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

All right. Let's go ahead and get started. Welcome, everyone, to the 42nd Annual JPMorgan Healthcare Conference. My name is Anupam Rama. I'm one of the senior biotech analysts here at JPMorgan. I'm joined by my squad, Priyanka Grover, [ Lorea ] Hall and Malcolm Kuno. Our next presenting -- our first presenting company of the day is Kodiak. And presenting on behalf of the company, we have CEO, Victor Perlroth. Victor?

Thank you, Anupam. It's my pleasure once again to be presenting here at your conference. I'd like to start with a quick recap of what 2023 represented to Kodiak. So on the one hand, as to the negative, it was an exceptionally hard year. Our twin tarcocimab DME registrational studies, GLEAM and GLIMMER, did not meet their primary endpoint due to an unexpected increase in late cataract events. These 2 studies were the 2 confirmatory studies we needed in one primary indication to file our BLA not meeting the endpoints was needless to say, a tremendous setback. To add more flavor, we didn't meet the endpoint, but it's how we didn't meet it. We were within the noninferiority margin through week 48. So it was very frustrating with the 60- to 64-week endpoint. We were very close. But on the other hand, speaking on the positive, we did meet the primary endpoint in two other registrational trials last year, for a total of 3 successful Phase III studies for tarcocimab, 1 in wet AMD and 1 in diabetic retinopathy in PDR, or nonproliferative DR. And what those 2 studies showed is very important. The wet AMD study demonstrated that tarcocimab and our ABC platform are safe on intensive monthly dosing and the nonproliferative diabetic retinopathy, the prevention study, along with all the other studies that we've run, continued to deliver. We continue to show the potential for 6-month durability across indications and for the majority of patients. In this case, the case of the NPDR GLOW study, we had 6-month dosing for every patient. So this is still unprecedented in the field. In addition, we continue to excel in our manufacturing and operational capabilities in 2023. In particular, with the activation and the regulatory approval of our Lonza-Kodiak joint co-manufacturing facility, Ursus. So it was an eventful and painful 2023. As some of you may have noticed, we've been quiet in the latter part of 2023. This has been a purposeful time for reflection for us as well as manufacturing execution. And we're now emerging with a thoughtful plan that we're excited about, and we hope that the community will see it the same way. We're planning to share details around this integrated plan, its priorities, its time frame, its costs, later this quarter. Often taking a step back, innovation is equated with fast-paced and immediate and sustained success. And that's truly not the case most of the time. Innovation equates most with breakthrough, where we have to figuratively and sometimes literally have to break through the status quo. That takes time. That is innovation, and it's hard and it can take longer than we expect. The question in the end is where we get to. So in summary, we ended 2023 with 3 main takeaways. First, that the ABC platform is a viable and validated platform to deliver meaningful, what I call Gen 2.0 durability. Second, we have 3 intriguing clinical prospects or candidates that can progress into late-phase trials, potentially in 2024, if desired, and that we'll talk about in a few minutes. And third, we can advance the 3 prospects without the need for additional funding. And we believe we can achieve valuable inflection points within our existing cash runway. As we move into the slides, please be aware of forward-looking statements and be -- and please carefully refer to our recent SEC filings, and in particular, the risk factors inherent to Kodiak's business. And this Slide 3 is really to continue to remind everyone of what we're trying to accomplish here at Kodiak, and that's to develop meaningfully better medicines for retinal diseases. And the way we're trying to do that is by leveraging our proprietary platforms and technology. Our enterprise and our molecules are complex, yet our fundamentals can be summarized in 3 key points. First, a foundation in science; second, our desire to develop truly Gen 2 medicines; and third, we have a 100% focus in ophthalmology and specifically, in retina. I think it's important to remember that retina is an attractive commercial area, with a few great companies, let's say, just Roche and Regeneron. But it's an arena of high interest to many companies, but they don't have a credible entry path. Kodiak, with our capabilities and our assets, we remain a credible contender here. Progressing into Slide 4. I've talked a little while about how we got here. Let's now move on to where we are today and what does Kodiak's future look like. I talked a bit in the intro that we have 3 intriguing clinical prospects. And here's a summary of them. I'd start by saying that we have the capital to advance these 3 prospects to meaningful inflection points within our cash runway. Importantly, this doesn't mean that we would not continue to invest in our platforms, technology and pipeline, which can translate into business development opportunities as well. Most of you who are listening know about tarcocimab, previously KSI-301. We're planning on running one more pivotal study to have a set of 4 studies across 3 indications, and that could support a BLA package of nonproliferative retinopathy, retinal vein occlusion and wet AMD. And importantly, we're planning on running these studies with a go-to-market formulation or drug product that we believe can improve injectability and usability and thus, improve some of the liabilities we've seen with the clinical formulation in terms of its gel-like nature. And we can talk more about that in the Q&A, if desired. As we continue to progress our pipeline, our next molecule, based on our proprietary ABC platform, is KSI-501ABC. It's a first-in-class bispecific molecule. It comprises an anti-VEGF trap targeting VEGF-A, B and the placental growth factors as well as an in-line anti-IL-6 antibody. So while anti-VEGF therapy is the gold standard target for retinal vascular and exudative diseases, we know that inflammation is a major component of the disease, and it's not addressed by anti-VEGF monotherapy. Specifically, IL-6 has been demonstrated to be upregulated across all retinal diseases, and inhibiting both pathways may provide a significant advantage for KSI-501. We'll be presenting our Phase I KSI-501 study data at the Angiogenesis Meeting coming up on February 3. And as we think about our clinical activities through our cash runway, we're considering to initiate a rapid Phase II/III pivotal program this year, with inflection points within 2 years, let's say, of study initiation. But we've not made any final determinations as of today. A third exciting prospect is KSI-501P, which is the same KSI-501 protein, but it's unconjugated. So it's just the antibody trap portion, like a regular biologic. The obvious question is why we would not use our signature ABC Platform? And the answer is simple. The ABC Platform's purpose is enhanced durability. We're developing KSI-501P for inflammatory diseases where durability is not the main priority. So in addition, by decoupling the protein from the ABC Platform, we have a clinical prospect with uncorrelated risks, effectively expanding our portfolio attractiveness, and I believe its overall probability of success. So it's important to mention that we're focused equally on these 3 exciting prospects. Of course, it takes operational excellence to develop 3 programs at the same time, and we have demonstrated that we're very good at that. So in summary, and you can see here in Slide 5 in these sort of pictographs, across the 3 molecules of tarcocimab, KSI-501ABC and also KSI-501P, we have a pipeline with a clinical synergy, which is helpful from the standpoint of execution, but it's also a pipeline with a diversification. Let's run fairly rapidly through a bit more detail on the 3 programs. We'll talk quickly about tarcocimab. So we have 3 successful Phase III studies with tarcocimab in nonproliferative diabetic retinopathy, retinal vein occlusion and wet AMD, with compelling durability demonstrated. With the NPDR study, every patient on 6-month dosing. In the RVO study, we demonstrated strong durability at the primary end point at month 6 and also at year 1. And in wet AMD, the goal wasn't durability. The goal was to demonstrate safety on intensive dosing, which we did demonstrate. Now across -- on Slide 8. Across the pivotal program, our durability with tarcocimab was consistently demonstrated and in multiple indications that delivered on 5- to 6-month durability for the majority of the patients. In RVO, we doubled the interval to 8 weeks versus aflibercept Q4. In DAZZLE, our Phase IIb study, we took 60% of the patients to 5-month dosing, and they did very well. In the DME studies, GLEAM and GLIMMER, although we didn't meet the endpoint, 2/3 of the patients coming out of the loading phase were on 5- and 6-month dosing and doing very well, and 50% were on 6-month dosing at the primary endpoint. In the GLOW study, as I mentioned, all patients were on 6-month dosing. And again, in the RVO BEACON study in the second 6 months, half the patients didn't need to be treated at all, even though they had received several doses less than the aflibercept comparator. So it's a powerful platform for durability, and it's built with a science of durability, a Gen 2 durability platform and agent. On Slide 9, the other thing that's really important is that tarcocimab and the ABC Platform have demonstrated a favorable safety profile with now more than 2,400-patient years of exposure across the full clinical program, with rates at intraocular inflammation across the pivotal studies very comparable to that of other marketed anti-VEGF agents. On Slide 10, in the recent GLOW1 study in PDR, tarcocimab established superior efficacy within every 6-month dosing and reduced the risk of developing site-threatening complications by more than 90%. So a nearly 30x increased response ratio of 41% versus 1.4% against Sham, in terms of the primary endpoint of 2-step improvements in their diabetes in the eye. And a key secondary endpoint, a 90% or 89% risk reduction of the patient having a site-threatening complication. That was phenomenal data that really helped us refocus on the value of tarcocimab and the ABC Platform for patients and the community. In Slide 11, in RVO, we showed strong durability, matched efficacy and a comparable safety profile with fewer doses than aflibercept. On Slide 12, as I mentioned, in the wet AMD DAYLIGHT study, we showed that month dosing of tarcocimab had a favorable safety profile, non-inferior efficacy and supports monthly dosing of tarcocimab across its potential label. On Slide 13, I think it's important to say, well, if there's a high probability of success in the fourth confirmatory pivotal for tarcocimab, and let's imagine that we run that study and it's successful, well, what is the potential of tarcocimab in a complex commercial marketplace? While we believe that tarcocimab's breadth of indications, differentiation with the science of its durability and design and the thoughtful commercial strategy may translate into meaningful commercial opportunity for the molecule. Importantly, the signature durability profile of our platform and of tarcocimab has gathered early positive feedback from practice owners and retina physicians in terms of 6-month dosing, with the majority of patients in our Phase IIIs having done very well on 6-month dosing, twice a year. We also have the flexibility of monthly dosing in the label from the DAYLIGHT study, the potential for filing a single BLA across 3 large indications at launch, the underlying science of our design and durability in our platform being highly differentiated versus other newly commercialized molecules that I termed Gen 1.5 and the ability as a biotech to have flexibility in our commercial strategy. So in summary on Slide 14 for tarcocimab. Across the entire program, tarcocimab and the ABC Platform have demonstrated consistent and differentiated durability as well as favorable safety. We have 3 successful pivotals in 3 large indications. And we have a plan and are planning or thinking of running a fourth confirmatory pivotal, which will also bridge in our go-to-market formulation. And that we can complete that fourth confirmatory pivotal, we believe, within 24 months. Moving quickly into the KSI-501 program on Slide 15. We have sort of some nice electron micrograph pictographs of how our ABC Platform, with the biopolymer and the antibodies that is conjugated to, bind their targets. And in this case, we have binding both to interleukin-6 inhibition and also VEG-F inhibition. So it's kind of fun to take a look here at the pictographs. But on Slide 16 for KSI-501, which is a first-class bispecific molecules that inhibit interleukin-6 and VEGF. And we're going to look to develop it both as an unconjugated protein for retinal inflammatory diseases, which is a broad new category and also as a conjugate, which will represent as a bispecific molecule, all of our learnings across the tarcosimab program put in also into the KSI-501 program as a conjugate. And why is the addition of a new mechanism, say, within KSI-501ABC useful? And here on Slide 17, this is an interesting plot, which really shows the tremendous variability of response. In this case, to aflibercept across our pivotal studies with anti-VEGF monotherapy, showing that there is a substantial portion of patients that are underperforming and that additional mechanisms of action, together with durability, such as our KSI-501ABC, there's tremendous potential. So Slide 19 shows a little pictograph of how the KSI-501 protein binds to the different targets. On Slide 20, just a quick preview on the picomolar potency of KSI-501P and ABC to both targets, in comparison to the Roche IL-6 and also, of course, to aflibercept. On Slide 21, looking at a little bit of preclinical biology to share. Looking at how the KSI-501ABC molecule has a powerful potential for new biology. And on Slide 22, that you can confer superior normalization with the bispecific than you can with either of the monotherapies. And what about very quickly on the KSI-501P program, the unconjugated bispecific protein on Slide 24 and 25? I think it's important to recognize that there is still a very high unmet need for safer, disease modifying therapies for patients -- retinal patients who have inflammation with or without fluid. And the opportunity is primed for a safe and effective locally administered biologic to move up the treatment pathway towards first-line therapy, with anti-VEGFs being third or fourth line, systemic biologics being second or third line, and immunomodulators or local steroids being second line or first line. So there's a tremendous opportunity. Outside of the anti-VEGF market for physicians and retina doctors are very interested for such a biologic as KSI-501P. And IL-6 plays an important role in retinal inflammatory diseases and has demonstrated clinically meaningful improvement in vision and resolution of edema in uveitic macular edema patients. And the addition of anti-VEGF in Slide 27 to anti-inflammatory agent therapy may provide further clinical benefit than anti-IL-6 alone. And on Slide 28, briefly, a little bit of introduction, again, on the KSI-501ABC program. Our objective here is to think more about bringing the ABC Platform, again, for high prevalence retinal diseases, to maintain the signature durability of the platform, while benefiting from the learnings of the tarcocimab program. So it's currently in its Phase I study in DME patients. And as I mentioned, we'll be presenting data from that study on February 3 at the Angiogenesis Meeting. And further studies will be conducted with a similar commercial/go-to-market formulation that represents the learnings from the tarcocimab program. And we're exploring initiation of dual pivotals in 2024 in the high-prevalence retinal diseases. So with that, I'll stop here, and we can have a little bit of a discussion. Thanks, Anupam.

Thanks, Victor. So I want to remind folks, and some of you have probably heard this, there are 3 ways to ask a question. Raise your hand. I'm happy to call on you. You can send me an e-mail or you can submit a question to the portal, and I'm happy to ask on your behalf but I'll start out. I guess as you think about it today for tarcocimab, what are your latest thoughts on sort of indication selection for that next pivotal trial? And what's on the table and what's definitively excluded?

Thanks, Anupam. With the tarcocimab program, as we mentioned, the idea is to have 2 successful studies in 1 of the indications and then Singleton studies in adjacent indications. And with the GLEAM and GLIMMER studies not meeting the endpoint, we're left with 3 successful Phase III studies, one in each of 3 different diseases. So the goal is to determine, well, which of the different diseases could we run in a confirmatory pivotal to bring that to have 2 studies. And then the other 2 indications would have 1. That's consistent with our prior FDA interactions. So the question then is, well, what's the right indication to run the study? And what's the framework in which we want to make the decision? And fundamentally, we look at either wet AMD or non-proliferative DR. So sort of another wet AMD study or another GLOW study, like a GLOW2. So there are pros and cons to either one. We haven't made a final determination. I think one element of the framework is we would like to have a very high probability of success in the study, not just our own view, but that the perception of the PTRS would be high, say, from the standpoint of analysts, such as yourself or investors. So people don't feel like they have to hold breadth through 2 years of study. Rather, we'd like them to feel confident as soon as we announce the study and begin executing it, that it can be successful. And then people can turn their attention not on the estimated probability of success of the study, but rather, okay, let's assume the study is going to have the same safety and say, positive efficacy, right? Well, let's now talk about the commercial opportunity for the molecule based on the profile. So I think we haven't made a final determination. Our view would be though that we would want the study to have a high perceived PTRS. And so therefore, within wet AMD and the non-proliferative DR. Those are really the 2 choices. And I would say, at this point, we haven't made a final determination.

If I could push you a little bit, though, correct me if I'm wrong, but GLOW is just a Sham control, right? And so if you're thinking about probability of success, why wouldn't you pick non-proliferative DR?

Right. Well, that does seem like an obvious choice. I think while we look across the whole pipeline of the 3 programs that we want to be developing and let's say moving expeditiously into pivotal studies and maybe even pivotal study starts this year, running a GLOW2 study with relatively low complexity. And again, Sham makes a lot of sense. I think we just maybe, in our own minds, want to make sure from a regulatory standpoint that that's going to get us to where we need to go.

GLOW2 has got a nice ring to it.

It does. Doesn't it?

Yes. When you think -- questions from the audience? I guess, if you have to look back on the tarcocimab, everything that you've learned across these studies. What are the key learnings that you have to apply, whether it's a wet AMD or GLOW2.0 at this point?

Well, I think it's what I said that being an innovator is hard. Our platform, building antibody conjugates for retina, it delivers the differentiation of that true like Gen 2 durability that Vabysmo doesn't have and EYLEA HD doesn't have. And as a biologic, even if you look several years in the future, we can still be a core differentiated and important agent in the market. And the durability signal is very strong. Where we really fell short was on the cataracts, where we had 2x the cataracts and GLIMMER than aflibercept did. Only in diabetics patients and let's say, severe diabetic patients. So the key learning there was a gel-like nature of the drug product where we really had ratcheted up the concentration, where it took 10 seconds to dose rather than say the 1 to 2 seconds of aflibercept or Vabysmo. Really turns out that both the injectability parameters and usability for the physicians wasn't good. And also the gel-like nature meant that it wasn't dispersive, let's say, enough in vitreous after you dose it. So in some cases, let's say that the gel will get trapped between the capsule, the back of the lens and the vitreous itself. So for me, the key learning is, as we move to a commercial go-to-market drug product or formulation and as we scale that, the ability to make small course corrections, right, to be an innovator who listens to the science and says, "Wow, look at everything that we've learned." The science of the durability and the design can deliver 6-month dosing for the majority of all patients but in GLOW, for 100% of patients. That's still going to be really important over the next, say, 5 years or 10 years for patients as a biologic. We can still get there. And so the key learning, I think, is to listen and how to make agile, small course corrections along the way and to say, "Look, we're not [indiscernible] that we're going to keep pushing our platform in our molecules just because we have the capital." The point is to put it in the context of our portfolio and to say, "We believe we know how to get the molecule approved. And on the basis of having approval in our discussions with the community, they're excited about it." Patients didn't want to get off the medicine when we shut down the studies. The majority of the patients were doing really well. It turned out it was really a few tens of patients with cataracts that destroyed the GLEAM and GLIMMER studies, but the majority of the patients did well. So the small course corrections can fix that or certainly, bring the rate down closer to background. And so that's the key learning, that we're close, and we can apply those learnings across the ABC Platform generally. And some people will say, "Well, I'm not smart enough to really know whether the ABC Platform is good or bad or what you're going to do, but the KSI-501P program, which is the protein alone, right, as a novel bispecific agent, allows those investors, let's say, to buy into Kodiak and our pipeline development because that molecule alone is innovative and can create a whole new category of locally administered biologic, right, for retinal inflammation, which almost every patient has."

And then maybe for KSI-501 that you just mentioned, that data is going to be at Angiogenesis next meeting -- next month. What's going to be the size and scope of that data? Maybe talk about the severity of DME patients in that study? And like how would you define a win scenario and moving forward?

Yes. Well, that Phase I program is done with the KSI-501ABC molecule. And from there, it's a jumping off point both for the protein as well as the 501ABC conjugate. So we want to be looking at the data from the standpoint of both of those, that dual future. What's important is that the study shows safety, number one. That the study show signs of bioactivity, both in terms of vision as well as OCT, right, fluid. The Phase I study, obviously, for a new molecule -- in this case, DME, should showcase, as I said, safety, vision and OCT fluid bioactivity improvements. And the population that you go into in the Phase I is generally fairly severe DME previously treated patients. Luckily, I think half or slightly more than half of the patients in our Phase I with KSI-501 are treatment-naive patients. And so as a result, that's really the subpopulation, let's say, of patients where we can take a look at, in a more clean manner, the vision and OCT responses. So for me, we have a nice design of that Phase I. It's in one of the slides here as well, where we actually dose escalate through 4 different dose levels, and each patients gets 3 injections once a month. So it's a Phase I that's really, say, a Ia, Ib, or I/II, or say, half the patients are treatment-naive. And so we should be able to have a useful view of safety, vision and OCT in a Phase I study. And then that provides a jumping off point for us. I think to -- across 4 different dose levels, right, a jumping off point for us to think about the KSI-501P and the KSI-501ABC. And the other thing that, Anupam, that's sort of important is that there's a Roche anti-IL-6 molecule. And although they dose escalated through many different dose levels, for example, in the 2 pivotals that are moving forward for approval in uveitic macular edema, they've actually only chosen, I believe, their 2 lowest dose levels for development. And that's really driven by some liabilities of the molecule. So as we look at our Phase I and we look at safety across the 4 different dose levels, if we had safety across all 4of that, that's actually a positive feature for this molecule versus competing molecules in development.

Questions from the audience?

On the EYLEA variability slide, I think, '17 or '18, so it's clear that there's a lot of variability between patients, but it also looks like there's variability within, like, some patients, like, from week to week, right? How does that kind of inform your strategy going forward? Like how do you think about that? Because that seems like it maybe creates a lot of noise there.

Right. Well, a lot of that is the variability associated with EYLEA. Let's say, aflibercept, not having sufficient durability between injections, right? So you either have a variability of flutter on vision and/or you have a flutter on the fluid. So one thing that Kodiak has shown is a stronger durability. And so we believe that's important. So this is a very interesting and unusual data set that we decided to showcase. It also shows the complexity of doing drug development in retina, where best corrected visual acuity is the outcome measure and it's very noisy. So part of it says, "Well, within the KSI-501ABC program, right, we want to develop our conjugate with the KSI-501." Well, what is the right development strategy, right? So if you have anti-IL-6 on top of the anti-VEGF -- and also remember that what's kind of cool about our KSI-501 protein is it's a VEGF trap in line to an anti-IL-6 antibody. So it has a slightly different and more interesting VEGF biology, which EYLEA also has. So we get a bit closer there. We bring the signature durability of the platform, and we're bringing what we call sort of that go-to-market kind of mixture of the unconjugated and conjugated in this defined ratio. So KSI-501ABC represents the culmination in some way of like more than 10 years of Kodiak's innovation. And the point here isn't that we know exactly which large subpopulation the anti-IL-6 is going to solve, okay? But we may decide, let's say, to develop KSI-501ABC kind of broadly as treatment-naïve within the retinal vascular diseases and maybe bang that out for approval with pivotal studies because we have the capital, let's say, to do that, okay? And then once it's approved, to try and let the physicians really help us figure out which large subpopulations within this variability, right, are going to be best served by this next-generation molecule. Okay. So that represents both opportunity in retina, right, and also some of the complexity of doing drug development retina. But that's why there's really just Roche and Regeneron in the space because it's hard. But let's look at all of these other companies on the outside looking in this thing. Everybody has 2 retinas and with diabetes and with aging, right? There's a tremendous demographic opportunity. And everybody wants to pay, let's say, hopefully, right, to keep -- to retain vision. So -- and we want to be part of that.

Makes a lot of sense.

Any final questions from the audience? Thanks so much, Victor.

Thanks so much, Anupam.