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

Okay. Thank you, everybody. And we apologize for the delay. We're going to get started because we have a lot of material to go through. We'll start by saying at Kodiak, we had clinical failures with our tarcocimab investigational therapy. We had 3 successful studies and 3 failed studies. We've reviewed our data. We've gleaned insights and translated these into thoughtful and important changes for tarcocimab. The reason for the clinical failures had nothing to do with tarcocimab's durability, but had to do with other factors: the immediacy of clinical effect; the trial designs; the gel-like nature of the old formulation. And we believe that we've addressed these factors. Durability is still important and tarcocimab is the only biologic medicine designed for durability from the beginning and with an underlying science of durability. Today, Dr. Brown will talk through the science of durability, where he has been a thought leader for many years from the beginning, and he will also disclose some exciting new data on tarcocimab's durability in humans. Dr. Brown will then talk through the important change that we've made to the go-to-market formulation for tarcocimab and the data supporting this course correction and its expected impact. There are also important learnings from our clinical trials and their designs and execution. Dr. Charlie Wyckoff will talk us through the 3 molecules in our clinical pipeline, and will talk through in some detail the designs and the rationales for their ongoing and planned clinical trials. Importantly, our course corrections, they represent platform-wide changes and are now reflected both in tarcocimab and in KSI-501, our bispecific conjugate. With the 6-plus years of clinical and more than 15-plus years of nonclinical and manufacturing experiences at Kodiak, we understand both of our ABC molecules much better now, and we believe deeply in the changes that we've made. In this past year, as we mapped out our future, we didn't stop there with those 2 molecules. We've added a third exciting clinical molecule into our pipeline, KSI-101, a bispecific antibody that's outside of our ABC Platform, and represents a different or greenfield opportunity and brings a nice diversification to our clinical pipeline. As a committed R&D company, an important part of today is to present to you where we are as a company and the science as we see it, and to allow all of you to internalize that and come back to us with questions and comments. So with that quick pre intro, let's get started. First of all, the presentation today and the discussion will include a variety of forward-looking statements, and it's important to understand those, that we have a lot of risks that we're engaged in as a drug development company. Please take the forward-looking statements seriously. Many companies fail for not testing the hypothesis, you know, for poor execution of the plan. At Kodiak, with the large number of studies that we've run, and the large number of patients we've run through our clinical trials, and the significant level of manufacturing that we've performed, we hope we're not in the bucket of not having properly tested the hypothesis, which is an ongoing battle. But at Kodiak, we have a tremendously talented team, and technical leadership and ownership is in-house across the board, enabling timely and cost-effective execution, and here are the members of our broader leadership team. Today, we have members of Kodiak management who are in attendance and will be part of some of the roundtable discussions and some of the audience Q&A, including myself, Victor Perlroth, the CEO of Kodiak; John Borgeson, our Chief Financial Officer; Dolly Chang, who recently joined Kodiak from Genentech Roche, worked on the late-phase pipeline on Vabysmo, and also worked on the early stage and had responsibility for the early stage Genentech ophthalmology piping through early clinical studies. We're glad to have Dolly join Kodiak and have her discuss Kodiak's earlier pipeline towards the end of the day. Almas Qudrat is our Head of Quality. We have a substantive quality group, and it oversees, among other things, the Ursus commercial facility that we built and partnered together with Lonza, which is Swissmedic-approved and operational and was used to manufacture the enhanced formulation of tarcocimab, which is now in several pivotal clinical trials. And we also have Dr. Velasquez-Martin, Pablo, who brings both an artisanal approach, which has been critical in our clinical trials, and important. He brings the transparency that we need to be able to run so many pivotal studies in parallel at Kodiak with a small team. So we're glad to have the Kodiak management attendees. Moving to our 2 special guests, both important for their clinical leadership and also their scientific leadership. So first of all, with Dave Brown, a young overachiever back in the day at Baylor and then to Iowa for his ophthalmology training, Dave goes back to the early days of Lucentis and Eylea, pushing durability from the very beginning with Lucentis in the HARBOR study and more recent work on Vabysmo and Eylea. He also is a business leader in the field of ophthalmology, working closely with the Retina Consultants of America, or RCA, as they're called by all of within the business Charlie or Dr. Wykoff, also a serial underachiever. His undergrad at MIT as Phi Beta Kappa, a PhD at Oxford on the Marshall scholarship, an MD at Harvard. His ophthalmology residency at [ Brigham ] (sic) [ Bascom ] Palmer, where I think he had a stint as chief resident. More recently was on and is on the ASRS, American Society of Retinal Specialists Boards of Directors with a lot of impact on safety decisions in the community, and he's very active in translational research and clinical trial design. It's a pleasure to have both of them bringing their own perspectives and to talk about Kodiak Science and our progress over time, and to put that in perspective of the broader field. So what about our agenda for today? As I mentioned, Dr. Brown will talk through the science of durability of our medicines and then talk about the science of our enhanced formulation. Those 2 important science updates are critical for people to understand where we are today with Kodiak. We'll have a little bit of a roundtable. From there, Dr. Wykoff will talk about Kodiak's late-phase clinical program, our 2 ABC medicines and our new KSI-101 product, and also talk about the pivotal clinical trial designs we have for those 3 molecules. I'll talk briefly about the commercial opportunity, or the Kodiak opportunity, with a bit of a roundtable discussion. Dr. Chang from our team, will talk through our [ newer pipeline ] medicines using what we call our ABCD Platform, where we bring small molecule and peptide drugs into the biopolymer. And we'll have some summary and takeaways with an audience Q&A. It's an R&D-heavy kind of agenda, but we think that's relevant for Kodiak today. So why Kodiak? Well, we believe that today, Kodiak is primed for near-term and long-term success. I think the core element of that is the agile R&D mindset that we brought to looking deeply into the data, starting with the negative surprise of the DAZZLE Phase IIb wet AMD study in 2022 and then following the progress of Kodiak's data through our broader pivotal studies, and more recently, the GLEAM and GLIMMER studies. We've made important course corrections in our molecules and in our studies, and we've implemented those very rapidly in our new pipeline. From an execution standpoint, we have a track record of running a lot of studies and a lot of patients with a high level of care and attention. As I mentioned, the key element is to have technical leadership and ownership in-house in the company, which allows us to execute in a timely and very cost-effective manner. We have a diversified late-stage pipeline with 3 shots on goal, each in a BLA-facing development plan with filings as early as, say, mid-2026. Tarcocimab is 90% of the way completed both from a spend, let's say, and standpoint. The clinical and CMC costs have already been incurred. An the enhanced formulation is designed to deliver both the pulse and the durability in the patient. And we believe that there's still a very poised commercial opportunity in the marketplace for our ABC medicine support for tarcocimab and also for KSI-501, which is our bispecific conjugate, and brings the potential for both greater efficacy and the signature durability of the ABC Platform. And KSI-101 represents a greenfield commercial market, the ability for near-term data starting in Q1 of next year. And as an uncorrelated asset to the ABC Platform is important as part of our broader diversification and is a fast follow dual MOA possibility with very high strength. And as a company, it's important to remember that we have tremendous independence. We have the flexibility to make the right choice for each Kodiak stakeholder at each moment. We own global commercial rights to all of our molecules. We built to complete it in approved high-volume commercial manufacturing facilities for Kodiak's ABC medicines. And we have the freedom and flexibility on how best to commercialize our molecules if approved. And our cash runway is expected to support operations into 2026. We try to be a learning organization, and through our journey, we've gathered key insights and transform these learnings into actions. Starting with our Phase Ib study, where we dosed patients initially in January of 2019, what we saw was a well-tolerated profile with a 6-month predominant durability profile and we moved quickly to put tarcocimab into pivotal studies. What we saw in our DAZZLE study was strong durability, but we also saw the first sign of immediacy deficit in wet AMD in terms of the overall activity through the loading phase. And we also had an overly aggressive study design, which at that time was maybe asked for and appropriate but was really not needed. Our BEACON study in RVO reinforced our durability, and in GLEAM and GLIMMER studies, although we saw a 6-month predominant durability profile and strong immediacy, an increased cataract rate specific to DME was noted. The GLOW1 study generated tremendously strong data with all patients on 6-month dosing. The Ursus facility is now approved, as I mentioned, for commercial manufacturing, and it's already working and has made the enhanced formulation which was released in November of 2023. We decided to expand our pipeline in February of this year, adding KSI-501 with its durability and the potential for greater efficacy. And we also added KSI-101 to the pipeline at that time. The GLOW2, DAYBREAK and APEX studies are now enrolling and just more recently in the last month, we finalized some analysis of our human ocular PK data supporting our signature durability profile. So a number of very positive findings. Tarcocimab and the platform are well tolerated. A differentiated 6-month durability is real. And ocular PK data supporting that durability. So it's a true platform, but we also identified really critical and important issues that we decided we couldn't live with and we needed to fix: the immediacy deficit in wet AMD; the increased cataract rate specific to DME; and taking overly aggressive bets on our clinical study designs. So the actions were to build an enhanced formulation to course correct issues identified both in wet AMD and DME; new study designs educated from prior studies anticipated to have a higher probability of success; and to diversify the portfolio into KSI-101. We thought we would address at the beginning 4 questions that we've gotten from investors. And we would tackle those head on before we got started. So why are we running another wet AMD durability study for tarcocimab given the failure of the first study? Well, why did DAZZLE not meet the primary endpoint? The first is that we undertreated a minority of patients forcing all of them up to Q12-week dosing, which was not tolerated by that small number of patients. And we compounded that by having a weaker immediacy of our drug through the loading phase, which we hadn't known before the study. And the combination of those led us to fail the noninferiority endpoint. So how do we solve the undertreatment? Irrespective of the drug tested, some patients require frequent and monthly doses. So in DAYBREAK, our new wet AMD study, we had a fourth loading dose. We allowed dosing as frequently as monthly. We do proactive or what I call, preactive treatment until dryness. And we detect disease reactivations earlier using new tools and new approaches that Dr. Wykoff will discuss. And solving the weakness in immediacy, the problem was the slow loading of the high molecular weight conjugate in some patients and Dr. Brown will discuss that. The solution is a basal-bolus type enhanced formulation combining conjugates as well as unconjugated free protein. In the free protein portion alone, for tarcocimab, has the VEGF binding equivalent to 1.3 milligrams of aflibercept plus the full power of our conjugate. And therefore, we continue to bring the uncompromising durability. What about question 2 here. Will the cataract imbalance seen in the GLEAM and GLIMMER studies impact the success of tarcocimab and KSI-501? So why did the GLEAM and GLIMMER studies not meet the primary endpoint? And the reason is an unforeseen increase in the onset of cataract adverse events. So this was specific to patients with advanced diabetic eye disease. So the problem is the mechanical insult to the lens capsule, which is known to be more fragile in DME patients who have swollen lenses during hyperglycemic episodes due to the gel-like consistency of the tarcocimab old formulation and the high force needed to inject it and the prolonged time to do so. And notably, what's important is that the cataract rate in wet AMD patients on monthly tarcocimab was lower in the DAYLIGHT study than it was with aflibercept, the comparator. So the solution is the enhanced formulation of tarcocimab with less biopolymer amount significantly reduces the injection time and force, which decreases, or eliminates, any potential insult or issue to the lens capsule in DME patients. So we believe that the enhanced formulation would be expected to bring the cataract rate back down to the levels associated with all biologic if we were to run another DME study. Question 3, what's the commercial opportunity for tarcocimab in 501 in an increasingly competitive landscape? And we'll spend a little bit of time here because I think it's an important overview. So is there space for new treatment in retinal vascular diseases? So intravitreal anti-VEGF biologics are the mainstay of therapy for patients. And despite limited differentiation, Gen 1.5 biologics, Vabysmo and Eylea HD, they have significant market uptake. They have high efficacy, but limited true durability, and our guests today will be able to discuss that together with us. There remains a high interest and a high need for a mainstay biologic with high efficacy and high durability. So tarcocimab and 501 have the potential to be meaningful products in this marketplace. So the science of our platform and of our enhanced formulation delivers immediacy and durability. Tarcocimab and 501 are being developed as mainstay biologic monotherapies for all patients, treatment naïve, treatment experience, mild patients and severe patients. And pivotal studies support a wide range of dosing from monthly to 6-month dosing. And if successful, the initial BLA package for tarcocimab could provide approval for 3 indications at once, diabetic retinopathy, retinal vein occlusion and wet AMD. And we have complete commercial flexibility in our approach to how we would commercialize these medicines. So with a differentiated clinical profile, along with creative and thoughtful commercial strategies in a $14 billion plus and growing marketplace, tarcocimab and 501 have a meaningful commercial opportunity if approved. So what do we mean when we say a mainstay biologic? Well, tarcocimab and 501 are being developed through our development plan as mainstay biologics, monotherapies that provide high efficacy and durability in a flexible 1-month through 6-months label. So if you look at Eylea and Lucentis, or Eylea HD or Vabysmo, that's what the current industry is set to manufacture, to approve and to provide to patients. Initially, Elyea and Lucentis as Gen 1, and Vabysmo and Elyea HD now, but the jury has come back and we understand in the clinic and for our retina physicians how those agents are performed. I think they don't show true durability necessarily. They may show some mild improvement, but they don't definitively solve the problem where you want to have high efficacy and high durability. And that's really where Kodiak is developing towards with both tarcocimab and 501. They are intended to follow and be Gen 2.0 agents that follow directly from the biologics and as a mainstay biologic. Whereas other novel therapies that are, say, earlier in development than where Kodiak is, the TKIs and the depots or whether it's gene therapies, those are largely alternative therapies or adjunct therapies for different subsets of patients after evaluating the mainstay intravitreal biologics. So maintenance agents to be used in a subset of patients after intravitreal biologic monotherapies have established disease control. Question 4. Will our cash runway last through top line read outs of ongoing clinical trials? Well, our cash runway is expected to support operations into 2026. We believe we have increased our probability of success utilizing our historical learnings and corrections. So in tarcocimab, the GLOW2 has an additional dose compared to the unequivocal GLOW1 data that was very strong. DAYBREAK optimizes treatment for each patient and incorporates our enhanced formulations. And these address our 2 key insights of undertreatment and immediacy deficit. So with tarcocimab, we're running that in DAYBREAK as a durability arm and in 501, we actually have a different objective and are running 501 to explore optimal efficacy with intensive dosing. And in KSI-101, we bring an uncorrelated asset into the pipeline with an early 16-week primary endpoint against sham. So we believe and anticipate a high probability of successful in these studies, with readouts, let's say, for GLOW2 in early 2026 and DAYBREAK, in the first half of 2026. And KSI-101 beginning to generate data, open label starting in Q1 and Q2 of next year and dosing pivotal studies through 2026. Our studies are run fully in-house without a CRO, increasing efficiencies and reducing cash requirements. And we own commercial rights to all our molecules, both ex U.S. and U.S. So we can decide to act on the option to partner and retina is an area of high partnering interest. So we anticipate approximately $200 million in cash and equivalents at the end of the third quarter of 2024, just in the next couple of weeks. And importantly, we remain committed to raising capital at the right times and in the right ways to enable cash runway through critical milestones while maximally preserving value for existing shareholders. An important summary of our clinical programs and timelines is included. So we're actively enrolling, for tarcocimab, the GLOW2 study. We're actively enrolling, for tarcocimab and 501, in the DAYBREAK study. So we hope to complete enrollment for the GLOW2 study by the end of this year or early next year. And for the DAYBREAK study, we hope to complete enrollment towards the end of Q1 or early Q2 in 2025. So for these 48-week primary endpoints, that drives the timing for top line data readouts, okay, into early Q1, let's say for GLOW2 and we'll say mid-Q2 in 2026 for DAYBREAK. Meanwhile in the KSI-101, we have the APEX Phase I open-label study. That's similar to the Roche DOVETAIL study on their anti-IL-6 and we'll be sharing open-label data on different cohorts and different dose level cohorts in patients with inflammation. And we plan to begin enrollment in our PEAK and PINNACLE pivotals in early 2025. So in summary, with 3 clinical programs leveraging Kodiak's 15-year history, Kodiak is at a decisive moment and represents an exciting investment opportunity. So tarcocimab and 501 is our conjugates with upcoming milestones coming out of the GLOW2 and DAYBREAK studies. And for KSI-101, there's an interesting novel separate greenfield opportunity, our Phase Ib APEX study clinical data will start to be disclosed in early -- in the first half of 2025, with Phase III initiation similar time frame. So what is our vision for Kodiak in 2026? Top line data readouts for KSI-101 from twin pivotal studies in inflammation with the BLA in preparation. Tarcocimab, a BLA filing in diabetic retinopathy, RVO and wet AMD supported by 5 successful studies, including GLOW2 and DAYBREAK if successful. For KSI-501, where we want to be exploring the potential for better efficacy than existing agents, we'll be 1 study away from registration. And we expect between now at the end of this quarter, we expect approximately $200 million in cash to support our operations into 2026. And very quickly, and we'll talk a little bit more at the end, the concept of continuing to further our ABC Platform and evolution that allows us to bring drug cargo into biopolymer and where you can have the antibody or not. But we've shown that the biopolymer alone has the same exit profile and the same half-life as our conjugates, so bringing diverse APIs like peptides and small molecules into the polymer, having a drug antibody ratio of up to 100 or 200 as needed, we've shown we can do that, we can release those on command and it leverages the proven safety record of the ABC platform. So a new combination of targeting high drug loading, mixed API formats and tailored drug release is real with application to ophthalmic and systemic disease. And we believe, based on the fact that it really leverages all of the knowledge and expertise and capabilities, resources and even commercial manufacturing capabilities of our existing ABC Platform, we can remove those new programs that Dr. Chang will talk about quickly into the clinic. On that note, with that introduction, we're going to bring Dr. Brown up here and he's again, going to begin to walk us through some exciting science of durability.

Thank you very much. Charlie and I are happy to be here. Unlike a lot of my colleagues around the country that do Investor Days, Charlie and I are not monkey to wind up and you hear what the company tells you. So we're both very candid for better, for worse. I'm going to start with some data that I presented at Retina Society and some initial data before that at Macular Society that I presented to Kodiak and I said, "You need to do this because this may prove why you actually ought to continue on this platform." And this talk is talking about -- I was asked initially to give a talk for the Royal College of Surgeons of England as the keynote speaker on why is 12 weeks the new norm. And so the problem is, spoiler alert, for most people, 12 weeks isn't the new norm. And both clinicians and the investment community, thanks to -- this drug's a 12-week or 16-week drug, but in the clinic it's not and why is that? So MARINA ANCHOR, this is the paper I wrote 2006, seems like a long time ago, that started the anti-VEGFs with ranibizumab. And you see if you give an effective anti-VEGF consistently and constantly, you have a vision gain that maintains that gain. If you look at a lot of our trials over the last 3 or 4 years, you have that gain and then it tapers off because we undertreat. And it's all an attempt by marketing to say that I've got a 12- or 14-week drug et cetera. So when we tried to use ranibizumab quarterly, this is the PIER trial, and what it shows is 3 doses and then you lose all the gain on average over the course of 12 months. So the interesting thing is there were some patients that did okay. And even on quarterly dosing with ranibizumab. And I went to Genentech and I said, "Can the reading center tell us if you maintain the dry retina with quarterly dosing, did you maintain that vision gain?" And the reading center couldn't do that. I had written this paper with Carl Regillo, which find the few things that we look at a lot today, subretinal fluid, intraretinal fluid, sub-RPE fluid. And I get the original pictures from PIER, and we had 2 images, and I read them all in a mass fashion, 6,000 images, I'm crazy. And showed that if you maintain the totally dry retina, in other words, no intraretinal fluid, no subretinal fluid, not 100 microns, not losing vision, not all the stuff that we don't do clinically, you maintain vision with that orange curve versus if you didn't control you on quarterly, you lost that vision. So this is the first time we showed that absence of OCT fluid correlates with vision gain. But for the purpose of this discussion, the most interesting thing to me is what's different? Why do 1/3 of the patients in PIER do okay with quarterly dosing with the drug that we know the average patient can't get by with? And I'm going to convince you that one of the major components is drug clearance. We know that axial eye length, pseudophakia, vitreous syneresis, that means more liquification of the gel, all those things increase drug clearance. We thought, but what improvement until this that anti-drug antibodies probably increase drug clearance. This is public knowledge from the FDA filing data of faricimab showing that if you have any drug antibodies, your need to go to a more frequent dosing Q8 is statistically significant, 0.004. What do we know about the half-life of drugs? The most one, the one we've given 80 million doses, aflibercept, we don't know that much. All we know is a 5-patient study. Let me repeat that. 5 patients is the only thing out there on drug clearance of aflibercept before some of the data I presented at ASRS. And what it shows that the mean, somewhere around 10, 11 days median, but the interesting thing is if you look at the 5 patients, 4 of them were along that curve, but one of them had a half-life of 3.7 days. This is a published paper by Diana Do, showing that some people really clear drug fast. What do we know about clearance in other drugs? This is the best data we have. And it's in the Genentech filing data, it's public knowledge, FDA.org -- .gov. And you see that the average half-life of faricimab is a little less than 8 days, but here's the most interesting part. 10% have a half-life of less than 5 days and 10% have a half-life of over 11 days. And does that correlate with increased dosing? They did the work for me. They actually show the logistic progression of probability of needing more frequent dosing, like if you clear a drug faster, your statistical needing more frequent dosing is 0.0005. This is in 628 patients, not 5. The FDA made them do a lot after brolucizumab, and Roche wanted to do it and they published the data. So this is requiring Q8 or Q12. Again, if you clear drug fast, you need more frequent dosing. This is in diabetes, you see the same kind of curves, interestingly a little bit less frequent half-life. That's probably due my constellation and maybe break down the blood retinal barrier but the frequency of needing more frequent dosing is directly correlated. So how does this relate to the waiting room? My waiting room. These are all angry patients waiting for their injection. And all our trials are treatment-naïve including all the Kodiak trials. So in a treatment-naïve population, there is a normal distribution, where you got average clearance in the middle and then you have lucky people like this guy. And he's a slow clearance guy. So he is the type that even with ranibizumab can probably go 10, 11 weeks. You know with aflibercept that's that amount of number. Where is he in my clinic? He's not there very often. And he's only coming in every 3 months. Then you have this part of the bell curve, if he's not there very much, really goes away. So our clinic and the actual drug volume of sales is really skewed to those patients that need drug all the time. Most -- a lot of my clinic is guys like this guy. This is a Muppet and he's saying, why me? Why do I need dosing all the time? And he is the type that has fast drug clearance. So this part of the bell curve and where is he in my clinic, he's their every month. So there's 3x as many of those as the slow clearance guys. So all my clinic is this guy. So this is anti-VEGF molar equivalency. In other words, if you take aflibercept 2-milligram as one, then ranibizumab is about 0.5 to 0.6. Faricimab should be 2.3. And by definition, aflibercept's 4.0, right, because it's 8-milligram, take half-life, go down to 4 milligrams, another half-life, you're down to 2 milligrams. That should give you 2 extra half-lives. So if people come in with a distribution, my answer -- or my question, is 12 weeks the new norm? If your half-life is 10 days, and you double it, you get 21 extra days or 3 weeks longer. If you're this lucky guy and your half-life's 14 days, you double it, and you get a month longer. But 1/2 my clinic is these guys. They've got a half-life of 5 days. You double it, he's only getting 10 days longer. So is 12 weeks of the new norm? For the average person, maybe. For lucky slow clearance side, he even may go longer. But this guy, the guy that's 1/2 my clinic, 5 weeks is his new norm. Even with the fancy new faricimab an 8-milligram Eylea. But it's better than monthly. So what I'm telling you is if you have a drug that actually lasts longer in the average patient or in any patient, it's incrementally better. And if you look at it, aflibercept's incrementally better than ranibizumab and it's a $10 billion drug, right? It's not that much better, but it's better. So the question is, can Kodiak do something to make their drug where you get the same curves that maintain but go longer between dosing. So currently available options, you increase dosing frequency; block additional targets; molar blockade. And with the ABC Platform, theoretically, we should be increasing drug half-life. Does that pan out? So what I'm going to show you is, is it needed? It's definitely needed. Half my clinic is needing more frequent dosing than Q7 or 6 with faricimab with 8 milligram. They showed in their trials and the problem with the trials was it started with brolucizumab with Beovu. Beovu, the agency lets you do this trickery where you had 1 group that you treated 3 doses in Q8. And the other one, you had to have so much fluid or so much vision loss and in real life, I would never let your mom have increased fluid or have loss of vision before I retreat. And so it's like a horse race where you handicap one group, but then you'll let the other run fast, right? And so we're definitely -- you're going to show that you have less doses because that's just built into all the trials. So with that, and you know the trials, but they showed 5- to 6-month durability with their dosing criteria. But it's hard to correlate that to the other drug because it wasn't a fair horse race. So the conjugate design should be with a bigger molecule, should make a slower half-life. This is the data from Genentech, and this is published, where they took ranibizumab and they made ranibizumab bigger, different formulations and they showed that the half-life correlates with the size of the protein in a linear fashion. This is published in transaction (sic) [ translational ] vision science & technology. So this is a big boy. This is sort of a sumo wrestler of our drugs. This has -- instead of being 140 kilodaltons, this guy's 950, ranibizumab if you remember, 48 kilodaltons. Brolucizumab was 26 kilodaltons, really, really small, that's why they go pack so much in. And if you look at molar equivalency, here in this big molecule, you still have pretty close to the molar blockade of aflibercept HD, 3.5 versus 4. So what about potency? If you look at binding affinity, and again, they gave me not much time to go over 8,000 slides. So this is all online, and we can talk about it more. But basically, if you look at a inhibition curve, tarcocimab really close to aflibercept. Actually, may last a little longer down at the end of the curve. The IC50 is almost right at the same. This is the core thing, ocular animal half-life, and that's one thing where you can test. You can get rabbits, you can put all these drugs in. You see that the distribution of tarcocimab goes to the retina. There was some argument that big molecules couldn't get there. This is interesting. Their ocular half-life is 11 days in rabbits versus about 4 days for aflibercept and faricimab and about 3 days for ranibizumab. The question is how does that correlate to all that human data I showed you with faricimab? And this is the cool new data that they did, and I'm pleased to show it to you. So in the Phase Ib study, which was the longest, biggest Ib in the history of retina that I've ever heard of, they actually had patients, and 47 patients that as they were needing redosing, they got AC taps every time they are redosing. So in a lot of these patients, there's 4 or 5 doses that you can get a good dose response curve. You really only need 2 but if you have 4 or 5, you can really see if it's linear. And this is all the patients here on the right, and it shows you, just like the faricimab data, that there are some patients that clear tarcocimab faster than others. But in these points, the select individual patients, you'll see that it's a pretty nice linear distribution. Like in those patients, you're really seeing a nice linear distribution to tell half-life. So of their ocular distribution, 45% had a half-life over 20 days. That's bigger than anything you've seen in any of the stuff on faricimab and I presented in a poster at ASRS, that's bigger than anything we've seen with aflibercept. 36% are 11 to 20 days, and 19% are less than or equal to 11 days. So this is superimposing the data of what I showed you from faricimab and then tarcocimab, showing you that on average, it's got a lot longer half-life. So it really should last longer if it has the strength to control the disease. This is in that Ib. They had AMD, RVO. Here you see that distribution varies in all the diseases. But if you look at the half-life and most of the mean-medians that sort of -- it's certainly greater than 11 days, which only 10% of faricimab had. So here you have a human half-life and according to the others. And so you see that it's way bigger and it goes along with that molecular weight, and that's probably the reason. So the durability, like I said, it's going to get tested in a more head-to-head trial that Charlie is going to talk about, DAYBREAK. If you look at the DME didn't meet their endpoint, that's the visual acuity. Victor mentioned this, if you look at pseudophakic patients, the visual acuity on that top curve looks the same. Here you see an AMD that didn't meet their endpoint. If you had those that could go 5 months, it did. And again, you're getting rid of 1/3 of the patients that couldn't go that long because they sort of went along the pathway of brolucizumab trials and the faricimab and HD trials where you try to make it look like that everybody is 16 weeks. I've shown you before, that's not real in the clinic, like we don't have patients -- I don't have anybody in the clinic with either faricimab or 8 milligrams that can go 14 and not 16. In other words, if you don't tolerate any fluid, no one can go that long. This is their diabetic retinopathy. This is actually some of the best data they have. They show a 29% (sic) [ 29x ] increase over sham which is the approvable target for 2-step retinopathy prevention or improvement and then a reduction in the risk of 89% of sight-threatening complications. In RVO, it's probably their best evidence of durability. This is not my favorite graph to show it. I think maybe in Charlie's stack. But basically, you had considerable number of patients in the evenly matched last 6 months that needed less dosing with tarcocimab than you did aflibercept. It's about 30% plus. So what does this mean? If it's really more durable, one, is there a problem, first of all, with their trials on the loading dose? Like, on the loading dose, it didn't look like it dried as good, and that's a problem because you need it loaded. And that's why they came up with this -- that immediacy gap's what they call it, not being as strong as the reference product, which is unconjugated. And that's why the new formulation is kind of like a sustained release where you have a load -- a bolus of instantly available drug plus your tarcocimab, which is released at the rate of the biopolymer. So the enhanced formulation we're going to talk about. Basically, it's taking 80% conjugated, 20% unconjugated. Victor talked about 4 plus 1, it should be the same amount of active drug but you're given one that should be enough to control the initial curve while your other drugs are building up. Here, you look at -- go back to those molar equivalencies, that tarcocimab 1 milligram should be about the same as 1.3 milligrams of aflibercept, which certainly should control AMD. It's actually equivalent to more than a ranibizumab dose. In my first slides I showed you, MARINA ANCHOR where ranibizumab 0.5, I think we would all take that efficacy if it was our eyes or our patients'. How much unconjugated protein is it? It sure looks like it should be for AMD. They're not doing a DME study, so jury's out on that. But for AMD, it's really bioavailable, it should be enough to control the disease. And like any other trial, you don't know 'til we run the trial, and that's why they're spending the money to do it. Here you see the -- you got to show contribution of components to the agency. The unconjugated protein should give you a pulse that should help you bridge that gap that we saw in the first loading doses of RVO in AMD. And closing that gap is what needs to happen because even if it lasts a long time, if it doesn't dry out the retinas, it's not commercially viable. The potency looks the same. The lines are on top. If you're looking at VEGF findings in preclinical assays, it's the cell-based assay. And bringing it all together, I bring my same people here. The lucky guy is going to be fine with having an immediate bolus of the enhanced formulation, but then should have quite a long time of extended dosing. The frequent flyer, as we call it, the ones that need lots of injections, he's not going to go 5 months, right? Here he is, we got him out 60 days, but if you can get any of those out to 8 weeks, that's a commercially viable product. It's actually a commercial winner. Like, if I can take my 4- and 5-weekers and actually get them to 7 or 8 weeks, there's -- that will -- patients will demand it, doctors will demand it. Will this product do it? We'll see. We got to run the trials. And Charlie is going to talk about that. Are you sacrificing durability? You should have quite a bit of conjugated drug. You should -- the 4 milligrams should do it, again, it's sort of a compromise, and I think it's a reasonable scientific-based guess at what they should do, and I agree with this. What about the cataract? Is it going to fix it? I don't know. I mean, it's really kind of hard to figure out why the cataracts are there in the first place and I'll show you this. If you look at DAYLIGHT, there was actually statistically a little more cataracts in the wet AMD with aflibercept than there was with tarcocimab. Is it something just with a diabetic lens? It may be. Diabetic lenses swell. And every time you're hyperglycemic, your lens swells and you get a hyperopic shift. And every time you get hypoglycemic, it shrinks. And it may be that, that lens capsule, because it's having to change so much, something changes in it in the diabetic eye. We don't know. This is porcine, pig lenses. And if you stress on what's a lot of stuff, almost anything gives a cataract, but particularly glucose, hydrogen peroxide. Certainly, if you microwave it, that's probably not a great thing to do your crystalline lens. And so what they did is they looked at these pig eyes and they put the untreated eyes and then the old clinical formulation and then the enhanced formulation and they're all the same. Like, the drug itself doesn't seem to cause cataracts more than the untreated with the original formulation. Peroxide certainly does, microwave certainly does. So if you look at it -- and this is a little tough slide to understand what it's showing, but what this is, is what if you combine a hyper glucose situation with their drug, could it be the combination of the 2? And what you're looking at in this second panel here, the green is oxidative stress damages. Untreated versus old clinical formulation versus enhanced formulation. The second column is all of high glucose. You're not seeing any difference. So it's not really the combination of the drug and the glucose. It must be something to do with the lenses itself. One hypothesis is the diabetic population is younger, so they have a more formed vitreous gel. If you have more formed vitreous gel, it may be that you're -- if your drug is more of a big gel, it may sit around on the back of the lens capsule and cause more problems. There's a secondary effect of doing this enhanced formulation where you basically dilute it with the unconjugated drug, that makes it way easier to inject and it makes it where it disperses easier. So it's possible this will decrease that problem in diabetics. They're not doing another diabetic trial. So until the drug gets out there, we won't definitively be able to tell this. But it's reasonable to think that it may help and you don't really know until you do it. So the solution is -- this is the new drug versus the legacy drug. The new drug is injected in 2 to 3 seconds versus 8 to 10 seconds. I've given a billion injections and 10 seconds in the eye is a long time. And you can see where maybe some of that was lens damage, maybe the needle hit the lens, although they were really good investigators in this trial. I can tell you, from a commercial perspective, now that we have SYFOVRE, we're used to really viscous drugs. And so we're better at injecting viscous stuff. But still, commercially, people like drugs that don't take as long, and is not as hard to inject. So if this drug comes up commercially, it will be better to have this formulation. The injection force is reduced by 1/2. This is a model just showing you the old formulation versus the new formulation. To be honest, this is more like an old vitreous because old vitreous is more liquified. Young vitreous is more like Jello. So in a diabetic population, you may get even more pooling than what you're going to see here at the top of the eye, by the lens, than you do in this. So this is the old formulation and the new formulation. And you see, the new one is already done while you're still squirting in and pushing on the old clinical formulation. And so that dispersion should help potentially with the cataracts, but also with commercial acceptance if we get that far. So this same formulation is in all the new medicines he's telling you about, the combination with the polymer and the VEGF trap that's got IL-6 and VEGF blockade. So that's in the same principles of decreased injection and more disbursement equally applies to KSI-501 as it does tarcocimab. Same VEGF, same anti-IL-6 potency as the preclinical assays and less time and force, which is always a good thing for those of us giving 8 million injections.

We're going to have a roundtable discussion now and -- we're good? Open mic, you can ask about anything. Trump, Harris and the British parliament. We're good. Yes.

Thank you, Gena.

Gena Wang from Barclays. Well, thank you very much, Dr. Brown. That's -- and also Victor. That's super helpful, very comprehensive overview. So I do have a few questions, i.e., you brought up -- one is the longer half life is so important for durability. But we do see the initial loading dose cannot control well. I think it could be slightly different mechanisms. Maybe if you can elaborate a little bit and if the initial loading dose control cannot be as good, just simply have a longer half-half, if we compare to say faricimab, would that be overcome? Because then you have a slightly maybe different durability profile. Just we cannot just say, looking at the half-life if one drug control a bit better than the other. So the half-life maybe is not the same comparison. You have a different factor we need to take into consideration. So the second question is we talk a lot about the half-life about antibody. What about the biopolymer? Could that be -- I know we talk about viscosity in terms of injection, but what about biopolymer, the half-life in the eye? Was over time the accumulation of biopolymer, would that also would be the underlying cause of, say, safety concern there? the would that be any talk about this cost for the injection, but what about from biopolymer's half-life in the eyes? I mean, over time, the accumulation of the biopolymer, would that also be an underlying cause or a safety concern?

So in terms of the half-life, the unconjugated part should have a half-life. There's no data on this, but that should have. It's the same size as another antibody. It should have a half-life pretty similar to what you see with faricimab or Eylea. They're about the same size, that unconjugated part. Your hope that you control that initial disease activity, and then you have enough ongoing VEGF suppression to keep it. It looks like in their RVO data, at least in that second 6 months, they're able -- with the same head-to-head retreatment criteria, they had less doses than aflibercept. So I think once they get loaded up with the conjugated, you're okay. Certainly, Victor and Pablo has spent a lot of time thinking about this, and I'll let them answer. Your thoughts?

Yes, I think the data that we see, we believe that the conjugate just takes a little bit of time to get to where it needs to go and to be effective. And so now any time we give a dose, we're always going to have the 1 plus 4. So you're always going to have the 1 of the unconjugated that's going to essentially bridge us to where the conjugate then has had time to get to where it needs to go and to do its thing. And so in any situation, you're going to be, essentially, better off.

Yes. The way I think about that question, it's a good question, is to separate, just kind of dive in the weeds a little bit more about the drying capacity and then the durability capacity and those can be separated. So for example, if you look at TKIs out there, they're in development. I hope they all get approved. I'm very supportive of those programs. But we've learned through most of those programs, that they don't have a really strong drying capacity. And that's why you see the trial designs as you do in all of those programs. They're selecting a subpopulation of wet AMD indication that are important, but they're not the high-need patients because you may not have that immediacy effect with TKIs. But there may be a huge durability benefit even among those patients that don't need that immediacy effect. And so from my perspective, I certainly hope that this 4 plus 1 or 1 plus 4 dynamic can improve that efficacy. But worst-case scenario, even if it doesn't, then this need [indiscernible], there's still potential value because I do believe the PK data and I see that durability as a huge value add at the base. So if you can have both, that's the optimal outcome. But I think it's important to be able to look at those factors independently.

In terms of the safety, this molecule seems pretty safe. I mean whenever you look at a clinical trial, you always see 1%, 2% inflammation rates because we're looking at the anterior chamber every time. And we have different investigators looking. And we all call things different, like trace cell versus 1 plus cell. Like there's no -- a retina guy will call it nothing and then my uveitis partner will go, "Oh, there's 1 plus cell there" or whatever. Really, it's not important unless you think it's clinically significant enough to give prednisone or to give treatment for it. And so that rate really was no different than the aflibercept, right? Correct?

Yes. I think --

You know the data better than me.

Yes, I think maybe our rates might have been 1 point north, for example, but more similar to faricimab. Honestly, across the studies, I think our blended rate is 3 something.

Yes, it's close to 3 and just similar to the blended rate of faricimab in their first studies.

Yes. When I think about safety, I'm glad these questions keep coming up. There was a question earlier about safety and then you're bringing up safety. To me, to see a monthly dose trial through a year, DAYLIGHT, I think it was called, I get the names confused, but to see monthly dosing for a year in wet AMD population and not see a difference in the cataract rate and to have this consistent inflammatory rate is quite meaningful to me. Because the thing I was hearing with your question is, is there a build up of the polymer and is that going to change the safety profile over time. I think we've, I mean as definitively as you can in Phase III programs, answer that with monthly dosing consistently in a Phase III program with no differentiation. But we're smart as a field. I'm glad you're looking at this. I mean, back in the [ Vikipar ] days, in [ bevacizumab ] days, like, there were wholesale missing of safety events that we should have been identifying in Phase III programs earlier on. And so I'm glad that you're looking at this. This is super important. And there's a lot of programs out there with safety challenges. And we're learning to recognize those and manage it early. I think everything that I'm seeing so far, especially with the huge number of injections, I think the team said over 15,000 injections in 2,500 person years of management with tarcocimab is quite reassuring. I think we've learned that you may not know the full story until you're commercial, right? These really rare events, we've seen that with the pellets, right. It wasn't seen at all on a very large program. So as clinicians, we do think about the safety more used to because we've seen it in multiple programs and it's real. But so far, I'm not seeing anything here of concern.

Thank you. That's really a great dialogue. One thing to remember is that our polymers are conjugated permanently to the antibodies, as part of the antibody viral biopolymer conjugate platform, so nothing falls apart, right? It has the activity and all of that is leaving eye, right, on a consistent basis. So there's no, like, accumulation of the polymer stays separate in the protein. That's not the design. This isn't something that degrades over time, leaving a polymer residual in the eye. There's no residual. It's always leaving and we feel very good about the amounts that we have from the data in the DAYLIGHT study. And the other thing that's nice is that we're not conjugating a PEG, which I think is fairly standard for other agents, and is really a platform extension approach for other medicines. We're really the only company that's conjugating something very different, which is our phosphoryl choline. And so that's important, another differentiation in the overall platform. Because PEG is complex. Was part of a lot of the COVID vaccines, and so there's a lot of complexity perhaps and concerns around things that relate to PEG being injected into the eyeball where once it's there, you can't get rid of it. With that, I think we'll move on to the next part of our story. So we'll move on to Charlie.

Yes.

Thank you.

Thank you. Good. Great. Great presentation, Dave. Okay. Again, it's great to be here with all of you and to go through our development pipeline essentially for the molecules that we've already been discussing today, yes. Am I going the right way? Yes. Okay. So there's essentially 3 sort of sections here that I'll go through. I'll go through each of the molecules, where we are in the development pipeline and what we're thinking about. But I guess the big picture that I would give you from my perspective is we have quite effectively benthic epidemiologic curve of blindness with our current anti-VEGF agents. And they are dramatically beneficial for patients. And it's amazing to see that at a patient level. And this is not new; we've been doing that now for 20 years. But that said, I do firmly believe that we have huge unmet needs still in the field and these fields are huge, tremendous number of patients in the U.S. and around the world getting injections. I see 2 big unmet needs. As Victor pointed out, I think durability is a huge unmet need. I think the burden on patients and caregivers is tremendous and [ can't ] be overstated. That said, patients are willing to come in frequently because they care about their vision. We've seen that repeatedly. The durability is a huge unmet need and I think it remains so in the field. And then secondly, I think new mechanism of action is the second really, really unmet need out there. We have been so lucky that VEGF inhibition has gotten us this far for so many different diseases. We've got to expand beyond VEGF, and the field is actively doing that. There's a lot of really promising molecules out there across retina and I think that we'll talk about one of those in particular today, IL-6. So with that background, you've heard about the platform. Looking back at some of the earliest data we presented, 2020, I think was the first time we presented some of the Ib data. We all looked at that data and said, "This is uncontrolled data", which looking back, misled us in some ways, and we called pulled out of those failures directly. But I think that durability profile that we saw, that signature durability, I think that, that has continued to resonate through the trials in direct comparisons with aflibercept. I think that, that is being driven by the biopolymer conjugates. I think that's still quite meaningful to me. I think those have a useful place potentially in our clinics. So here are the 3 molecules we'll talk about. We'll go through each of them specifically. So tarcocimab, obviously, a tremendous amount of data around KSI-301 or tarcocimab, over 6 studies. I mean it's amazing that the team has persevered. I mean, there's a lot of incredible development across retina over the last 20 years. I think most programs would have quit. They would have seen some of those failures and walked away. I know that, that was some of the discussion early on after GLEAM and GLIMMER that I was part of and those were hard decisions. I think the team did see value there. I see value there. And I think that there is still a road to be had, and I think the end isn't that far away, and that's my hope from a commercial perspective because I think the durability signal is potential real and the DAYLIGHT study is going to make that definitive because they've really put their money where their mouth is with that design, as we'll talk about. And then the last point here is that the development pathway is now relatively clear. There are 3 successful trials and they have 2 ongoing that are highly derisked, from my perspective. So again, this is the durability data and each of these trials was designed with durability in mind. So of course, they all have that benefit and flaw to them. The 2 that I would call out to me as most meaningful are the RVO. Dave summarized it well. But this is the highest VEGF burden disease of all of these exudative retinal diseases that we have. And so to be able to say 75% patients are going 6 months with no or 1 injection is quite meaningful in a disease that is a tremendous burden for known frequent injections long term. And the second one is, of course, the wet AMD data, that DAZZLE, 59%. Of course, it colored because it was a failed trial, but that durability signal, again, I do think is real based on what we've seen, and we'll look at that in some detail. This is a summary of the positive trials in blue and then the ongoing trials over there in purple. The positive Phase III trial of wet AMD, RVO and DR, laying the foundation. And then the thought would be that they may just need 1 more trial to get registration, but then having 2 ongoing that hopefully give the entire package of all 3 indications as early as 2026, because I think it's good for the space. So this is a summary of some of the FDA interactions. I've been involved with some of this directly, in particular, with the GLOW2 discussions. When GLOW1 was positive, the FDA gave quite positive feedback that they were in alignment with study design with GLOW2, in particular with the control arm, which we'll talk about as well as the dosing frequency. And then DAYBREAK, also there's alignment here with the FDA on study design. I think it's a quite unique study design from a retina perspective, where 2 different molecules are being incorporated. I think that's elegant. They can move the space more rapidly forward. You've seen other spaces do this. Oncology, for example, is doing a lot of these combination trials, and I hope that's where retina can go in the future. If we can learn more, more efficiently, it's cost effective; it's time effective; it's better for the space the more we can learn about all of our molecules in development, and there's alignment there with the FDA again on the control arm. And then finally, on the right, there's this concept of having a new formulation. And that always does bring up a wild card, right? There's a lot of programs out there that would love to just switch their formulation in the middle of the development program. It's always a question mark, "Well, how does that work? You have to start over again." And it's nice to have clear feedback from the FDA that a single additional Phase III program would be enough to successfully bridge that formulation to the go-live formulation. Okay. So, we'll dive in a little detail on the study design. This is the ongoing GLOW2 trial. This is actively recruiting at sites across the U.S. Remember, the output in GLOW1, summarized at the bottom, was highly significantly positive. So 41% tarcocimab patients achieved a 2 or more step DRSS improvement compared to sham, quite meaningful. And so one of the thoughts was let's do exactly the same trial. Why change it at all? I think it would have been successful if that's what we've done. But there was one change made to it and that change was to add an additional loading dose. The question is, well, why? From my perspective, that was not needed from an efficacy perspective. The difference between the control arm and the active treatment in GLOW1 was overwhelmingly positive. An additional treatment, I do not think, is necessary to achieve a larger 2-step DRSS improvement to ensure security. Rather, I think that that's a valuable addition because of potential future on-label dosing frequency. A reality in clinical practice is that the label matters quite a bit because it's how the carriers determine how frequently we can give these medications in clinic. And you're seeing that in particular with the aflibercept 8-milligram dose, where the shortest interval was 7 weeks, 49 days. And as Dave beautifully described, there's a lot of patients that need more frequent dosing with both Eylea HD as well as the biosimilar. And so we in clinic are seeing these patients really need frequent labeling, even in drugs that are designed to be super durable. And so that's one of the reasons why that 4-week interval is included there. So there's 3 monthly doses, eventually, when this is approved, for diabetic retinopathy. The primary endpoint, again, at 1 year of 2-step DRSS changes. And then this DAYBREAK. I think this is a great trial design. And we'll talk about the blue boxes here with tarcocimab, and we'll come back to the same slide when we talk about KSI-501 in the next sort of subsection. But we really incorporated the learnings from the previous studies into this to increase the chances of success. So what do you see? Well you see 4 loading doses with tarcocimab. And then you see these monthly PRN visits. So this is essentially a capped-PRN design where patients will be treated every 6 months definitively, so they don't go longer than 6 months. And then they're eligible for up to every month dosing if needed. And we'll talk about what the retreatment criteria are. And again, I want to come back to the green boxes. Let me get to 501 in a second. So this summarizes the flaws from DAZZLE and then how those have been brought into DAYBREAK. And it's amazing in a conference, I guess, to be able to talk about flaws. And one of the first words Victor said was failure earlier. Those words are not usually used in conferences like that. But I love that, because we have learned a lot. We've had 3 failed trials. And we would like to obviously avoid those trials, but we have had failed trials and they give them amazing opportunity to learn from those. I mean, in our space, we want to go from Phase I to the Phase III across all the indications and all the new medicines. I get it because we want to get there fast. But sometimes, if it doesn't work out, it's good to pause and then incorporate those learnings and that's what we've done. So the first flaw with DAZZLE was the underdosing concept. And that's why there's that fourth loading dose. It's also why we're now having the unconjugated component. The second was the reactive dosing, right, to wait until the recurrence of disease is not optimal. And therefore, there's a very clear treat-to-dry approach here with DAYBREAK. We'll talk about that in the next couple of slides. Loose retreatment criteria, this is still very much a debate in the field, right? Is it CST, it's vision, what's the right criteria to retreat? As you'll see here, we're being extremely strict. And then lack of immediacy again, that enhanced formulation will hopefully address that. So this was the DAZZLE data that Dave and Victor showed earlier, right? 5-month durability and 59%. That percentage stands out to me as differentiated, right? These OCT and visual acuity curves are just that subpopulation and with that, comes the challenges, of course, [ without ] looking at the rest. So this slide sort of outlines why we've changed the retreatment criteria. I think this is super important. And this is highly differentiated from every other program out there for retreatment, right? We're always talking about the DAA, the disease activity assessment, the disease activity criteria. And so the questions that we put forward is well, okay, let's say the OCT of 354 micron. Let's say the patient has lost 5 letters. Let's say they've lost 5 letters from their best corrected visual acuity. What do all those mean in a vacuum? Well, it's really hard to know what those mean in a vacuum, unless you have the entire clinical picture. And the clinical picture, to us as retina specialists really revolves around OCT imaging, and that's where we're going now with DAYLIGHT. Really, almost exclusively OCT-based decision tree, which is what we do in the real world. So what about this? What if you look at the central 1 millimeter. That's what most programs do if they look at CST or something a little beyond that. And again, this image is -- shows them because it's a little ambiguous. Right? What are you looking at? You got a little [ PD ] there, there's no intra-retinal fluid. You've got subretinal debris. Is that vitelliform debris? I don't know. You can't tell. But if you look a little broader, and this is the 3-millimeter zone, here it becomes quite obvious. And I would say almost all practicing retina specialists would look at this and say, "Yes, there's obviously disease activity and that eye deserves to be treated" even if they haven't met some very select disease activity criteria. That's why this trial has moved to a very fluid-based treatment algorithm across the 3-millimeters zone centrally to catch any and all disease activity. And this is a summary of it. This gets into the details. But essentially, what's happening in DAYBREAK, the ongoing actively enrolling trial is that OCT images are being uploaded into an algorithm, which will give you, essentially, a real-time output of the presence and amount of intra-retinal and sub-retinal fluid. And then based on that, decision to retreat is being made. And again, this assessment is across the full 3-millimeter central area, and it's essentially at the level of detection. So there's not some large amount of fluid that needed to be retreated. Essentially, if there's any real fluid that is going to be treated monthly in these patients, and again, that's to ensure optimal efficacy in these patients, because we believe the durability will pan out. Some patients will not need it very frequently, but we want to make sure that patients that need it more frequently do get it, so they get the optimal outcomes as well. That's the first chapter probably the longest of these chapters. The second is KSI-501. This brings in all of the learnings that we've had over the previous 6 trials, the conjugation, the bioconjugate into a new molecule. Victor and David talked about this. This is a VEGF trap. So different component of VEGF inhibition; similar affinity as tarcocimab; and then also brings in IL-6 inhibition. And the reason I like this is this is the second biggest unmet need that I mentioned, which is we need new MOAs. And the goal here on the far right of development here is can we achieve better efficacy? This is different than that durability play, which I think is relevant. This is can we do better than pure anti-VEGF inhibition. This, again, is the design, sort of the core molecule itself, very similar, near identical to tarcocimab, except now it's been altered to have IL-6 binding. And on top of that, is the VEGF trap component which captures the simple growth factor, VEGF-A and VEGF-B, very similar to aflibercept. Now there's a lot of data stretching back years looking at IL-6 as a potential target. We see a lot of other sponsors out there looking at targeting IL-6. This is looking at aqueous humor IL-6 levels and they corelate it with anti-VEGF treatment response in wet AND. So, in particular, among patients that were less responsive, we saw higher IL-6 concentration. This is data from a large wet AMD trial, HARBOR, and also a large DME trial called READ-3, again showing a correlation here of IL-6 levels with outcome. And in particular, higher IL-6 levels, both of these with the highest quartile of IL-6 levels in the blue trajectory have achieved less visual gain in patients with lower IL-6 levels in both of these disease states. This is a preclinical model looking at blood-retinal barrier integrity. And you can see the dual inhibition here with KSI-501 directly seems to restore integrity of the vascular endothelium at the bottom, as well as the RPE cells. Again, both of the blood-retinal barriers appear to be restored with inhibition of VEGF and as well as IL-6 with KSI-501. So what data do we have so far with 501? We do have human clinical data. This is the Phase I data that's been presented, which was 4 different doses of KSI-501 up to 5.25 milligrams per injection of the antibody plus the bioconjugate. And this was at 5 clinical sites across the U.S., monthly dosing and then observed through 24 weeks. Limited number of patients here, so I'd avoid over-interpreting this. Seven treatment-naïve eyes, 9 previously treated eyes, and most importantly, I think, to me is the blue boxes here. I think it's a lot easier to interpret treatment-naïve patients especially when numbers are small. And you're seeing a robust improvement in anatomy, most importantly to me, on that bottom-left graph, which is staying dry for 16 to 20 weeks with some slight increase there at 24 weeks, again, suggesting that the drug is wearing off at that time point. On the previously treated eyes, similarly, you're seeing a similar visual acuity gain and then similar anatomic improvement, again, small in values and previously treated patients are notoriously heterogenous in this -- in a Phase I population. This is looking at categorical visual gain. The 10-line gainers is the majority of patients in both. And then 15-letter gainers. I think the 40% is what I would expect and want to see in a treatment-naïve but to see 1/4 quarter of patients there, improving 3 lines or more in a previously treated population does potentially suggest added value of IL-6. So now back, the same slide we looked at before, but focusing on the green trajectory, right? The nice thing here is that it is, in some way, a direct head-to-head comparison with aflibercept 2 milligrams, right, aflibercept 2 milligrams is the control, 3 monthly doses and every 8-week dosing. KSI-501 is 4-monthly doses, again, incorporating the learnings that we've had over the 6 trials with tarcocimab. And then from there is every 8-week mandatory dosing, but again, allowing more frequent dosing with the same retreatment criteria used in the tarcocimab arm. So a very sensitive dial there. Any fluid will drive additional therapeutics, treatments up to every month. And again, the gold area is to look not at durability for this molecule in this trial, but it's to look squarely at efficacy. Right? Can we be better for these patients with IL-6 inhibition? Which I think is a very important question for the field. So do we really need to go back to monthly? And this gets back to what Dave was talking about. I really think that the commercial landscape has skewed the clinical perspective. It's not reasonable for patients to come and expect them to go every 16 weeks, unfortunately. It's led to a lot of frustration among patients to be told something commercially when it's just not panning out to be the case. Many patients do need monthly dosing, for example, with aflibercept 2 milligrams. This is the best paper looking at this. This was Glenn Jaffe's work published in Ophthalmology in 2016, and this a great paper. On the left, it shows that monthly aflibercept compared to every-other-month aflibercept are identical outcomes, critically among patients that are dry. So the OCT really can predict outcomes. But then on the right side of this graph, you see a separation of curves and what's driving that separation? If I'm a patient, I want to be in the light pink or the light purple curve at the top. I want to have more vision gain. And that's where my patients want to be. They want to see better, and they'd rather get more injections and see better than get less injections and see worse. And what you're seeing here is that the dark purple line is every 8-week injections of 2 milligrams aflibercept compared to the lighter purple at the top, which is every month injections among patients with persistent fluid. So we know in clinical practice that fluid correlates with need for dosing. It's not a perfect correlation of visual function, but it's what we use clinically to drive frequency in these patients. All right. The last of 3 chapters here is KSI-101 and we'll talk about the design of this molecule in a second. This is differentiated because it does not have bioconjugate, otherwise, it's the identical protein structure, the antibody itself as KSI-501, again without the conjugated form. And the concept here is to go into a different space. So this is inflammatory retinal diseases with associated macular edema with some fascinating possible means for accelerated development. So we know that uveitis and inflammatory diseases more broadly are common causes of vision loss and within those patients, macular edema is one of the most common causes of vision loss. So we currently do have therapeutics for this. This is not a population of patients without options. Or they all have specific challenges with them, both local therapies that are steroids, as well as systemic therapies, all have their limitations. Again, coming back to data on IL-6. We've seen elevated levels of IL-6, both in aqueous on the far left, in the vitreous in the center there among patients with inflammatory diseases and including, specifically, uveitis. And then on the right, the interesting observation here is that VEGF may be playing a role in these patients with macular edema in the setting of uveitis and ocular inflammatory diseases. So this may be a multi-component disease process and targeting both IL-6 and VEGF may be important. This is great data put forward by Roche. This is from the DOVETAIL trial that was presented last year. I'm looking at a monoclonal antibody inhibiting IL-6. And importantly, this was proof of concept in humans that an intravitreal injection with, again, monotherapy IL-6 inhibition can lead to visual improvement and OCT improvement in a dose-dependent fashion, indicating IL-6 is important for this patient population. Critically, however, when I look at this data I see here on the far left, you see meaningful improvements in IRF in the proportion of patients without internal fluid, but then the other side of that is, there still is about 50% of patients with persistent IRF after 3 monthly doses, again, with IL-6 monotherapy in this program. So that brings us to KSI-101, but again, this is the same protein antibody structure here with the IL-6 inhibition as well as VEGF-A, B and placental growth factor being blocked by the trap formation there at the top of the molecule. And again, this does not have the bioconjugate formulation to it because the play here is not a durability play. The play here is can we be as effective as possible in this patient population and that's why the formulation is even stronger at 100 milligrams per milliliter. So this is the design of the ongoing APEX study. This is a Ib trial, looking at both DME patients at the top and what we're calling a MESI population or macular edema secondary to inflammation, again actively recruiting. There's 5 loading doses in the DME patients and 4 loading doses in the UV macular edema and also a broader inflammatory population with macular edema, with the anticipated primary endpoint for the Phase III program, being there at 16 weeks. So very early to read out, again, agreed upon with the FDA at this time point. These Phase III trials, PEAK and PINNACLE, Phase IIb/III, are not ongoing now, right? This is dependent on the readout from the Ib APEX trial, which is ongoing. But if those are positive, the plan would be to look towards this sham-controlled trial in patients with inflammatory genes with macular edema, 4 monthly loading doses and then an as-needed retreatment paradigm from there with a primary endpoint, again, early in this trial, again at 16 weeks. The other potential interesting angle here is an even faster mechanism to get to clinic, which is to look at a pediatric uveitic and broader inflammatory disease population with macular edema, and this represents a substantial number of patients in uveitis clinics. So this is just summarizing all of those 3 molecules that you've heard many times today, tarcocimab, KSI-501 and 101. If I depict 1 or 2 things that are most important to me for each of these, I would say from a tarcocimab perspective, it's that we've had a lot of learnings over a lot trials. And I think that those had been well understood, as well as we can understand at this stage, and then meaningfully inform the design of both GLOW2 and DAYBREAK going forward. KSI-501, very excited to have a new mechanism of action in the space. I think that needs -- we need that in the space to move it forward to get better outcomes for patients. And importantly, we apply to all of the learnings from the bioconjugate from tarcocimab into the 501 development trajectory. And finally, 101, right, steering us in a completely new direction, I think, helps diversify the team. And I think it's good for patients with inflammatory diseases to see possibly a combination of VEGF and IL-6 inhibition. With that, I'll pass it back over to Victor?

Great. Well, thank you, Charlie. That's a wonderful overview and we appreciate that. I'm going to run through what some of that means. What is the opportunity for Kodiak with what we've described so far? So one of the things that's important to remember at Kodiak is these are very high-prevalence diseases and there's a tremendous opportunity for the patients to solve these problems but also a tremendous commercial opportunity. And we're playing in the big leagues as a mainstay biologic, trying to access some of the benefit for the patient and the revenue. So it's a $14 billion market growing towards $20 billion. In wet AMD, what we call the established Gen 1 anti-VEGF agents, achieved modest vision gains in the real world and as we've heard today, require frequent injections to maintain the vision. And the Gen 1.5 agents provide modest dosing interval extension in the real world. So I think what's important and what we've heard multiple times from our experts is that the Q16-week or the every 4-month numbers that may be quoted in the direct-to-consumer advertising actually are not being achieved or even close to that in the clinics. And interestingly, although the 45%, Q16-week number is what's shown in the faricimab wet AMD clinical trials, in recently independent U.K. from real-world data, from what we see is actually that Q16-week patients represented maybe 18% of patients in the real world or in a year 1 study from Liverpool, Q16-week patients represented 0% of the patients on Vabysmo. And again, the Gen 1.5 agents don't provide additional vision benefit over the Gen 1.0 agents. As Charlie mentioned, the data that we've generated in everybody on 6-month dosing in the GLOW1 diabetic retinopathy study was very strong data. With U.S. prevalence for diabetic retinopathy patients at around 8 million patients, fewer than 1% of whom are currently treated, this represents a very interesting opportunity for tarcocimab and for public health. And to a large degree, I would say that's due to treatment burden with the existing anti-VEGF therapies. So it's really a watch-and-wait approach. And what we know is that disease progresses when it's untreated. Despite the limited differentiation of these agents, whether it's Eylea over Lucentis or the Gen 1.5 agents or Vabysmo or HD over Lucentis and Eylea, I think it's important to remember and recognize that each incremental improvement has resulted in blockbuster commercial opportunities for these mainstay biologics. And I think it's been especially illustrative to see how quickly Vabysmo has built their commercial potential despite what people would presume to be a complex and perhaps saturated market, and remembering that Avastin still represents around 50% of injections. So despite that, we've seen Lucentis grow into Eylea, Vabysmo grow much more rapidly than people anticipated and Eylea HD will, we think, fix their not having monthly in their label profile and broaden their profile and perhaps grow their revenues more aggressively. But if you look at what we're trying to build with tarcocimab, what we call the Gen 2 profile of good efficacy, strong immediacy and the best durability, we'll have, we believe, if successful, a very broad label and something that may even represent more than incremental improvement, with each 1% share in the market estimated to translate into $200 million to $300 million in net sales which is quite substantial, given where Kodiak is trading today. And as we mentioned earlier, our objective with our development is to develop tarcocimab and 501, our ABC medicines as mainstay intravitreal biologic monotherapies that themselves prime the patient and provide the durability, the high efficacy and the high durability with the label that goes from every 1 month through every 6 months. So hopefully, more than incremental improvement on top of the mainstay agents and delivering a level of durability that suggests why would we want to go to more exotic technologies when a biologic can service -- when the biologic -- the Kodiak biologic can service the unmet need. Just to talk for a second about the opportunity for KSI-101. We think it has the potential to be an important and differentiated medicine in the retinal inflammatory conditions, which we see as a greenfield market segment. And stepping back a little bit to talk about a relevant case study, we don't pretend to be able to achieve the level of success of TEPEZZA in thyroid eye disease, but I think what's important to recognize is that, at its time, was also a greenfield market opportunity launched into a non-existent market. There was a high unmet need with no approved therapy. And sales approached blockbuster status even in the first year of launch, substantially outperforming the management expectation of $30 million to $40 million in year 1. So with KSI-101, there is a greenfield market opportunity, a broad patient pool, a high unmet need with no approved locally dosed intravitreal biologics, and furthermore, it's a very interesting gateway indication from, MESI or UME, or macular edema associated with inflammation, all of those in a bucket. Those are gateway indications for our molecule to a broad set of diseases, which could include, down the line, all retinal diseases that have inflammation and macular edema. And we have spoken with the agency on this question. So what's special about KSI-101? The dual inhibition, what I call a predominant anti-IL-6 mechanism supplemented with the anti-permeability of the anti-VEGF trap and a synergistic effect both of them together on normalizing the blood retinal barrier function versus anti-IL-6 monotherapy, the data that Charlie shared. So the potential for it to actually have some level of disease-modifying capability and a very high-strength formulation at 100 mg per ml. So therefore, a very high potency provides the firepower needed to treat angry inflammation and the macular edema. Sort of a sledgehammer-type approach for these patients and we're exploring different types of accelerated development for the molecule, including options in pediatric UME patients. So Kodiak's clinical portfolio has the potential to provide continued revenue stream starting from 2027 with a built-in life cycle management and risk diversification. So 101 is a revenue potential diversified away from our ABC medicines. 501 has the potential for best efficacy and signature durability as a next-gen product to fast follow our own tarcocimab. And meanwhile, tarcocimab has the potential to bring the best durability and strong immediacy. And incremental market share can translate into meaningful revenue for Kodiak in these time frames. And let's remember that Kodiak owns full commercial rights to our portfolio, which allows us the flexibility in our commercialization decisions to support adoption of our products. And also, I think another important element of Kodiak's capabilities and resource and experience is the long-standing and significant investment that we've made in our own commercial-scale manufacturing facility which positions us to launch multiple ABC products into large and growing markets. And here, you see the grand opening of Kodiak's Ursus facility together with Lonza, which, as I mentioned, mechanical completion was in the first half of '22. It was commissioned as a GMP facility in January of '23, and we released GMP commercial scale product that we enhanced tarcocimab formulation in November of 2023. And many -- a number of the different PPQ batches that are required for the BLA have already been completed. For example, the antibody components of tarcocimab, those PPQ batches are already completed as a drug substance. Furthermore the PPQ batches for the OG1802, what we call the biopolymer that's used for the conjugation, those batches have also been completed in Ursus, and we're working now to begin the PPQ batches for the tarcocimab drug substance, and that will take place in the first half of this year. So that is a unique asset that positions Kodiak well as a commercial player if we're that lucky and successful. Now this is something that we haven't shared before exactly. And this is what we call our VETi or our visual engagement technology and imager. And this, we believe, can form part of Kodiak's commercial franchise for example, tied into the launch of tarcocimab. And, essentially, being in Silicon Valley as we are, we're located in Palo Alto and having access to important technology talent from whether it's Apple's Face ID team or the Google X team or other high-tech groups, Kodiak has built, over the last several years, a goggles-based imager, okay, that delivers an OCT that's a 2-dimensional OCT for the patient. And that can be translated into a tool that could be bundled in some fashion together with tarcocimab and our other ABC medicines through their commercialization and there's an element of bringing the AI and bringing a series of different apps and AI potential. And we mention this because we're now using some of that more modern AI in the execution of the DAYBREAK study when we look at the presence of the intraretinal and subretinal fluid in the DAYBREAK design. So as we think about DAYBREAK being a study that generates data for how the molecule truly will perform, okay, and then the ability for it to also predict how the molecule can be used once it in the market. Kodiak's VETi tool is something that we think can form part of our interesting commercial potential. So what is the Kodiak opportunity? So in Kodiak's 15-year history, we've made missteps, mistakes. In the 3 clinical programs in our pipeline, utilized the learnings from these mistakes to increase the probability of rebuilding significant value over the next several years. So our 2 ABC investigational medicines built with our proprietary science of durability and of our new science of the enhanced formulation, both being developed as mainstay intravitreal biologics, along with the other majors to address unmet need for high efficacy and high durability. With 90% of tarcocimab's clinical and manufacturing costs, they're behind us and we anticipate a high probability of success in our late-phase clinical trials with the potential to launch in 2.5 years. We have significant flexibility in commercialization to support adoption as well as the ability to enter into partnerships if desired as a full-fledged retina franchise R&D company today in an area or a market where incremental market share gain translates into meaningful revenue potential into a $14 billion plus growing market, with built-in life cycle management with KSI-501 to fast follow tarcocimab, and an uncorrelated bispecific KSI-101 product in a greenfield market with near-term data readout early in 2025, where we may and are exploring the ability for accelerated development and approval options. So today represents an exciting investment opportunity, we believe, for Kodiak and we think that the science is what needs to take us there. So rather than moving into the roundtable discussion, in the interests of time, we're going to have Dolly Chang now come and present on some of our deeper pipeline.

Maybe we do the Q&A now with Dave and Charlie, because they need to leave. Okay?

What do you think? What time do you guys have to go?

1:20.

We're okay.

Very good. Okay. Yes. Okay. Thanks, [ Tom ].

So I'm very thrilled to be here as part of the Kodiak team. As Victor mentioned before that, I was at Genentech and Roche, responsible for the early clinical development portfolio. And what drew me to Kodiak is the very exciting and tremendous potential of the portfolio that just heard before and also the strong commitment to the science. And over the next couple of minutes, I'm going to walk you through some rapid advancing of the ABCD Platform mentioned by Victor, and also some of early pipeline as well, which I think that it provides tremendous potential to change the way that we treat complexities in ophthalmology and even beyond. So if you just review the current drug development landscape, it's pretty much most of the drug followed the "one drug-one target" approach. It works in many patients, but it also brings some limitations. It's oversimplified for their complex biology, it may miss some of the potential synergistic effect. And there are some breakthrough in the field that we know, but they all come in with their key limitation. So just take a couple. For examples, if you look at RNAi therapies, there's tremendous breakthrough, but it is still limited to deliver to the muscle and also have very limited cellular uptake. If you're looking at the oncology field, the ABC field still has a limit of DAR, which is the drug loading to 1:8 ratio. And so if you just imagine, if there's a foundational platform that can address multiple disease and multiple targets all at 1 time and then the best thing is if this new drug concept is designable, it is manufacturable and it's being safe to the patient, then we may change the way that we treat patients. So as Victor mentioned earlier on, for the ABCD Platform, we think that this could potentially bring this kind of vision to life. And it coming with, I would say, 4 main advances and advantages. So first of all, it allowed us to conjugate a wide variety of payloads, including small molecule, peptides, oligonucleotides and proteins as well. And in terms of the second benefit, it accept high DAR, which is the high drug loading ratio up to the dozens and the hundreds. And we are going to walk you through a little more of how to achieve that. Number three is we continue to release profile of the payload by customized linker. And then number four is the drug can be embedded in the phosphorylcholine backbone that has been proved to be biocompatible and in many different tissues then with tarcocimab in the eye. And so how does the ABCD Platform work is it's highly modular. And for each component, it can be customized design and developed. So let's start with the A, which is the antibody and the biologics. So within Kodiak, they're very strong protein engineering team that can create different proteins that are targets towards different tissues, specific tissues or modulating a pathway. And for the biopolymer is it's highly customizable. And we can customize different size and loading with different drugs. And let me just take a quick example of ocular applications. So let's say, we can choose a 750 kilodalton, the larger biopolymer, with a 10% of drug loading, and that can achieve a DAR of 250. That's pretty incredible. And then let's say, for systemic applications, we can take a smaller biopolymer as, for example, 250 kilodalton with a drug loading of 3%, that can achieve a DAR of drug loading, let's say, 25, and that has not been achievable with the ABC fields or the RNAi therapy. And lastly, to talk about the drug cargo. We have talked about it that it has highlighted the capability of we can work with different types of property of the drug cargo. So -- and that's including different size, different solubility and different [ charge ] as well. And with the customized linker, we can have varying release half-life as well. So let's just put a couple of the design concept into what we mean by this kind of ABCD Platform. So first, we can take look at a small molecule and a peptide that we can embed a lot of them with high DAR up to, let's say, 250 into the biopolymer. And we are able to improve the solubility and then carry the durability in the eye. And with the releasable linker, we can also tailor the release profile as well. So these 2 are the foundation for the dry AMD and the glaucoma [ pioly ] pipeline that I'm going to talk about in the next couple of minutes. And just give the flavor of the potential of the ABCD Platform. If you're looking at RNA-based therapy, we can conjugate with the antibody to target specific target tissue with carrying repair with the small biopolymer to carry multiple copies of RNAi therapy to overcome the limitation of the cellular uptake. And if you look at that, lastly, we can also potentially create in oncology field a second-generation ADC that increase in the DAR, so that we can choose different payloads with less toxicity and with a higher number to overcome the safety and efficacy issues. So now let's just go into the 2 early pipelines that we have. One in glaucoma and one in dry AMD. And as my background as a glaucoma surgeon, I do want to highlight the unmet need in glaucoma field. So we know that most of the medicines are targeting for lowering the eye pressure in the eye because that's the only modifiable risk factor that we can tackle with. But we know that patient either can do the eye drops, cannot tolerate eyedrops, or they don't want to adhere to eye drops, or they may still have high eye pressure despite multiple drugs and multiple surgeries. And at the same time, even with controlled eye pressure, the disease can still progress. So that's probably why glaucoma remain the #1 irreversible blindness worldwide and that demonstrates a tremendous unmet medical need in this field. So I'm really excited that with the BCD component of the ABCD platform, we are combining 2 mechanisms of action of 2 different small molecules. Number 1 is the neuroprotective agent, NLRP3, to -- and the second small molecule with IOP lowering to achieve 1 intravitreal injection with 2 mechanism of action and they only require 1 injection every couple of months. And the reason why we choose NLRP3 as a target is because the NLRP3 inflammasome play a key role in glaucoma progression. The inflammasome complex is the intracellular complex that drives inflammation and cell death. It is actually a very exciting target for many of the disease area. Many pharmaceutical companies are targeting for this pathway but it has been very challenging for the eye because of the fast clearance of the small molecule and also the solubility issue. But I'm very proud that our team were able to develop a novel NLRP3 inhibitor, specifically designed to work with the ABCD Platform so that we're able to bring this to the glaucoma patient. And then to Victor's point, that how could we know that without the antibody the biopolymers still work in the eye with extended durability. So from the rabbit PK study, we can see that if we look at the ocular half-life and the biodistribution of the biopolymer alone compared to tarcocimab, which is the biopolymer with antibody, the half-life is very similar. So for the glaucoma program, we're going to take a biopolymer that is very similar to what we have used in tarcocimab and so we are anticipating long-acting delivery in the eye. And so there are a couple key highlights, key insights that when we work with the ABCD Platform for the glaucoma Duet program. So first of all, we already talked about how we pick the biopolymer. And we also have shown that we're able to incorporate 2 different small molecules in a single biopolymer. We also have seen consistent release simultaneously for the 2 different small molecules. So we think that this will be very exciting and a tremendous opportunity for the glaucoma patients that we can bring the neuro protection and in lowering the eye pressure both in one. Next, I'm going to talk a little bit about the dry AMD program. And what we mean by dry AMD is that the advanced form is geographic atrophy that we are quite familiar with and also the earlier disease, which we call intermediate AMD. And we know that there are some treatment options for the geographic atrophy patients but the outcomes are pretty limited. It moderately slowed down the GA progression. And there's no therapy for the earlier disease like intermediate AMD. So our approach is to combine a macrocyclic peptide for complement inhibition and also incorporate the novel NLRP3 inhibitors, which you know that inflammasome also play an important role in AMD. So by combining the 2, we think that we can achieve better clinical outcomes because we think that by dual mechanism of inhibition, better tissue penetration, and we are able to achieve dose higher with the ABCD Platform so that we potentially can improve the efficacy for the treatment of geographic atrophy. And secondly, because of the extended durability, hopefully, we'll be able to provide a gentler treatment just like the DR scenario that we're able to treat patients earlier and prevent the progression into late-stage diseases. So those are the 2 programs that I just want to provide a high-level overview of what we are working on and addressing 2 very important unmet medical need and very poorly addressed market. Now, I'm going to turn it back to, Victor. Thank you.

Thanks, Dolly. So just as a quick summary before we move into the Q&A, a summary of the 3 clinical programs that we're advancing that we've talked about. A nice introduction. Thank you, Dolly, onto the ABCD Platform, which brings 15 years of development and also brings our commercial facility in light where we can build these conjugates in the future in the commercial scale. And we'll bring our time lines up on the screen, perhaps while we go through the audience Q&A to remind everybody that our focus as a company is on late-stage clinical execution on the tarcocimab program, the 501 program and the 101 program. With GLOW2 actively enrolling, we're trying to target completion of enrollment towards the end of this year or early 2025 and DAYBREAK, also actively enrolling, as a 2-year study and looking towards a completion of enrollment, let's say, towards the end of Q1 of next year if we can achieve that with 48-week endpoints and top line data very close to last patient, last visit, taking us into very early 2026. And then exciting data we hope from KSI-101 both DOVETAIL type data, we hope, open label early 2025, the decision of which 2 dose levels to move into the PEAK and PINNACLE pivotals driving the initiation of target enrollment start PEAK and PINNACLE and then target enrollment complete with a 16-week primary endpoint. So looking at an exciting execution 2025 and a large number of really important readouts for us and hopefully the field in early 2026. Thanks very much.

Good. Let's do some Q&A.

It's Michael Yee from Jefferies. Maybe for the doctors, I appreciate all of their clarity and commentary. Maybe just taking a step back, assuming the data for tarcocimab play out as you expect and given the enhancements and improvements to the study design, maybe, Dr. Brown or Dr. Wykoff, could pontificate about how you would expect to use these types of drugs in the first few years, what types of patients? It's obviously a complicated field, so that would be helpful in terms of trying to explain how one might use that type of drug profile? And then if an IL-6 compound is also promising, where does that fit in and how would you use that type of drug?

Yes, I'll start with the first one. If it's -- you're looking at a situation where you're probably going to have a bunch of aflibercept biosims. And so I assume that there's going to be step therapies from insurers, et cetera, to step through that. Depending on where they market it and price it, if you indeed can go 1 or 2 weeks longer with this drug or even 3 or 4 weeks longer, there's going to be a big push for all those that don't. Can't go 8 to 10 weeks with 2-milligram aflibercept. So I think the devil's in the details. It all depends on how long people can go with this trial design and you'll be able to see it because you can get up to monthly and it's capped at 12. If you can really get most people close to 12, I think it's going to be very commercially successful. IL-6, I don't treat uveitis, so. He doesn't either, but.

Yes, I think that the perspective there is that most retina specialists tend to look at their high-need patients as the first one to transition, so to switch. You see all that with the bias mode, with IV HD. And it's those high-need patients that you're seeing every month in clinic that they have referenced, that are constantly asking and are remarkably well informed in many cases about, "What's next? What's next? I want something better than what I've got now." And so those are the ones that are the lowest hanging fruit to transition. And I think if the half-life data bears out in the DAYBREAK study, and we see a meaningful proportion of patients that are beyond 8 to 12 weeks, I think that, that will confirm what we've seen so far with the PK data and that would be meaningful for patients. Look, yes, we think we can get you meaningfully further. And that doesn't have to be 4 months longer. As Dave said, an extra few weeks for patients is highly meaningful from a quality of life perspective. Incremental benefit is incredibly valuable for our patients. And then where I see the 501 -- a lot of feedback up here, I don't know if that's a me thing or something else. But the 501 to me is not a durability play. It'd be great if it did have that durability angle. It should, given the structure of the molecule, but to me, I'm most interested in, can we do better with the additional MOA. For me, that's why I was excited about Ang2 for a long time. That's why I'm excited about many of the molecules in development as I truly believe we can and will do better than VEGF monotherapy inhibition, but we're not quite there yet as a field and maybe IL-6 will move us meaningfully forward. I don't know.

Right. So that will be on PCBA and other endpoints or how should we...

It's all of it. So when I say A -- it's a good question, right? There's a nuance discussion there about what do we really mean about efficacy. But ideally, from a patient perspective, it's vision. Right? That's sort of a holy grail. Can you see better than we can do with the gold standard anti-VEGF therapy? And that would be ideal. Because usually, if it's not vision and it's anatomy, that's a durability play versus an efficacy play. So really when you talk about efficacy, you're talking about visual function. And there's complex ways of looking visual function that we don't get into: contrast sensitivity; visual field; all these other things, which we have not been able to quantify well in our Phase III trial to show benefit. But patients talk about all these other additional benefits when [indiscernible] goes away.

Mike DiFiore, Evercore ISI. As well, thank you for Kodiak putting on such a comprehensive informative event here. Just want to pivot back, and had two questions. Number one, just want to pivot back to the formulation. And correct me if I'm wrong, I recall that a while ago when we were just announcing this new formulation, I think the -- was it the 70/30 conjugated versus unconjugated mixture? Now, it's 80/20. And my question is, what gives you the confidence that you've struck the right balance or ratio? Is this built on actual experimentation or is it built on modeling? And I have a follow-up question.

Thanks, Mike. Well, the formulation for the tarcocimab go-to-market is a 20%/80%, the 1 plus 4. And the formulation as, I could say, go-to-market for the KSI-501 is 30%/70%, or a 1.5/3.5. And the reason we did that is we believe the 1 plus 4 for tarcocimab is the right strength. Okay? And Dave went through a little bit of some molar comparisons for the anti-VEGF, right? That a 1 plus 4, right? The 1 is equivalent from a VEGF binding to a 1.3 mg of Eylea or 0.7 mg of Lucentis or a 2 mg of faricimab. Okay? Now in KSI-501, that bispecific protein is slightly larger than the antibody is in tarcocimab. So tarcocimab, the antibody is, let's say, 150,000 molecular weight conjugated to the polymer. But for the KSI-501, that bispecific protein is, say, 190,000. And so because it's a little bit larger, we wanted there to be a little bit more room, right, because to get a certain number of molar equipment or the stoichiometry, we've shifted up from 1 to about 1.5 mg of unconjugated to achieve a high molar amount to be unconjugated vis-à-vis the conjugated. So that's the type of balance that we drove. And we believe in both cases that the objective is to have this unique signature durability, the promised land of longer in interval durability. And in order for patients to be able to get there to have the unconjugated so that we can try it on every patient and have that early strength needed to see where the patient could go as a mainstay biologic therapy for all patients.

Got it. Very helpful. My second follow-up question is just on the DAYBREAK study, just fully acknowledging that this is a non-inferiority trial versus Eylea, I guess, the risk here, if there's a risk, is if KSI-501 doesn't appear to be at least directionally more efficacious than tarcocimab, what -- how do you think about that? And what are your expectations for 501's efficacy here? I mean, I assume it's -- is that it would be a lot or some degree of directionally improved versus tarcocimab. But any color you can add to this would be helpful.

Sure. I think for the 501, the study design with an objective to try and really push for trending better efficacy, it's very hard to achieve that when you look at the broad mean population. And so I think what's a nice model is to look a little bit at the Vabysmo bispecific, where they went for superiority and did not show it. Okay? They showed noninferiority. But now there's quite a lot of analysis to try to identify certain large subpopulations of patients within the broader population to try to really identify well, where actually this medicine with its bispecific mechanism, actually made it be able to achieve something special. So we may not need to look for the mean curve showing a superiority, although a trend would be nice. I don't think that's necessary at all, and we shouldn't set that as the standard. The objective is to explore at a high-intensive dosing, whether there are certain subpopulations of patients where we have a very different behavior. Certainly, within aflibercept control patients and at Kodiak, we have many thousands of patients now when we evaluate data. There is tremendous unmet need across the disease areas even with drugs like aflibercept where they have disease reactivation or significant increases in subretinal fluid across the diseases. And the question is, with the KSI-501, do we think there's a different pattern? It's hard to know. We could spend 2 to 3 years in Phase II, trying to figure it out. We're going to look now in a Phase III setting combined with tarcocimab in a very cost-effective manner, right, in enough patients over enough time, right, in a non-inferiority setting for the threshold for success, right? But hopefully, to be able to find like Vabysmo is trying to do, large opportunities to describe differentiation in different ways.

Yes. Along those lines, if you can kind of at this moment any -- what any predefined subgroup analyses may be or predefined? Again, even subgroups might be in study or is it too early?

No. I mean, I think the nice thing is that the primary endpoints set and then we have some time to think a little bit about secondary analysis.

I mean the one layer I would add to that is that the typical subpopulations are the larger lesions, more fluid, worst vision at baseline, the polypoidal patients. There's subpopulation of patients in wet AMD that the field's pretty aware of are typically a little more difficult to treat and have a worse outcome.

Andrea?

Andrea Tan, Goldman Sachs. Victor or John, just a question here on your cash runway. You mentioned $200 million ending 3Q, which gets you into 2026. Just curious if you can confirm that we will see some of these top line data reads before that? And then how much does that account for any type of development or discovery work from the ABCD Platform?

Yes. I can take it first and John can provide perhaps more color. We don't make any promises about -- I think what's going to be important in 2025, say, end of Q1 or early Q2 of 2025, where will Kodiak be. We'll have data that I hope will be DOVETAIL-equivalent data out of the KSI-101 program, and we'll be selecting dose levels to start the PEAK and PINNACLE pivotals for the KSI-101 program. We will also hopefully have completed enrollment into the GLOW2 study. So we'll have visibility into the time line and top line data readout for GLOW2. And I think also towards the end of Q1 or, hopefully, early Q2 in 2025, we'll have completed enrollment in the DAYBREAK study. And so we'll also have visibility into the time line with the 48-week primary endpoint readout for top line data for DAYBREAK. So I think when you look at our cash position, we're not going to make any promises exactly whether or not we can take us all the way through, let's say, DAYBREAK top line data. I mean, as a public company, that's not what you do. So I think our commitment is that we're going to be thoughtful about when we raise capital and how we raise capital. I think whether or not we could get to GLOW2 data or whether we get to DAYBREAK data is something that's beyond where we think we need to discuss today. What we're looking at is, let's say, in early Q1, Q2 of next year, perhaps whether or not we have transparency and clarity about these time lines, and that will help us to determine what kind of capital in what form. Really, the type of capital we might look more really may only be 2 quarters. So it's really not a lot. Gena? Oh. I'm sorry. Go ahead. Yes. Thanks. In the back, yes please.

I doubt if you know me. I'm on the G40. If you go back to durability quickly, have you been able to flush out yet any baseline characteristics of patients who have lower durability?

I think that's very challenging. And if you think about the level of expense that some of the other technology companies are doing to do run-ins of all subjects, whether treatment-naïve or previously treated on the anti-VEGF biologics before they move into the port delivery system or to depots or even therapies, I think that tells you that people don't believe there is a way today to really know.

I've been looking at it forever. I don't think there's a way. If we have an assay like Regeneron's got a very good assay for aflibercept. They could probably do it with an assay in multiple tests, right? You can tell if you're a fast clearer or a slow clearer. But individual pseudophakic patients, more myopic eyes should be a faster clearer, but we've never been able to identify those and so we treat and extend to get them there.

Right. I will say that, for example, if we want to try and identify well, what percent of treatment-naïve patients entering wet AMD studies may meet criteria as well controlled, that may be in our analysis of, say, our data on historic studies, you're looking at around less than 25% of patients. So that's why when you look at what we're trying to achieve by going for 100% of patients, that's why we think we're developing a mainstay monotherapy. But it also shows you the challenges for other players, where they're selecting smaller groups and trying to generalize those results is very challenging. Maybe our VETi tool will help to solve that conundrum, but that's going to be a few years down the road. Gena?

Also follow-up the DAYBREAK data. So I think that, Dr. Wykoff, you mentioned that you wanted to see higher percentage of patients were able to achieve, say, 8 weeks to 12 weeks treatment frequency. So maybe wanted to know like what is the bar? What percentage are you looking for? And also, now we use different criteria like this is a fluid volume versus I think the faricimab, all the other trials are using BCDA and OCT sickness. How do we do comparison? Like a 45% clinical trial setting? How do we do that comparison?

Yes, it's a good question. I don't have a definitive quantitative answer to that. I think the important points I would make are that this is a highly differentiated trial, right? This is very different than every other durability trial that we've had in the space because they're artificially set and all the sponsors have done this, including Kodiak. They've talked about this beautifully before where they're set so that you're putting your best foot forward. And I get that. That makes sense, and that's been the standard in the space. And this trial really turns that on its head because it's really the first and foremost goal here is let's make absolutely sure that we hit noninferiority, and that we have really good drying. Right? Let's make sure the efficacy is first and foremost. And then built into that is the concept of [ our ] as a team really believe the durability signal, although the trials have been slightly artificial that they had across the entire space, we still believe in that signal. And therefore, in the trial that's designed to newly treat any amount of fluid, if you believe in the durability message, then those patients will still have to top and we'll still see a meaningful proportion of patients that are able to go on an extended interval. So I'd say, we're sort of trying to get both in 1 design focusing on dryness, which I really like. It's how clinicians practice in most cases. And that said, you have to really press them, well, what is that percentage? I don't know. I mean I could tell you that if 100% of patients are getting monthly dosing, that's not going to be great. That's obvious. One very extreme. I don't think it's going to be 59.4% of patients getting 5 months or longer, right? Some of the patients have some fluid. So where is that sweet spot? Well, if you look at [ CAT] and HARBOR and all these trials that were as needed after the first sort or few shots, some in the range of 5% to 15% of patients may not need additional shots through 1 to 2 years of follow-up after you get dry to begin with. So the percentage would need to be meaningfully higher than that, right? We can assume 10% to 15% of patients don't need any more shots and, therefore, they're definitely going to be every 6 months intervals. So I'd love to see at 3 months and on intervals, I'd love that to be 30-plus percent because that to me would be a meaningful differentiation from the other end of it there.

So if you'd like to add to that. If you look at view, about 60% of treatment-naïve patients are dry on aflibercept Q8, which means 40% aren't. If you can get more than 60% Q8 or longer, that's meaningfully different. The other thing is and what I applaud them is there is no biologic or scientific reason to acquire vision loss before you retreat. They only do that to decrease the number of injections to make their drug look better. Fluid is a pure biomarker. I mean you can argue how much fluid, I don't like any fluid, but the amount they're picking here makes way more sense than making you lose vision and have fluid.

Right. We've tried to design the DAYBREAK, to have GLOW2 be very -- we anticipate a very high probability of success for GLOW2. We like the addition of that third loading dose at the beginning. So that provides a lot of flexibility commercially for reimbursement to be used broadly in the diabetic retinopathy population. So that's very great. And then in the DAYBREAK study, we've really tried again to think about skating to where the puck's going to be, to say, let's not think about how other people have designed their studies, but let's actually think about how the physicians are going to take our data, right? Because we think we can rapidly move the data into a BLA and we have our commercial facility with all the PPQs and rapidly move that into product, and we hope we can do that in the context of a prefilled syringe. So we're thinking where is the field going to be and what kind of data do we need to generate to educate the physicians about how the drug really works, rather than trying to think about like well, how do you understand whether or not you've lost vision or 80% or 90% of the patients on 12 or 16 weeks. We're trying to do it a little bit differently, but importantly, have a very, very high profitability of meeting the noninferiority in the study for both molecules. Being a bit scarred from our experiences in the past has pushed us, I think, hopefully, to be more thoughtful.

I have 2 on the formulation and 2 on the trial. Should I split them up or? Okay, let's have the 2 on the formulation maybe. So the ratio of 1:4, presumably, it could have been 2:3 where there's more unconjugated in there, and that could radically have also obviated the need for having 4 induction shots versus 3 in the trial. What was the thought process, I guess, on your end? And then I have a few follow-ups.

Well, one is we weren't completely free. We couldn't drive the change in a -- you know, invert it, for example, because our historical data was predominantly conjugated. We wanted to stay close to that. So that's, I would say, one element. But in reality, our conjugate did extremely well in the majority of the patients and the majority of the studies. And so what we're trying to do by bringing the unconjugated is to drive a bit more immediacy, let's say, in the higher-need patients. And by looking at that percentage, right, and the molar equivalent, like, comparisons, we think -- we don't know exactly where we're going to need to go. But like I said, for me, it's arithmetic. We're going to definitively close the gap or narrow the gap. I don't think actually the field demands that we be exactly overlapped with Eylea, right, in the loading phase or that we be better. I think it was an amazing surprise when 8 mg aflibercept didn't dry better through the loading phase than 2 mg aflibercept. I mean, from a science standpoint, that's a bit of an earthquake. I think what that tells you is that you're saturating earlier. So with our 1.3 mg equivalent of aflibercept, right, in the 1 plus 4, we're going to definitively be a lot better. And the goal is to get the patients to the promised land for the majority of patients predominant, very long or longer interval dosing, but to do well at the beginning, and then for the physicians to decide like, oh, well, should really high-need patients be on Eylea HD or should they be on Vabysmo or can they actually be on tarcorsimab or KSI-501. So we'll find out. But we're not going to be weak, right? And we're going to still have a very strong differentiating and longer interval durability drug.

With AMD, we hope we don't need 4 loading doses. Right? That's -- say AMD patients dry up with -- they're either going to dry or not dry with that 2 doses and that third dose is usually just bonus. But hopefully, you don't need 4 for AMD with this particular molecule.

Got it. Maybe also -- and I want to get to the trial in just a second as well. Formulation changes, could you remind us what was changed in the formulation to go from gel-like to fluid-like? And is there any prior data on that? I just want to understand the tolerability profile as well because of the formulation changes.

Well, the main thing is that the tarcocimab old clinical formulation had 100% of the antibodies conjugated to a biopolymer. So you had a high amount of, let's say, biopolymer mass in that highly concentrated formulation. Okay? So for example, the biopolymer is 5x larger than the protein. Okay? So if you have 5 milligrams of protein, you have north of 25 milligrams of polymer. And so for a total of 30. But when you bring that down, so that you only have 4 milligrams of conjugate, then you have, say, 20 milligrams of polymer. Right? So the point is that it turned out that we were a bit on the steep part of the curve, okay, where when you have biopolymer amount, and the impact on different biophysical criteria like dose time and dose force, so that as we marginally decreased the amount of polymer from 100% to 80%, okay, you actually come down that steep part of the curve very quickly, and you can have dramatic impacts and dramatic improvements. So in the end, whereas we leaned into the changes in the formulation initially from a manufacturing standpoint, as we moved to commercial scale in our Ursus facility because the drug substance was quite gelatinous and somewhat difficult to work with. And similarly, led to some complexities in the clinic as they did dose preparation, right, and as we talked about the dosing. But as you mildly decrease the biopolymer amount in the concentrated formulation, you do have a dramatic improvement in many different qualities. So that's what -- so it's going from the steeper part of the curve down into a flatter part.

Got it. And maybe just to wrap it up. KSI-501, I think you've shown just about 100 drop on the CSTs. I'm just trying to compare that versus Roche whether we should or shouldn't make that comparison? As well as on DAYBREAK, remind me, is the timing of primary endpoint measurement at week 48? And I ask because the way I'm looking at the trial design, because the way dosing is structured, Eylea last injection happens 8 weeks before the primary endpoint measurement, whereas for 501, the last injection happens 4 weeks prior to the primary endpoint measurement. Could that make the regulator and/or a physician ask whether 1 arm was just designed a little better than the other based on the timing of last injection?

So I'm going to answer your second question. So the primary endpoint is a combination of weeks 40, 44 and 48 in DAYBREAK. So there's -- it's a mix.

How does that work? So if I was a statistician, how would I do that? Just take an average of all 3?

Yes, exactly. It's an average of the 3 basis, which is a more fair comparison.

For all arms?

For all arms, compared. Correct.

And that's predicated. The Vabysmo studies do the same thing. There's many other sponsors that have done it. We actually think it's more fair.

It's a good more realistic comparison, as you say. Like instead of like how long after the last visit is the primary endpoint, this is a combination of the 3.

Okay. Got it.

And sorry. Could you repeat your first question about the 501?

On 501, the CST, the depth of CST responses versus what you saw on Roche?

I think that it's impossible to compare like between trials, like if that's -- it's -- yes. What do we see, like, what we see in the 501 Phase I data is that there's clinically meaningful improvement in BCVA that corresponds to that response in CST. That's what you want to see. It's a small sample size. So it's directionally as good as you could get. But it's really hard to compare to something else that you can see. It depends on the baseline characteristics and so on.

Victor, I guess, how do you think about that? Because I feel like if there's one thing I got really wrong on Kodiak first time and exactly this issue, CST was weak last time and it perfectly correlated with the Phase III problem. And I'm just kind of seeing that on the 501, and I'm just trying to make sure I don't get that. Clearly, DAYBREAK, you've adjusted for a lot of that because there were some changes. So I'm just curious about that.

Well, don't forget that the 501 now has the enhanced formulation technology. And so therefore, if there were an early weakness in OCT associated with 100% conjugate, now it's a 1.5 3.5 for 501. So we're bringing a lot of power and the protein itself is very potent, right? In fact, on the VEGF, it's now a VEGF trap on the 501 design rather than any anti VEGF-A antibody. Whether that's helpful in terms of early power or control is still up for debate. But I think Umer, we agree with that. That's why we brought the new technology in.

And the other issue, and again, it's hard to cross draw compare, but they're also different reading centers. Like the Vienna reading center, Ursula, is very different than the reading centers that were used. So that's closed now. And so now you're going to have more -- the new trials, I think, will be more -- they're moving to Wisconsin. Right?

Yes.

Yes.

Okay. Well, great. I think that's a bit of a tour de force of Kodiak R&D and now you know as much as we do. So we'll work hard on the clinical trials and their early pipeline. Thanks, Umer. Thanks, everybody, for coming.