Kodiak Sciences Inc. (KOD) Earnings Call Transcript & Summary

Thanks, everyone, and welcome to the 40th Annual JPMorgan Healthcare Conference. My name is Anupam Rama, I'm one of the senior biotech analysts here at JPMorgan. I'm joined by Priyanka Grover, Malcolm Kuno and Caleb Smith from the team. Our next presenting company is Kodiak and presenting on behalf of the company, we have CEO, Victor Perlroth. I just want to remind all the attendees that there is an ask-a-question feature in the portal. If you would like to put a question in there, I would be happy to ask on your behalf. With that, Victor, go ahead. Take it away.

Thanks, Anupam. Hello, everyone. I'm Victor Perlroth, the CEO of Kodiak. I will be running through our presentation, which is posted on the Kodiak website. So first, we'll just do a quick reminder of the forward-looking statements. So please read those carefully and we'll be describing potential future events that we anticipate. So first of all, it's my pleasure to be here today at the virtual JPMorgan. Certainly, being virtual, it pulls into focus the continuing pandemic. Therefore, I want to be highlighting today -- what I want to be highlighting is Kodiak's continued strong progress on execution, in particular, on the clinical development front. And the road map of pivotal study, top line data readouts from our 6 ongoing registrational studies. So before we jump into those granular details, maybe a quick slide, Slide 3. To orient newcomers a little bit to the Kodiak story. So Kodiak was founded in 2009, and in the concept was to develop new science, new technology medicines. We call those our ABC medicines, our antibody biopolymer conjugate medicines. We're focused only in ophthalmology and retinal and we're focused on high-prevalence diseases probably targeting many millions of patients globally. KSI-301 is our lead molecule. It's for retinal vascular diseases. That's a $13 billion-plus market. That's growing and where there continues to be a strong unmet needs. And so KSI-301, we are exploring it in studies that look at the leading cause of blindness in older -- elderly people in the developed world as a treatment and also evaluating it for the leading cause of blindness in the developed world for working-age people for diabetic patients, diabetic retinopathy. So it has as well as several other indications. So it has a tremendous potential and we're exploring it in a deep study. We also have our pipeline behind KSI-301, KSI-501, a bispecific ABC medicine and then deeper in the pipeline, we have very interesting scientific medicines, our new triplets. And we're very excited about some of the emerging signs in our pipeline. But let's go back to where Kodiak is today. Looking at Slide 4. The KSI-301 clinical development program. Where we are today? Well, we dosed approximately 2,500 patients, across 6 pivotal studies, exploring all 4 of the major anti-VEGF indications, wet macular degeneration, diabetic macular edema or DME, retinal vein occlusion or RVO, and also nonproliferative diabetic retinopathy more of a preventive indication. So our goal is to combine the first 5 of these pivotal studies together into a package to form our initial BLA submission. And then our 6th study, the prevention study. But we hope to have that data before pricing. And that's our nonproliferative DR, the GLOW study. So of the initial 5, looking at the Slide 4. What we have outlined here is a little bit of a road map for all of you so that you can track kind of where we are today and what the expectations are for top line data readouts across these 5 studies. And at the same time, we highlight somewhat the enrollment progress of where we are within each study. So the DAZZLE study, obviously, is coming up. We're poised for top line data readout. The time line that we guide to is the second half or the latter half of the first quarter of 2022. So we're coming right up on that. So that's approximately 550 patients in treatment-naive wet AMD. It's fully enrolled. And I think a critical element is noninferior vision as the primary endpoint versus EYLEA with a dramatically lower number of injections in the patients. DAYLIGHT is our monthly wet AMD study. We're very pleased to see north of 50% enrollment in DAYLIGHT, very strong engagement by the community. It's really leaping forward. And actually, we expect DAYLIGHT to be able to read out actually before the GLEAM and GLIMMER pivotals for diabetic macular edema, which are both north of 90% enrolled at this point. So we're getting very close to last patient last visit for GLEAM and GLIMMER with the 52-week endpoint. And BEACON, our RVO study actually is going to be the second study that will readout as we had through a press release. That's a 100% enrolled now as of last year with the 24-week endpoint that brings us really into the June time frame for last patient last visit. I think we're hoping across the 6 pivotal studies with each study to have a shorter time between last patient last visit and top line data readout. So if we have, let's say, a little less than 12 weeks for DAZZLE, we hope to pull that a little bit shorter for BEACON. And then a little bit shorter again for DAYLIGHT. And then hopefully, GLEAM and GLIMMER will be poised for rapid top line data readout after last patient last visit in the context of a little bit more than a year from today. So that gives a nice view into the cadence, okay, of data. And I think also in the context of the pandemic as well, does showcase the strong work that Kodiak's clinical team and our supply chain team is doing to keep those pivotal studies enrolling, and once enrolled to keep those patients coming and to getting the data that we need to be able to put the full package together for KSI-301 for all of you who are following the story. So on Slide 5. Well, what do these pivotal studies actually tell us and where will it take us with KSI-301. I think it's somewhere very exciting, is certainly the potential based on the design of the studies and the breadth of the program. So we're looking at the broadest label for an anti-VEGF. So we're looking across a wide range of dosing intervals, okay, within each indication to maximize the flexibility and the reimbursement confidence for physicians and patients. So we're looking at a label that can go once a month, all the way to every 5 months in wet AMD coming out of the combination of DAZZLE and DAYLIGHT studies. And we're also looking in diabetic macular edema every 1 month all the way through every 6 months based on the design of DAYLIGHT and into the GLEAM and GLIMMER parent studies in DME. And in retinal vein occlusion, which is labeled for monthly dosing by EYLEA, we're looking at every 8 weeks. Those 5 studies to be packaged into the initial BLA. So 3 major indications for the broadest labels. And then, of course, our GLOW study, which is the prevention study, which is a little over 20% enrolled at this point, which is very good for that kind of indication during the pandemic, I suppose, looking at every 6-month dose. So again, a very powerful label network that we're working towards. Looking at the next slide, how should you all and ourselves, think a little bit about the profile for KSI-301 in the context of the commercial opportunity or the mission or the opportunity for patients themselves or for the clinics. What profile can meaningfully change the current paradigm for patients with these diseases. Well, our view is that we designed and are executing well on a pivotal program that is designed to explore 5- and 6-month predominant profiles, okay, across these major indications. And that's what we see as clear differentiation versus all of the other medicines, okay, that have basically read out their top line data whether it's predicate medicines like Lucentis or EYLEA or once a month or every other month regimen or whether it's faricimab as an up-and-coming medicine that brought, say, slightly less than a majority of their patients, let's say it, every 12 and 16 weeks, every 3 to 4 months. Our objective in the program is really to try to bring a majority of patients to the 5- and 6-month level predominant, the 5- to 6-month predominant. Of course, that's my own aspirational goal for the molecule, really having been involved since the company was founded back in 2009, and really working over that period of time to design the medicine with the teams and to develop that medicine and to be exploring and [ to use in these ] pivotals. Having said that, when we speak to the community, really, I think a middle of the road here on this chart anyway, a 4- to 5-month predominant profile that we bring a majority of the patients to 4 and 5 months in DAZZLE, that is a phenomenal outcome and a clear differentiation from the next closest medicine, which is really faricimab and EYLEA, which really are 2 kids in the camp running a little bit together. Okay. So hopefully, that helps us to think a little bit about the type of data that we can be anticipating as we turn the data card for DAZZLE in the latter half of this quarter, Q1 in 2022. Thinking about the durability thinking about noninferiority [indiscernible] on the efficacy, right, the vision, the BCVA, and then the safety, of course. And also, we would expect the [ time or rate ] sort of data to be thinking about really would be the noninferiority of the vision. That's the primary endpoint, the durability proportion that we just talked about in terms of differentiation, the safety as well as a direct comparison on the OCT in terms of the imaging at the 1 year interval. And so that's a little bit what we think about when we think about top line data for DAZZLE in the latter half of this quarter, meanwhile preserving some important data for physicians to be able to talk about on the podium to the retina community, hopefully, following shortly after top line data release. So let's just step a little bit deeper from there, kind of where Kodiak is today. And I think fairly quickly, we could give you a little bit of a view a little bit more maybe of a rapid fire view through what KSI-301 is for those who are a little bit new to the story. So looking at Slide 7. Just a quick introduction to the platform, right? Our antibodies that are novel, are phosphorylcholine, biopolymers that really deal with water in a very novel and important way to build these conjugates that are large and have a number of very novel, very powerful properties, okay? And you call them same where it matters, different where it counts. In other words, their intravitreal biologics, injected in a similar manner, similar paradigm. Yet the science underlying their design is tremendously different and has the potential to drive very, very different and important opportunities. Moving on to Slide 8. Today, traditionally, different companies compete really on the trajectory of molecular weight in terms of size, getting larger as well as formulation strength getting more drug inside of a given volume injected into the patients. In KSI-301, despite all the new science that I just kind of mentioned competes very favorably. We're the largest agent, and we have the largest very high molar dose, high formulation strength. So even on the more traditional metrics, it's a very well-designed medicine on its own base of molar dose alone should deliver well. On top of that, of course, we have our new signs of the ABC platform. On Slide 9. Briefly, just on the top right, you see an electron microscope image of KSI-301, which I think very cool. It really showcases a true visual reality of what our ABC medicines, what our KSI-301 conjugate looks like versus the little [ bit to it ]. That's what we're injecting, okay? And when you look at some of the data, whether it's the class-leading half-life, the residence time, how long it stays in the eye, its ability through that bioavailability to dramatically access the retina in an amazing manner, okay. What we call a deeper potency is actual power as a medicine is actually very strong, despite its large size. And although it acts big in the eye when it gets into the body, it's actually cleared very quickly. Again, an element of its [ scientists ] design. Moving on to Slide 10. Stepping into, well, why do we have some confidence that some conviction coming into DAZZLE, and its top line readout? And also, why are we running all these pivotal studies in parallel, right? And why are we investing in manufacturing that should be able to provide more than 10 million doses per year or more, actually. Why do we have that conviction? -- part of it is around the design and the science of the medicine. But part of it, of course, is the clinical data that we generated in our Phase I/II or 1a/1b studies, the 1-year data. And with that program in treatment-naive patients that are overall similar patient populations to what we're testing in our pivotal studies. We tested the wet AMD indication, we tested the DME indication, we tested the RVO indication in treatment-naive patients. And generated really stunning data, safety and efficacy in line with today's first-line agents. And really a durability that was disruptive and clearly differentiated if we can repeat that in the context of what I said are our 5- and 6-month predominant profiles, right. That we're targeting in the design of our pivotal studies. So in the Phase Ib, we took 2/3 of the patients to 6-month intervals well beyond what really is needed in the context of an anti-VEGF biologic, to be able to have the leading molecule a new first-line agent, what I call sort of a new North Star for the field. So we've shown really, really tremendous data in terms of durability. And when you look at a study -- a slide like this and you look at the interval at year 1 and you look at 1 month, 2 months, 3 months, 4 months, 5 and 6, and you look at the percent of the patients right, that was on each 1 of dose intervals. It helps you understand that in each 1 of these diseases, it's really like a distribution. Some patients need monthly dosing, in fact, with EYLEA, a number of patients need twice-a-month dose. But some need once a month, some every other month, every 3 months, all the way out. And we want to have a medicine, and that's why we're running, say, our daylight monthly study. because we want to be able to provide a medicine that can be pulled out of the fridge and put in any patient that requires an anti-VEGF agent, be they once a month or maybe even twice a month or whether need to go to 5 months or 6 months or maybe even beyond, the physician can grab it and then they can get reimbursed. That's our strategy, and that's why we're running this broad program. And that's why we're really investing with conviction with this new data card DAZZLE being turned within the next few weeks, let's say, the latter half of the first quarter. Moving to Slide 11. Another view of the Phase I/II data, okay, in terms of number of doses, okay, in year 1, the actual improvements in the vision, the primary endpoint for the pivotal studies. In this case, in the pivotals as compared to EYLEA and then also looking at the changes in the retinal anatomy to really showcase that the drug is able to have an immediate impact. Yet the primary endpoint, of course, being changed in the vision. So moving into the Slide 12. I think it's a reminder. Our focus at Kodiak is on the pivotals and on the execution and turning those data cards to show you what our data looks like across the major indications through the first 5 pivotals, which are going to go into the BLA and the GLOW study. And it's also investing in the science for retina, for retinal medicines, for ophthalmology, for our community of patients and clinicians. And KSI-301 really is just the beginning. It's here on the top right? It's an antibody is conjugated to our polymer to form that conjugate that inhibits a single target. But KSI-501 is our bispecific that inhibits 2 different and powerful targets of relevance for retina. Interleukin-6 and VEGF together, and there's going to be an important place for that molecule moving in our pipeline for patients. And as I mentioned, we have tremendous and very exciting and interesting progress on the science and the development of what we call them triplets that bring small molecules and biologics together in the context of the water, the special water structure that our phosphorylcholine conjugate allows us to bring many hundreds of copies of powerful new bioactives. So we'll save those pipeline elements for the future. Right now, our focus is on high-quality execution, both from a clinical development standpoint, a manufacturing standpoint, reading those data cards out. This quarter for DAZZLE, BEACON midyear, then DAYLIGHT, and then GLEAM and GLIMMER, and putting those together into the BLA. That's our focus. And thanks for listening. Anupam, happy to take some questions.

Yes. Victor, if you want to introduce the broader team on the line, we can get started.

Absolutely. We have Jason Ehrlich, our Chief of Medical and Development Officer on the line. Hi Jason...

Hi, everyone.

And we also have John Borgeson, our CFO, who's in the room with me here for any financial questions.

Yes. We have -- just a reminder to the audience that we -- if you want to submit a question in the portal, I'm happy to ask on your behalf. And we actually do have a bunch of questions in the portal. So maybe I'll start there, which is -- the first 1 is can you speak to the rescue criteria differences between Phase Ib and DAZZLE?

Well, the intent of the study designs are to keep patients in the studies. So just like EYLEA in its pivotal programs where patients were taken to 8 weeks and then not rescued in between those 8-week doses, it's the same in our pivotals and especially in DAZZLE, where patient's hot from every 12 weeks, every 12 weeks or every 16 weeks or every 20 weeks. So there's no rescue injections given as part of the program. And what happens is that physicians perform their assessment, they take those values into the computer and then basically the framework of the clinical trial will tell the physician to grab [ kit 12 or kit 17 ]. And if they're an EYLEA patient at the 4-week interval they're going to get a Sham kit. If they're a KSI patient on the 4-week, week interval, they're going to get a Sham, otherwise, they would get whatever the appropriate, EYLEA or KSI. And so it's run based on the assessments of the physicians and then it goes into the framework of the study, and then they engage in the appropriate injection for the patient.

We have another e-mail question here, which is do you get blinded safety looks at DAZZLE? Have there been any additional cases of intraocular inflammation in the Phase Ib? And what do you think is an acceptable rate of intraocular inflammation? So 3 questions there.

I think at the highest level, the question is asking, well, what type of visibility do we have out of, say, the 2,500 patients that are in the pivotal program at this point, who are receiving injections through the course of the program, whether it's DAYLIGHT once a month or whether it's GLEAM and GLIMMER every 6 months? We do have deep visibility into safety across the study. Of course, we don't know what drug the patient is on, right, whether they're on EYLEA or whether they're on KSI. However, we have deep visibility into safety. We're continuing to see a very consistent safety profile, consistent with what we have indicated before whether it was in the queue that we had in the last few months or whether it was broader concepts around inflammation or other untoward safety events. We continue to be very pleased with the safety profile that we have. We continue to see that as a safety profile in line with existing predicate standard of care agents. And certainly, safety will be an important element of the 1-year DAZZLE data, as it will be for BEACON, as well as DAYLIGHT, and GLEAM and GLIMMER. So I think at a high level, the right answer is to say that we continue to think our molecule is acting like 1 of the standard of care molecules and not anything that would be more similar to [ Beovu ] or like a bigger part of the agents that clearly had a differentiated and negative safety profile. That's not what we're seeing in the context of the data that we see, and we have had deep visibility across those 2,500 patients.

And I guess, just to put it out there, what is an acceptable rate of IOI, if you could just put that into context, which was the third part of the question?

Well, I think it would be low single-digit percent. I think we have an active comparator in the study. So we'll be able to do that comparison. To a certain degree, it might depend a little bit on the views of the physicians in the study and how they decide to grade inflammation or other events because the preconceived notion can drive say EYLEA's inflammation rate from 1% to 2%, or from 2% to 4.5%. But we consider EYLEA to be kind of the standard that we want to compare ourselves to. And we hope that will be at or below EYLEA's level of adverse events in the study. So we'll have to see when we read it out.

Another e-mail question, which is, can you ask given you have 3 active arms, are you splitting alpha across the 3 arms? Or is it a hierarchical analysis?

Jason, do you want to give a little bit of detail kind of on how we actually design DAZZLE? And how people have been doing it in the industry?

Yes, sure. That's -- thanks, Victor. It's a good question. So the important thing to understand is that for the sake of the statistical comparison, it's 1 active arm of KSI-301 being compared for noninferiority against 1 active arm of EYLEA. So the study is 1:1 randomized. Within the KSI-301 group, patients can receive the 3-, 4- or 5-month dosing based on those criteria, as Victor described. But when we do the statistics, it's that, all those patients considered as 1 population compared to the EYLEA patients. And that's the same approach that was taken, for instance, with -- within the faricimab studies or within the brolucizumab studies in the past. So the number of doses that each patient receives is customized, but there's all 1 group. So there's no splitting of alpha. It's not 3 different comparisons against EYLEA, it's 1 comparison.

So thinking about retreatment here, it's a 12-month study. So after the lowering doses, a patient who get KSI-301 at, say, 3 months, which is -- or 6 months, which would be a 3-month interval, but then at 11 months, which is a 5-month interval, how is that patient accounted for, I guess?

I mean the simple way that I think about it [ out of the box ] is all of the Kodiak patients that are receiving KSI-301 at the primary endpoint looking at vision. It doesn't matter what group they're in. They're all pulled together into 1 group. And that group's mean change in BCVA is compared against the mean change in BCVA for the EYLEA group. So the durability doesn't matter. We're giving fewer doses, but we're being evaluated similarly. So some people say, if you show noninferior vision and so many fewer doses, I mean, in some way, that's like a superior. Let's take things 1 step at a time.

Okay. Okay. Another e-mail question we have here is based on your time line, when would you expect regulatory approval and launch?

Well, right now, I think what's important for us is really to look at -- if you go to the deck, which is on the web, you go to the Slide 4. And I think what's important for us is progress on execution, right, with 90-plus percent enrollment. I think what do we say, GLIMMER at 95% and GLEAM catching up quickly. We expect it's a 52-week endpoint for GLEAM and GLIMMER. So that's the tail that's going to wag the dog there. So I think with each study, we hope to have a shorter interval from the last patient last visit into top line data. When GLEAM and GLIMMER readout will have had a lot of preparations and we'll be ready and we'll be organized. And so had some weeks from today to finalize GLEAM and GLIMMER at 52 weeks, add some time to go from last visit to top line data and some time to package that into an integrated safety database, well then into our BLA that helps to give you a visibility into timing. But from our standpoint, we're going to be focusing on execution of these studies of the top line data readout, and obviously, the CMC section required for the BLA, we'll be not providing any guidance on BLA or commercialization today. But we're going to focus on what's right in front of us, and we obviously have those time lines in mind in terms of what we believe to be aspirational versus what we believe to be achievable. We hope to achieve our aspirations.

Another e-mail question here, which is -- so if you start at every 16-week interval, do you stay on an every 16-week interval through the entire study?

Jason, do you want to talk a little bit about some of the nuances there.

Sure. Yes. So the way that DAZZLE works for the patients on KSI-301, they get their 3 initiating doses, the 3 monthly loading doses. And then the default is that the patient is on 20-week dosing. But if they don't make it that far, that is they may meet any of those criteria for the disease activity at either 12 or 16 weeks after the last loading dose, then they can drop down to 12- or 16-week dosing. And then that process repeats itself over time throughout the first year, right? So a patient makes it to 20-week dosing. They get dosed 20 weeks after the last loading dose. They can go another 20 weeks, but that assessment process repeats. So if they make it to 20 weeks, they stay there, if they need to get 16 or 12 weeks, they get that. So there's multiple points along the way at which they can have their treatment interval customized based on their disease activity. If they drop down an interval in the first year, they can't go back up in the second year of the study, there's also -- there's flexibility in both directions. So in the first year, we give people option to get dosed more frequently. In the second year, they can get both more or less frequent based on how they're doing.

You haven't disclosed what portion of -- and this is another e-mail question. You haven't disclosed what portion of patients are in each of the 3 dosing schedules though. So what that would mean?

Not yet. Well, we did for the 1 week at the endpoint, and we'll be looking at that in the data, and we'll be planning to release that.

Yes. We have another e-mail question. It's actually 2 questions. I'll break it up. The first question is given the advancements of EYLEA and Roche reducing treatment burden from every 2 months to about 4 months, and that being approved in the next year or so. How do you plan on differentiating the VEGF with this evolving marketplace?

Well, we actually tried in that Slide 6 to articulate a little bit of a framework to answer this, a listener's question, in terms of the number of loading doses, faricimab, which can be more, right? And then looking at really, they have less than a majority of the patients that were in even the 16-week group. And really how they define that durability wasn't the fluid manner that Jason described, which we think is a better reflection. Rather what they did is coming out of their extended loading phase while they're still sort of exiting the loading phase they locked patients into a particular regimen and then call that the regimen at year 1. So I think there's a lot of nuances overall. In our framework, like I said, my personal aspiration would be if we can show a majority of patients in the 5- to 6-month predominant durability regimens. Fundamentally, even a majority in the 4 to 5 months, which is not what EYLEA shows, right, and that's certainly not, and not what faricimab showed. That's quite differentiated, and they had a number of patients even on an 8-week interval. I think fundamentally, faricimab should be thought of more as a very close to EYLEA type of medicine, okay? There is no science around the design of faricimab as it relates to durability, certainly. Now there was science around its design with regards to a second mechanism of action, the Ang2 inhibition. It's just not clear from any of the data that we've seen really that the Ang2 inhibition is really driving any kind of important therapeutic benefit for the patients, and it doesn't appear to be driving any kind of durability benefit. So that's why I say when we're in an attractive situation in that prior agents and faricimab have turned those Phase III data cards. And I think they leave still a tremendous North Star opportunity for KSI to be the single agent as a biologic that really showcases the new potential for patients. And with regard to non-biologics, I would say those data cards having turned as well with gene therapy or the port, and I would say, overall disappointing for patients. So again, a very nice opportunity for our medicine. Of course, we want to turn our own data cards across the 6 pivotals, share those and then be able to put that data in context for everyone.

Same participants. Second question was, do you have market research supporting physician interest for -- in diabetic retinopathy, which is largely left untreated by EYLEA approval and marketing?

Yes. I mean, Jason, you talk to that a little bit more. He's really, I would say, a world leader, having led the approval of Lucentis as a label expansion into sort of that preventive area. It's true that there hasn't been penetration into that important marketplace. And I think largely because it's really an indication where it really is like the tip of the pyramid where it's not getting dose penetration in those patients because of the durability, right? It's just not there. But if you have a differentiated durability okay? And you can really show the patients what their retina looks like, right, and where it's going because the data is very clear about the progression, okay, in those patients. And if you had, in this case, the outcome from our GLOW study, which is designed to showcase every 6-month dosing in the population on top of the data that we hope to show from GLEAM and GLIMMER, right, which, let's say, would be a 5- to 6-month predominant profile in diabetic macular edema pitch, right? So those 2 things together can be very powerful to change behavior. Of course, we don't need the NPDR indication to be a whole new paradigm for our drug and our company to be very successful. It's just that we believe that it is kind of a canary in a coal mine where it shows you that durability of today's agents is just clearly not there and that a new agent with a new kind of durability is needed, and that's what we hope KSI-301 can deliver. I mean, Jason, what do you think in terms of thoughts around market research. I mean the market research says it's with today's agents and today's paradigm. It's hard to build there. Jason, what are your thoughts?

Yes. Well, I think there's a few other maybe points to consider, right? If you look at some of the work that Regeneron is doing in this space in terms of DTC around diabetic eye disease awareness and screening, and what they have said publicly about the DR as a future growth area for them. I think people realize that it's an important part of the future for anti-VEGF therapy. And I think there's also increasingly more retina physicians who are using it right? It's a bit of a paradigm shift of vision loss prevention as opposed to vision loss treatment. And that requires people to think differently and be more aware of it. But I agree that the treatment regimens that are currently approved monthly or every other month, I think it's hard for people who haven't lost a significant amount of vision in their eye have to say, they want to get injections in their eye very frequently. So that's where we really hope that -- This combination of increasing awareness, increasing acceptance by the treating, the retina community as well as the optometry, general ophthalmology community, for referring these patients for treatment, plus the potential for a very long interval dosing therapy, like what we're hoping to show and go with every 6-month dosing, it's that combination of those things that will really allow more patients to be successfully treated.

I do think DME is going to be a commercial battlefield with EYLEA and faricimab and especially with faricimab being a single dose across all the indications. So it will be attractive financially for physicians and practices to use in competition with EYLEA. And so I think there'll be a lot of jocking in diabetic patients. And I think actually, though, if you look for both EYLEA and faricimab, the durability, for example, of faricimab in their Phase III's was 8 medium doses in year 1. Whereas that's a lot of injections in DME patients. And I think if you look at our Phase 1b data or hopefully, when we look at the readout of GLEAM and GLIMMER, the potential to really highlight that 5- to 6-month predominant profile for DME patients can be very disrupting. And so all of that marketing in that battlefield to really gather those patients, sets the stage well for KSI-301, and our commercial potential in diabetic patients, even in the existing battlefield of DME, not even irrespective of building future markets in NPDR.

We have another e-mail question here is what is your expectation of how the EYLEA arm will behave in DAZZLE? Are there concerns of better outperformance that have been seen in prior trials?

Well, when you look at EYLEA across its different studies, whether it's [ Beovu 1 or 2 ] whether it's as a comparator in the HAWK and HARRIER studies or in the TENAYA and LUCERNE studies, there is a lot of variability that 1 sees even for the comparator group, even in pseudo match studies like TENAYA and LUCERNE or HAWK and HARRIER. And in the context of the pandemic, well, what does that mean in terms of variability and comparisons to historical studies. I think that's why it's important to run it in our study. And then even within the study, to be honest, to be a bit humble given that wet AMD is a highly variable disease. Our objective is to show the noninferiority according to the noninferiority margin with substantively fewer doses. And anywhere within that range is a success if we're north of EYLEA, well, that would be nice, but we'll be humble about it if we're south. We'll focus, I suppose, a little bit more on the fewer number of doses. Having said that, the reality is that when you get down into it, it's not that useful to be very reductive because there is a lot of variability just in the patient population. And to be honest, that likely accounts for these [indiscernible] when you look at EYLEA across studies irrespective of comparing it against our [indiscernible].

We only have 1 minute, so I'm going to get this last question here, which is, I think, related to Slide 6 in your presentation. Which is why for a 5- or 6-month safety profile, would you be in line with, say, EYLEA, but 3 or 4 months, you may be slightly worse than EYLEA. And do you expect a low single-digit IOI at 3 and 4 months as well with KSI-301.

No, I think that was just designed based on a framework to think a little bit about the field. In other words, if somebody even delivers a 3- to 4-month profile, that's real, right, as opposed to EYLEA, which is a 1- to 2-month profile that the community could tolerate more adverse events if you really -- we're creating a benefit on durability. I think we don't want to rely on anything like that. Like I said, we're trying to be on the top level or certainly in the middle of -- in those middle or top levels in terms of commercial attractiveness, level of differentiation, you want to be in line from a safety profile with, let's say, EYLEA. That's certainly our objective. I do think the listener is right to say that, well, if you are fundamentally solving this durability problem, there may be the ability through benefit risk to have a different safety profile that we want. Having said that, I don't -- I hope we don't show that. And so let's not fall in trouble. Let's just see what the data show.

All right. Victor and team, I want to thank you guys so much for a productive session. We're running a little over. So I hope you guys have a great rest of the conference.

Thanks a lot Anupam. Thanks, Jason.

Thanks, Anupam.