how to use this drug. It's not a bolus drug. We have proven that. It is a maintenance drug, and as of now, it has no competition in that arena.
Okay. As you think about retina practices, and they obviously see a lot of patients, give them anti-VEGF, so they do a lot of injections with anti-VEGF. How can you make this type of therapy economically attractive for these practices and kind of talk about the value proposition to the practices?
Yeah. That's a great question. This is really one of those few drugs where there's a win-win-win proposition. We've done a lot of work, we're doing a lot of work with payers, obviously. As physicians, we understand how to run a practice as well. Let me just describe that to you. In the U.S. alone, despite having a very effective therapy with anti-VEGFs for almost four decades, there's a 40% dropout rate in the first year. That's astonishing, not just from a humanity point of view, but from also economic point of view. Almost half the patients drop out in the first year, and every one of these patients ends up going blind. The most expensive disease to treat by far is blindness. It's not strokes, it's not heart attacks, it's blindness, because your mortality doesn't change.
Your use of resources skyrockets, but your mortality doesn't change. We know that in the U.S., every blind patient costs a payer at least $66,000 a year. If we're able to reduce that dropout rate by even 10%, that's a quarter million fewer patients in the U.S. alone that are not going to go blind. The cost savings of that is enormous. So from a payer point of view, this is very, very attractive, in terms of not just humanity, but also in terms of cost savings, which we believe can be translated into a higher pricing point, which we believe can be translated into a very, very attractive economics for the doctor, because there is no overhead expense at all. There's not a piece of equipment to buy. That doesn't involve changing the workflow in any way whatsoever.
As far as the number of injections goes, the doctor will have as many injections as he or she wants to do, just not on the same patient over and over again, because the dropout rate will be lower. Everything we're talking about is entirely targeted towards wet macular degeneration. It doesn't really even include a topic that we haven't even talked about, which is diabetic retinopathy and diabetic macular edema, which is three and a half times as big as wet macular degeneration.
Definitely. That is actually my next topic. You are running another trial with Ocular Therapeutix, as you said, in non-proliferative diabetic retinopathy called HELIOS-3.
Yes.
During your recent earnings call, I think you noted that enrollment is progressing well. If you could provide an update on that, could you talk about how important patient selection is in this indication as in wet AMD with your drug?
Yeah. Thank you for asking about that. You are right. We are getting absolutely no credit whatsoever for diabetic retinopathy. This will be a major topic, following our submission for wet macular degeneration to the FDA. Recruitment is going extremely well. We are being very, very careful about patient selection for the same reason. As you can imagine, in diabetic retinopathy, there is a window. If patients are enrolled who do not have enough disease, one may not be able to show a delta. If patients are enrolled with too severe a disease, one may not be able to change the course of the disease. Although there is no competition, we could easily finish enrollment much, much quicker.
We are being very, very, very careful about patient selection in this disease, as we have been with wet macular degeneration. It is important to say that we are not just enrolling for diabetic retinopathy. We are enrolling for diabetic retinopathy and for non-central involving diabetic macular edema. Our expectation is that we will have a label that will encompass all of diabetic retinal disease, so diabetic retinopathy and diabetic macular edema. Our expectation is that we will not have to do another study again for diabetic retinal disease at all, and that label will be broad enough to encompass all of diabetic retinal diseases.
When should we expect the next update on how this trial is going? Also, if you could say whether this readout is expected ahead of the SOL-R readout.
Yeah, that's a great question. So, what I've said again is that the enrollment is going very, very well. I don't mean just in terms of volume, but I mean in terms of patient selection as well. It's always important to ask about that as well as the number of patients. As far as the emphasis on the disease is concerned, that is our major topic in 2027. At that point, we will give you guidance as appropriate.
Okay, great. If you could remind us of the design of that trial and what you need to show with this trial.
Yeah. So, we have a study called HELIOS-3. That's what we're emphasizing. What we have done is to remove the every six-month arm now that we've seen the SOL-1 results. Again, all of these decisions are being made from a position of confidence in the drug from what we've seen in the data for SOL-1. So, we feel that it's only necessary to go ahead and give this drug once every year. The important thing to emphasize here also is the novel endpoint. We have an ordinal endpoint, which means that a patient will be counted as three chances on goal, as we say. The patient will be counted whether the drug improves the disease by two DRSS scores, decreases the opportunity to go down by two DRSS scores, or maintains the patient.
It is an ordinal endpoint validated by a SPA, which we believe will increase the chance of success significantly. As far as diabetic macular edema goes, our previous study showed great results for diabetic macular edema. If this study is anywhere close to what we showed in the original HELIOS study, we believe we will have great success for treatment once a year to prevent a 30%-40% chance year upon year of a blinding complication without any competition whatsoever.
Great. I think that is all we have time for. Thank you so much, Pravin, for joining us, and thank you everyone for being here.
Thank you, Alice, for having me.