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

Great. Well, good afternoon, everybody. Thanks for joining us for the next session. I'm Matthew Harrison, one of the biotech analysts here at Morgan Stanley. Really pleased to have Kodiak with us. Quickly before we get started, I just need to read a disclosure statement. So please note that all important disclosures, including personal holdings disclosures and Morgan Stanley disclosures appear on the Morgan Stanley public website at morganstanley.com/researchdisclosures. So with that, pleased to have Victor Perlroth, who is the CEO of Kodiak with us. Victor, I thought maybe a good place to start is just in the last -- it is the last week.

It all blends together, but we got some -- we've got some data from Regeneron on high-dose. I think that just maybe a nice place to start is just how do you think about the competitive market at this point? There's been -- between high-dose EYLEA and aflibercept, there's been a handful of developments in the market.

Yes. Well, thanks, Matthew. It's good to be at the conference again. Just first a little housekeeping on our side. I may make forward-looking statements. There's a lot of risk in the business. Please review Kodiak's SEC filings for a more comprehensive view of the risks and the risk factors. Well, there's a lot of news in retina over the last week or the last year in fact. I think that's great for patients. Number one, I think it also reflects that retina is an important area with a lot of unmet need for patients and also a lot of value for investors and for drug developers. So that's very good. There remains substantive unmet need in the space. I think what was also nice to see based on the impressive top line data from the Regeneron high-dose EYLEA aflibercept, and was the nice market response with the Regeneron stock. I think that really does showcase the importance of durability in retina. And Kodiak has, I think, the predominant durability platform in retina. So that's important to remember. And I think we -- we think the Regeneron data is an upside surprise and looks strong. Of course, we're going to be very interested to take a look at the particularities and the details of the data when they would be presented shortly. The level of differentiation will be really important, obviously, in the marketplace that Kodiak's KSI-301 or tarcocimab tedromer, whichever nomenclature you choose to use at this point. When we drop into the market, what will our differentiation be? I think our aspiration has been and remains to really be targeting to be a definitive 6-month medicine in the majority of patients. We still think that's a differentiated place to be. Of course, it's up to us through our development and in particular, the ongoing Phase III studies to really demonstrate that. So we were pleased in RVO, which is a tough disease in BEACON to be able to showcase a doubling of the interval versus aflibercept and to show very strong data. And then in our GLEAM and GLIMMER DME studies, which I think based on the BEACON data and the strong immediacy that we saw there, really increase and improve our own confidence that we're looking at a strong outcome in GLEAM and GLIMMER. And where we go from Q8 week all the way to 6-month dosing. And then also, of course, in our GLOW study, which is in diabetes, non-proliferative, which is a prevention and treatment study, really where we are every 6-month dosing versus Sham in that study as well. So I think we want to put together the portfolio of studies well-executed with the data over the next year that showcases our capability of our platform and our lead molecule as a 6-month medicine for patients. It's a massive market opportunity. Durability is clearly important. There are existing medicines, as you mentioned, right from Roche and Regeneron, I think they both have strengths and weaknesses and I think they leave room for our value proposition. And certainly, not just for our lead molecule, which we're excited about and which we're executing well, but also for our pipeline with our dual inhibitor, the IL-6/VEGF, bioconjugate our KSI-501 that's very close to IND and we believe in clinic early next year as an industry-leading, best-in-class bispecific. I think what's important in retina is the tremendous unmet need, the tremendous market opportunity. Kodiak's 10-plus year investment and capability narrowly in retina, which is a broad area and as we look at the unmet need, it's not just the durability, but it's bringing additional mechanisms of action into retinal disease. And as you look, our platform is safe, it has a tremendous durability. We're going to come out with the bispecific and then as we look further forward, we're also bringing other small molecule mechanisms of action into our ABC platforms. So I think we have a lot of capital, a tremendous amount of capital. And we're I think deploying that in a thoughtful manner. We always remain very agile as we see new data and try to learn from that. I think circling back to the beginning we're intrigued to look under the hood at the Regeneron high-dose EYLEA data. And we think there's a nice spot for Kodiak with tarcocimab in the marketplace.

Okay. Great, great and perfect. And so maybe we could talk about BEACON to start out with. And I think a couple of things. So everything to a certain degree, I think investors are going to look at look through the lens of DAZZLE a little bit and say, okay, so BEACON was successful. What do you think led to that study being successful where DAZZLE wasn't? And then really what I'm getting at is you obviously made some changes to the GLEAM and GLIMMER studies to try and maximize success there. So what does BEACON tell you about those changes that you made?

Right. Thanks a lot, Matthew. The great thing about DAZZLE is it did show the majority of patients on the top regimen of the 5-month regimen and those patients did extremely well. So the concept again of Kodiak's medicine across the indications being a 6-month medicine in the majority of patients, that's our aspiration and we'd like to build towards that and to be that to get the medicine available for physicians to really prove that. Wet AMD always is and was a little bit of a different disease and the macular edema type diseases of retinal vein occlusion or diabetic macular edema being a little bit of part does DAZZLE results help us and the community understand the distinction between those 2 types of diseases. For example, in our Phase Ib study, of course, our RVO and our DME data, we're always extremely strong and analysts and investors and ourselves ask questions around, well, is our wet AMD data on the margin a little bit weaker than the RVO and the DME data? And I think what the DAZZLE result helps us to understand that in a small minority of patients was the conjugate weaker because a question coming out of DAZZLE as well, I mean, is the conjugate -- does it ever reduce potency. I think the answer from DAZZLE, our understanding is a small, is it single-digit percent or what if patients. It did seem to be weaker, and that had a big impact in the study. So we had a study design that dragged all the patients too far, which in the high-dose EYLEA study, they don't do, right? There's patients on Q8 week dosing and then they have patients that jump to the higher interval. In DAZZLE, we didn't allow that. But also is there a small percent of patients where the potency seem to be reduced, perhaps based on the large size of the conjugate. And our questions are how do we think about balancing durability versus immediacy? So that's an important development question. But when you think about retinal vein occlusion and DME, those are more intraretinal diseases where we saw such strong strength in the Phase Ib. And so when we -- what was important for Kodiak as we looked at the BEACON readout, first of all, in terms of mass data, it was very hard to see what drugs patients are on actually because they did so interchangeably well actually. But what was important for us was to look at the immediacy of the effect. And what we saw in BEACON is really at that very first data point at the week 1, we see tremendous strength of KSI vis-a-vis aflibercept. And then, of course, we see nice narrowing and convergence of the curves through the 24-week endpoint, both in the RVO subjects and then in all population. Why that's important for GLEAM and GLIMMER is I think we asked ourselves, well, would we see a potency deficit in BEACON broadly? The answer is we don't. And so that helps us understand DAZZLE a little bit. But then because we don't, we understand the drug is extremely potent. We do have that durability where we dragged all the patients, 100% of them to a doubling of the interval in RVO. And then for DME with this understanding, of course, that the potency and the immediacy we saw in the Ib, we expect to see in GLEAM and GLIMMER. And so we have a strong confidence therefore that we can have a strong potency in GLEAM and GLIMMER. Now the changes that we made based on DAZZLE and GLEAM and GLIMMER, the algorithm, DAZZLE, it's like -- when you look in retrospect, it's like an airplane that crashes and it doesn't crash for one reason. Often airplanes crash. There's like 5 sort of reasons that kind of combine and then the airplane crashes. DAZZLE in my view is a little bit like that. And one of the things in DAZZLE is our algorithm really allowed patients to kind of go too far before the algorithm caught them and reduced them from say 20-week dosing down to the 12-week dosing. And so in GLEAM and GLIMMER, what we wanted is these disease activity assessments that shift patients from Q24 to say Q8. What we did is we slightly modified them so that they were more sensitive so that we would catch the patient as their disease was worsening. But before the disease had really fully reactivated. So that will interfere on the margin with our durability proportions, but what's most important as we see in the Regeneron study with high doses to meet the endpoint. So I think we believe the changes that we made on the margin and the protocol on the algorithm, which remove subjectivity and catch the disease reactivation a little bit earlier, plus the feedback from the BEACON where we see that strong immediacy, we have a stronger confidence as we can in successful readouts in those studies. And then together with the GLOW study, which is fully enrolled and is more of a treatment and prevention study in non-proliferative where we dose Q6 months. So we think the BEACON data combined with GLEAM and GLIMMER together with GLOW and then of course, we're running the DAYLIGHT wet AMD study, which is our monthly study, that will answer a lot of questions about the potency across the broad population and wet AMD obviously be on a maximum interval. But it will provide a lot of information together with the readouts of the other studies, all of which we should have within the next 12 months, for us to really understand how we develop the value proposition of a 6-month medicine, right, in wet AMD as well.

Okay. Okay, good. So I think that highlights a couple of things. Why don't I take them in this sort of -- so one of the top -- just to go back -- and then one of the top questions that I think I get is about inflammation and about how to just think about that. So I guess the question is do you think you have an inflammation signal and is that a headwind at all? And just how to think -- I know it's very low, but how to think about that? Or is that just the rate we see across a lot of agents anyway? So maybe you could comment on inflammation.

Yes, I mean I think BEACON is a good study to look at or maybe you want to compare that against the high-dose EYLEA studies or you want to compare that against the faricimab studies. We're really clearly in the ballpark. I think of course, you start at the top and say well, is there any blinding type of inflammation, right? And the answer certainly is absolutely not. Okay. So that's the most important. And then the second is like, well, what kinds of inflammation and what severity do you see and are they responsive, right? And the answer is the inflammations that we see are very low grade. And I think the highest was maybe like a 2-plus vitreal cell in the BEACON study, which is really quite low and I think we had 4 cases, right, and I think I only had one case. So the point is, when we're doing early drug development, right, through even Phase III, to achieve this level -- low level of inflammation, there appears clearly to be no platform levels of inflammation, which is I would say a tremendous and great finding. And to compare against aflibercept, which is in commercial manufacturing, right, or Lucentis, which is in commercial manufacturing and making so many batches of substance and product each year is really like a fairly unfair comparison. Despite that, we're doing tremendously well. I think that's important. When you get to a situation where you're basically counting fewer cases of inflammation than your fingers, it's really not very useful and it's actually an unreasonable standard. The fact that we're able to compete with such low levels of inflammation is tremendously positive. So I think it's misguided. That's not to say inflammation is not important. I think what people are really saying is, well, the commercial landscape has products that have let's just say 2 main EYLEA and Lucentis that have a level of inflammation that's extremely low. And in our study, right, in DAZZLE, we showed 0 cases for aflibercept and I think one case in BEACON. I think in the high-dose EYLEA studies, it was a bit more diverse. I think one of the groups had 0 and other groups had above 1. Of course, back when Lucentis was in this lyophilized formulation when it did have 10% or 20%, but that's not a useful comparator. The point is that as you move to commercial scale manufacturing and as it just becomes more rank and file or more turning the crank on the manufacturing, we don't have a platform problem and so there may be elements that are on the margin and it will only become cleaner let's hope.

Okay. Okay, great. No, good. Thanks for addressing that because I think it's an important topic. Second question is just a follow-up to some of the comments you were making about the study program, et cetera. So I think one of the other questions I get is, well, you obviously will have a monthly wet AMD study and let's just assume -- let's sort of just play this out. So you've got positive RVO results, assume you have positive DME results. I think most people think if you have positive DME results, you're going to get positive results on retinopathy as well. And so then is that the point at which you go back and you say, well, we're going to figure out how to get an extended regimen in wet AMD or how do you think about that? Because I think one of the questions -- another question that I get a lot is, okay, so maybe this regimen can work and can be registered. But if you don't have a label for wet AMD, do people stock it because they don't want to stock 5 drugs in their fridge, they want to stock 1 and be able to use it across all their patients. So it's a commercial question, I guess.

Right. Well, we don't want to develop -- I mean, the goal is to think how we develop a medicine that can be first line, okay? And so that's what we think about, how do we be agile in terms of thinking about our development with our Phase IIIs such that we could be that medicine. I think in the United States, of course, physicians are more interested in the data packages than they are, I think, in what the registered label is because if you have monthly dosing, you can use it on any frequency. The reverse isn't true, whereas if you have every other month, you can't necessarily use it monthly, right? So we believe that we hear from physicians for the patients saying DAZZLE who were on the longest regimen that the patients were doing tremendously well, and they really like the medicine a lot. They would call as they were disappointed in the DAZZLE result, especially because we have a bunch of patents that clearly are doing super well on super long interval dosing with your medicine. So I think in wet AMD, I think the goal is to get the drug approved on the basis of DAYLIGHT in the context of the broader portfolio. And meanwhile, to be thinking aggressively, but in the context of being capital-efficient and investing capital at the right time to be thinking about when do we think we have the right amount of information to think about a better study for wet AMD that can really showcase with the decreased noise and the study can better showcase the immediacy as well as the longest durability and it's a massive market. As I said, there's tremendous unmet need. There's a lot of new papers coming out, I think, on an annual basis, really looking at -- even if you look at the TENAYA and LUCERNE studies with faricimab where aflibercept was the comparator at the end of year 1 to the end of year 2, they're losing, let's say, almost 3 letters. So there's a lot of opportunities here. And we think that we can build the right momentum around the drug with the current portfolio of studies and then thinking about what would be future studies done in a cost-effective manner.

Okay. Okay. So then maybe just to summarize, it sounds like the initial commercialization plan would be to have wet AMD on label with monthly dosing, let physicians use PRN dosing with that as they choose they want to use it there and sort of get the interval that's right for the patient. And then at some point, maybe starting before approval or starting after, depending on what makes the most sense, have a study to showcase sort of what the -- maybe potentially what the most -- what the average interval or whatever you want to call it for most patients that they can achieve in that setting. Is that broadly speaking sort of the right way to think about it?

Yes. Yes. And I think the anchoring on the DME and the science of the conjugate and its durability from say, a preclinical and scientific standpoint and then anchoring in the clinical data from the DME, I think, is a good place to start.

Okay. Okay. Good. Manufacturing, you've been working with Lonza for a while, so maybe just update people on where you are in terms of scaling and CMC package?

Yes, well we have a number of efforts obviously with Lonza and we greatly enjoy the working relationship with them. Obviously, when you have such a close and important partner, you have a lot of back and forth. Our dedicated facility for the manufacture of our conjugate is completed, and we'll be moving soon into kind of what they call the [ water runs ] and from there into early engineering GMP batches at commercial scale. And it's an impressive facility by design and uses say, the similar team members that manufactured our clinical batches. So there's a lot of overlap there. Manufacturing for retina is pretty serious business. You need your overall purity levels of say into toxins in water and buffers to be extraordinarily low. So there's a lot of work to do to make sure because you use many metric tons of waters and buffers in what you manufacture. And then as you concentrate, those impurities can be concentrated. So we've done a tremendously good job together with, let's say, Lonza and our other partners in our manufacturing to-date and as we now scale to larger scale for the antibody, the biopolymer and the conjugate and also in the drug product. So I think before DAZZLE, our objective was definitively to have the manufacturing exactly in lockstep with clinical. I think after DAZZLE, we wanted to be anyway I think a little bit thoughtful about aligning those readouts and aligning the commercial scale because they're quite expensive in terms of building commercial inventory and stuff like that. So I think our manufacturing may be pushing out a little bit, but that's really a little bit more in the context of being capital-efficient and our dedicated Ibex facility with Lonza is making very good progress. And also we've, for example, completed our first clinical prefilled syringe fill as well. And so all of these elements are kind of moving and converging and all of those pieces, whether it's the biopolymer, will shortly be made for validation batches at commercial scale, the antibody similarly. And then the Ibex is where the dedicated facility are called Ursus is where we bring those pieces together at commercial scale and we hope that we'll be able to begin those early engineering batches like before the end of this year. So I would say the top line answer is the manufacturing may be shifting a little out from the clinical that's on purpose. Having said that, aggressive time lines that we had aspirationally for the different pieces in most cases, we're achieving those. I think the prefilled syringe is also an interesting one. It's one thing to make the batch. It's another one to get it into people and then another one to see the data. But again, that's also likely to be helpful from the standpoint of inflammation and stuff like that.

Okay. Okay.

Yes.

Maybe can we talk a little bit about retinopathy. I think it's probably an indication that investors have spent less time on broadly speaking, obviously, a lot of potential patients there. Laser is a bad choice, right, just to put it bluntly. And -- but I think also potentially I'm curious to see if you agree, but a very long duration regimen could have -- could be very attractive for those patients because just like with DME, they're not really seeing at least the start of the disease, right, the disease, if you haven't experienced the progressive nature of it, right, you might not think it's that bad. And so if you hear you've got to get a needle stuck in your eye every month or whatever, that doesn't sound that appealing. So how do you think about that indication? And how do you think about long-duration regimen there as a potential maybe outsized area for something for a long duration to be showcased?

Great, for a technology and a product like ours, that's really ultra-long interval, this concept of treatment and prevention. We call it prevention of worsening is really, I think, important whether the marketplace today recognizes that there's a lot of value if they were or not. I think it's something that will come. I mean simply put what you can do is you dip your hand into red paint and then splash it on the wall. That's what the retina looks like for patients that have non-proliferative diabetic retinopathy. They have these microhemorrhages that are ongoing for a long time and they're currently untreated. So the community today could say, well, it's not a big deal because eventually they'll have worsening of disease, they'll get diabetic macular edema and then we can treat that. But if I'm a patient, if you say, well, you have to come in monthly or you have to come in for 5 loading doses or ex-loading doses. But in our case, we move into a Q6 month regimen. We do think it can be a game changer if you have the right safety. So again, it's a very large market today and I think it's growing. And also we even have our KSI-501 -- 6 VEGF bispecific, which we think will go deep in DME as well. So we think, again, the concepts around the GLOW study with its Q6 months and then the Q6 month element within GLEAM and GLIMMER, right, provide an anchor to lock the long interval dosing element and to get the drug out there. So people use it and begin to use it also in wet AMD and then we buttress that with subsequent studies that really lock in and load around our aspiration of being the longest interval medicine.

Okay. Maybe a minute on your next drug, the IL-6 -- just talk about -- obviously, let's assume you get an IND in next year, where do you go with that first? And what does sort of proof of concept look like with that drug?

Yes, well, we think we can get the IND in this year and that we can be live in, let's say, I mean early next year right aspirationally I would target Q1. It's really more of our resource element. We have a lot of studies ongoing with tarcocimab. And so bringing additional clinical studies onboard is just a question of priority. But it's an exciting molecule, right? There's a lot of inflammation that goes on for decades in all of these diseases. So this molecule uses the same biopolymer and it uses the same antibody structure that we conjugate into. It has a VEGF trap in line with an anti-IL-6 antibody. It's a fusion. So it should be quite potent on the anti-VEGF and quite potent on the anti-IL-6. So we're refining together with the clinical community a little bit. Obviously, like a Phase Ia type study, it's just a step loading starting fairly low, like we did with the KSI-301 where we did 3 patients at 3 different doses. In this case, as a bispecific, we may take it a little bit more slowly. But fundamentally, we'll step through early single-dose studies, I would hope fairly quickly. Then the question is we want to go into like a Phase I/II type study. Whether this time, we would include active comparator a little bit earlier, I think, in the Ib, that would have been helpful to direct us a little bit to catch a little bit earlier what type of -- what level of risk do we take in the next set of study designs would have been helpful in retrospect. So -- but as I mentioned, we think looking at DME with that bispecific, which is, I think, first-in-class, best-in-class is going to be a great area to go. And again, it may not just be about durability, but you may be able to look for differentiated types of efficacy out the gates. And then, of course different types of uveitic macular edema is another important area where there's a lot of patients. I think we want to broadly explore the medicine. The Phase Ia type study would be I think fairly straightforward and our protocol may bridge immediately more into a multiple-dose study. And so I think generating clinical data the beginning of next year and then towards the end of first half, more on the Ia, right? But even in the second half, if we move I think in the 301 Phase Ia first injection was July and then the first dose and the multiple dose part of the Ib was in January of the next year. So we should be able to maybe not do it quite that fast, but be able to really begin generating data for a molecule that I think is further validating the platform is a best-in-class molecule really drives home concepts around the inflammation. And then it's just a stepping stone as we think about our next set of multi-inhibitors with Kodiak. If you think about it for a small company, there's a lot of discussion and value discussions of course around KSI-301. But I think it's important to look at Kodiak with our 10 years of history and our ability to execute technically in retina and take a look in an objective manner at the level of expertise in the pipeline, right? So 501 is a very, I think, like I said, best-in-class, first-in-class potential, showcasing the ability of bringing multiple mechanisms. And [indiscernible] will be able to talk more about the conjugates where we bring hundreds of small molecules into the biopolymer together with biologics to think about a whole new set of next-generation retinal medicines that really showcases our unique franchise in retina.

Great. Victor, thanks for being here. I appreciate the time.

Thanks a lot, Matthew. Appreciate it.