Apellis Pharmaceuticals, Inc. (APLS) Earnings Call Transcript & Summary

Okay. Good afternoon, everyone. Thanks for continuing to join us at the Bank of America Virtual Napa Conference. I'm Tazeen Ahmad. It is my pleasure to have our next presenting company, Apellis. For the next 50 minutes, we'll be talking about all the exciting things that have happened, and we expect to happen for the company. Speaking for the company this afternoon will be Co-Founder, CEO and President, Cedric Francois; Chief Financial Officer, Tim Sullivan; and of course, Adam Townsend, who's Chief Commercial Officer. So gentlemen, good afternoon. Thanks so much for joining us.

Good afternoon. It's wonderful to be here, Tazeen. Thank you for inviting us.

So you've had some good news happen recently. Why don't you maybe start off, Cedric, with giving us a quick overview of the company, your recent approval? And then we can go into some more details on that and upcoming catalysts, if that's good.

Of course, thank you. So Apellis is a company focused on the complement pathways. Complement being a very ancient part of our immune system. And where in the past couple of years, us and others as well have discovered that by targeting this old part of the immune system, there are therapeutic opportunities in a wide range of indications across multiple therapeutic areas. What is unique about our approach in complement is that we target this very complex cascade of interacting enzymes, which the complement cascade is, by going after the nexus in this cascade kind of the central point. And the best way to think of the complement cascade is like an hourglass where multiple pathways converge on a central point. And from there, multiple effects emerge. And that central point is complement factor C3. Now with the approach that we have at Apellis by going after this nexus point, we are able to control all of the physiologically relevant downstream effects of complement, regardless of what the source of activation is. And as the company grew over the past 10 years, we focused kind of on 2 important therapeutic aspects of this approach. On one hand, we had a systemic program where we aimed to control C3 systemically in the whole body with subcutaneous injections and where our lead indication was a rare disease -- is a rare disease called paroxysmal nocturnal hemoglobinuria, or PNH. We then also have a program where we inject the same API intravitreally, so directly into the eyes of patients with the advanced dry form of age-related macular degeneration. So a lot of people are familiar with wet AMD, which is treated with anti-VEGF agents like Eylea and Lucentis. But the advanced dry form, which is almost as prevalent in terms of numbers, is a disease for which there are currently no treatments available. Approximately 5 million patients in the world are affected by this, 1 million in the U.S. alone. This has been a very complicated disease. I mean many things have been tried in the past, everything has failed. But based on a large Phase II clinical trial that we read out in 2017 and on which our Phase III program is based, we believe that we have robust data enhanced to be optimistic about a readout in Q3 for this particular indication. To kind of pull all of this together to understand Apellis, it's kind of easy to think about 2021 as a pivotal year for us, with kind of 3 important chapters. The first one is already behind us, and you mentioned it briefly, the approval of Empaveli in this rare disease called PNH, where we were fortunate to get a label that we believe will really benefit patients, and we'll talk more about that in the next couple of minutes, I'm sure. So that was very important for us and could not have been better we believe for patients and for us as a company as well. The next chapter is then at the end of Q3, we will have the readout of these 2 large Phase III clinical trials in more than 1,200 patients with geographic atrophy to find out if we can potentially have the first therapy to treat geographic atrophy. And then the third piece is actually coming up in a couple of days, in 2 weeks, are kind of all of the research programs that we have been working on for the past couple of years to go beyond PNH and geographic atrophy, and truly a horizontally and vertically integrated company, where we take maximally advantage of C3, not just in PNH and GA, but many therapeutic areas with new therapeutic modalities as well. On June 30, we will have an R&D Day at the Apella in New York to which everyone is invited and where we will discuss at length what we believe we can do in the next couple of years.

Okay. Great. So a lot to talk about. So let's start with the first drug approval that you recently received, as you said, for PNH. The drug is called Empaveli. Can you talk to us about the commercial preparations that you undertook heading into the PDUFA as well as what to expect now that you're approved? Investors are keenly excited about the launch, but also want to get every clue that they can from you on how to think about ramp rates, et cetera?

Yes. So we have our Chief Commercialization Officer, Adam Townsend, with us, who has been with us already for, I think, 3.5 years now. And that in of itself is an indication of how seriously we took the preparation for the commercialization of this product. For us, this is more than just about generating revenue for the company. It is a real mission to make this drug available to as many patients as possible as quickly as possible and in as many countries as possible. And Adam will now briefly talk about how he went about that.

Yes. Thank you, Tazeen. Thank you, Cedric. So yes, as Cedric said, we've been preparing diligently pre-approval, and we've been building great relationships with PNH doctors and also the PNH community. And obviously, COVID hit in that pre-approval time. So we very quickly pivoted to digital interactions. We started to get the name of Apellis out there. And we started to communicate and build relationships and that's continued into the launch. So obviously, we were very happy with our broad label for the treatment of adult patients with PNH. And we're also happy with how physicians responded to the label. They were very impressed with the PEGASUS data and the efficacy and safety of this product. And we've seen that continue into the commercial phases of the launch. Now our sales force have been in place pre-launch, and they're now out and about seeing physicians where appropriate face-to-face, but also interacting with physicians digitally. The launch is going just as I had hoped. We're beating my internal expectations where we should be at this time. We've had great interest from physicians. Our REMS process is up and running and giving us no issues at all. And we're getting a regular cadence of interactions. So I'm excited to say we're exactly where we are as we should be. We're right in the early phases of this launch. One thing you asked, Tazeen, was what do we expect from the patients, et cetera? Well, it's a rare disease. So we expect the hardest-to-treat patients to come on to our product first. Those patients with the highest unmet need. And that was something that we learned in our research prior to approval, and that's something we're seeing now we have Empaveli within the market. So generally, we're finding those unmet needs being the patients that are being pushed into the systems, and we're finding that those patients have a low and falling hemoglobin, and they're continually having to have their hemoglobin topped up by transfusions. And that's about 1/3 of the 1,500 C5-treated patients that are in the market. We expect that to be the initial bolus of patients that have a conversation about Empaveli. We also expect to have conversations about treatment-naive indication and treatment-naive patients as we progress. But the unmet need exists in the market, and we expect those to be the first wave of patients that we get.

Okay. And your label did include naive patients, even though your data didn't come out until about a week or so after you got the label. So in this environment, that's very impressive as it appears FDA is becoming or has become a little bit more particular about what it's looking for. So with that data set now in hand and with the label being as broad as one could have expected, how should we think about what the split of which patients versus naive patients would be initially? And then what should it moderate to over time?

Yes. Great question, Tazeen. So based on our data and some clinical research, we believe there are 1,000 -- approximately 1,500 C5-treated patients. So patients who are currently on Soliris and/or ULTOMIRIS and about 150 to 180 treatment-naive patients. As I said previously, we expect the unmet need patients, those patients who aren't doing quite as well with their PNH disease to be treated first. And they tend to be within the 1,500 C5-treated segment. So we look at that in thirds. So 1/3 of the patients of those 1,500 have the highest unmet need. And we are seeing those patients transition now to Empaveli as we expected prior to launch and as we've seen in reality. So those patients have a low and falling hemoglobin, continually require transfusions to top that up. And that is the patient segment that the physicians identify the easiest. They can find those patients and they see those patients more frequently. So that will likely be the first wave of our transition to Empaveli. Then physicians have told us they will transition through the next highest unmet need. So these are patients that still have a low hemoglobin and they still have the occasional transfusion, but they have all the signs and symptoms of PNH, fatigue, lethargy, all of those type of activities. And again, we expect those patients to be the next wave transitioning on to Empaveli. And finally, the final 1/3 of the 1,500, they have a closer to normal level of hemoglobin, but what's happening behind the scenes is their bone marrow is working overtime to maintain that. And we expect that to be the next wave of patients that physicians have a conversation about. Now we do expect and have already had conversations on patients who are treatment-naive, submitting all the right paperwork, et cetera, to start Empaveli. But let me just give you this quick analogy, which we got from a physician, right? If a physician in a community practice has 2 patients, if he or she wanted to prioritize those patients, 99% of the time, they would prioritize a patient who has a low hemoglobin or is really, really using a lot of transfusions prior to starting a treatment-naive patient. So -- the broad label is great, our data is superb to support it, and now we have print. We'll have a lot more to talk about in the treatment-naive indication, but we believe that the C5 and the switch population will likely be the first to start.

Okay. Great. So Adam, a few minutes ago, you said that the launch is actually going above or better than what your internal expectations would be at this stage. So can you share with us any kind of qualitative comments about what portion of this launch did you anticipate was going to be easier or harder than it is so far?

Yes. It's a great question. I'm thrilled with how the Apellis team is helping raise and elevate the standard of care in PNH. And we had some assumptions prior to Empaveli approval about the dynamic of getting physicians to sign up for REMS, for example, or how we would help patients transition from an infusion site or -- to a self-administration at home. And I'm pleased to say in all of our interactions over the last 3.5 weeks of launch, all of those things have gone better than we had potentially anticipated. So our REMS process are going through, physicians are easily getting certified. It's -- we haven't seen any major hiccups with that type of process. And also, we're getting really positive feedback from patients once they've been trained on the -- at-home or wherever they want subcu infusion, we're getting great responses such as, "I just saw the other day, this is so easy" or "This is much easier than I anticipated." So those were some of the perceptions and assumptions we had prior to launch, and I'm glad to say as we are executing, we're finding that we're moving through those without any headwinds at all. So things are progressing well.

Okay. Good to know. So we will look forward to your next earnings call, and I think people might have a question about what type of details do you expect on the early metrics. Will you be giving out patients on drug? Will you be giving out specifics about scripts, et cetera, that, obviously, we all use to try to project forward? So any sense of expectation that we should have on that?

Yes. It's a question we get asked a lot. More to come as we get closer to earnings, but one thing we've been clear on is we really wanted to work through the initial phases of the launch. So not a surprise, right, with the second or third to market within PNH. So I'll give you an example from the payer world. A lot of payers will put second or third products to market on new-to-market blocks or will have to go through P&T committees and approvals to get through. And that also has a way of -- once we're through that, we can have a real open and transparent conversation about being able to switch and transfer patients. I'm pleased to say all the payer interactions are going very well. But over the next couple of months, we are hoping to accelerate some of those formulary inclusions to allow us to get access to the full PNH treatment population. I think once we've worked through that and the early phases of launch, we'll start to have a good stable view of how the launch is progressing. The early indicators are very positive. We still have -- we're still only 3.5 weeks in. So there'll be more to come in the earnings call, but we've got a lot of hard work to do to get us to allow all those 1,500 patients -- C5-treated patients and the 150 treatment-naive to get access to this drug.

Okay. Excellent. I'll give Adam a little bit of a rest now and move on to GA. So this is an event that I get asked about pretty much every call that I do with investors now your upcoming readout of your pivotal study in geographic atrophy. So maybe one thing that we can start off with, Cedric and team, is that many companies have tried to develop treatments for GA and both within complement and outside of complement. So can you give us a sense of what it is that you think these other companies weren't able to do correctly? And how did you get confidence in pursuing C3 in particular to treat this particular patient population?

Yes. Thank you, Tazeen. So your comment is, of course, on the development of drugs in geographic atrophy has been incredibly hard with lots and lots of failures behind us, unfortunately. And before I kind of give a little bit of an explanation here, it's important to point out that our confidence is rooted in data. So the Phase II clinical trial, especially in much of the other work that we have done, is what gives us confidence going into the readout of the Phase III now in just in a couple of months. The story of complement in macular degeneration is interesting, right? It came out of kind of the first breakthrough out of the genome-wide association studies in 2005, established an association between complement and macular degeneration, but it was not very well known really at the time and still not, right? I mean, how is complement really involved in this disease. But like with everything, we also learned from our failures. And if you put all of the available evidence to us today in one guiding hypothesis, it comes together in the following, which explains why we believe the data that we have generated are real and why we believe we have a very good chance of success with this Phase III clinical trial. C3 as a protein, one should think of like paint. It flows throughout the whole body, and it wants to continuously bind to every cell surface in the body, coherently like paint. And every cell in the body has a job to do, which is to, first of all, inactivate this highly reactive protein on your cell surface so that it doesn't amplify and multiply. And then once it's inactivated, to internalize it and process it lysosomally and get rid of it. And you can say, well, that's a lot of work, why do we do this to ourselves? And that kind of goes back to this ancient form of immune surveillance. Cells that are sick or that are infected and that do not do a good job of cleaning up that paint from the cell surface, they become visible to monocytic cells like macrophages and they become a target for phagocytic removal. And it is that precise process that, quite frankly, goes wrong in PNH, but in geographic atrophy, specifically in the retina becomes disregulated. Because what's unique about the retina and to some extent, by extension to neurodegenerative -- many neurodegenerative conditions as well, is that the mechanisms required to clear that C3 product from the cell surface are the same cellular mechanisms that are needed to take care of the visual cycle. So if you're a retinal cell, you have a job to do, i.e., vision, and it competes with your cleanup job of that C3 product. And at some point, that competition comes to a clash and certain cells will start neglecting the cleanup of that C3 product from the cell surface. And like in the rest of the body, in the retina now become visible to these monocytic cells. And geographic atrophy doesn't start across the retina. It starts in one, sometimes a few specific spots in the retina. And from there, like a forest fire, like a wildfire, expands from there. So there is something that when there's a few cells that tip the balance, it kind of infects or affects the neighboring cells and from there, it spreads, and they all kind of go over the cliff subsequently. To correct for that mechanism, it is very important not just to reduce the amount of C3 that gets deposited so that these poor cells can start focusing on the visual cycle again, but to do it to such an extent that the cleanup capabilities that these cells have overwhelm the amount that gets deposited. So it's not just enough to slow down the deposition, you need to flip the switch like in bookkeeping. You need to go from building credit to building a surface, right? That's the -- and when you look at the failures specifically in complement in geographic atrophy, there are 2 specific complement targets that stand out. The first one is complement factor C5, which was tried with an antibody both systemically as well as an antibody administered intravitreally and failed twice. Now C5, and we know this very well from PNH, of course, has no effect on the deposition or the cleanup of C3. So hypothetically, that makes sense. It's consistent with this hypothesis. The other one is a little more complicated. Lampalizumab, which was the Roche product that went through 2 very large Phase III clinical trials, close to 2,000 patients, failed. And where you could say lampalizumab does affect the amount of C3 that gets deposited which is true, but lampalizumab is by -- exclusively an inhibitor of the so-called alternative pathway. And it leaves the classical pathway exposed and that leakage of C3 deposition, we believe, made it impossible for the cells to kind of flip over, if you want, from continuing to accumulate to actually starting to clean up. So it is the mechanism of pegcetacoplan, which controls all of the pathways and does it in an enzymatic fashion that we believe made the difference.

Okay. So -- how are you thinking about what the -- what level of efficacy you'll be able to show in the study? There -- where is the study powered to show? And is there going to be a difference between simply being statistically significant on efficacy, and we'll talk about safety in a second, versus being statistically significant and clinically meaningful?

Yes. So that's, of course, a very hot topic we face, right? But we go back to what we had in FILLY, where we showed a 29% reduction over the course of 1 year in the growth of atrophy in patients with monthly injections and 20% in patients dosed every other month. We have done the commercial work around that and the medical work around that. And on average -- and I will hand in a minute over to Adam to shed some color to that, on average, the 20% seems to be the clinically meaningful cutoff for retinal specialists. Now it is important to note that in the Phase II clinical trial, we -- that result of the primary endpoint was based on the mixed effect model for repeated measures. That is essentially an algorithmic function to compensate for missing images because the way in which we measure the primary end point in GA is to find out if the growth of atrophy in these patients can be slowed down anatomically by taking images of the back of the eye and looking at how it grows and seeing if whether on treatment that grows slower compared to the sham. In the FILLY trial, we had images at 0, 2, 6 and 12 months and then another one at 18 months after we stopped treatment. So not that many images. So this mixed effect model actually would penalize us a lot in the Phase II clinical trial. In the Phase III clinical trial, we have many, many more images and that matters, because if you really look at the observed data in the Phase II clinical trial, we were actually around 35% and 25% for monthly and every other month, respectively. So we believe that we are in a very good place. The 2 Phase II clinical trials were powered more than 95% to show that difference in the monthly arm. And with kind of the repeated measurements that we do and the more images that we have, the variability also goes down, which means that we use the powering based on the variability from the Phase II, but we have less variability in the Phase III. And maybe one word from Adam as it relates to the physician feedback that we have received on the clinically meaningful differentiation in GA growth.

Yes, absolutely. So yes, so we did some research with retina specialists, key opinion leaders and ophthalmologists. And we believe that most physicians believe that 20% less lesion growth over 12 months is clinically meaningful. But we got that consistently from our research and also at our recent GA Day, it was reiterated by various KOLs. We've got some great quotes within that research. Anything over 20% would be significant from a retina specialist, for example. I'd rather see upwards of 20% less growth, but I think 10% would be reasonable was another quote that we got. So clinically, meaningfully, it's about 20%. And so yes, we expect that to continue, and we're really looking forward to seeing the data very soon.

Okay. Great. So let's maybe talk about safety observations that have been made to date. So I think there is some difference in nomenclature about what's been observed through Phase II and what to expect in Phase III. So is it the case that the drug is inducing wet AMD or not? I think that's a question that a lot of people are still not clear about.

Yes. So to interpret that question, it's important to talk about what wet AMD means, Tazeen. And I want to start with that because it is -- there's a very important difference between CNV, which is choroidal neovascularization, and exudation, right? Choroidal neovascularization means that you have new blood vessels in the retina, where there is not supposed to be new blood vessels. Exudation means that these blood vessels are actually leaking. And the reason why that matters is that in patients with geographic atrophy, you can have new blood vessels present in the retina at the start of a GA study, for example, but these new blood vessels are not leaking yet. So with the conventional imaging techniques, they wouldn't show up as wet AMD, even though we have a membrane that predisposes these patients to be more exudative. It is important, point number one. And that is what we really observed in the studies that these exudations that we saw were primarily driven by that phenomenon. The second piece, which is really important here is that there's a bit of a misunderstanding that sometimes from the investors side that the occurrence of wet AMD, i.e., exudations is an abnormal phenomenon in geographic atrophy. That is not true. So exudations are relatively common in patients with geographic atrophy. We presented a study on 69,000 patients that we did retrospectively at AAO in a late-breaking abstract this last year in 2020. And what we saw there was that over the course of 2 years of follow-up in patients with pure dry geographic atrophy, so neither of the eyes affected with exudations, 8% of patients will develop exudations naturally. If you take patients that have already exudations in 1 eye, and you take the contralateral eye with geographic atrophy, that number is actually 22%. So that's not uncommon at all. And when you think about the patients that we enrolled in FILLY, we had 38% of the patients that were enrolled that came from that demographic of starting off the study with already wet AMD in 1 eye and geographic atrophy in the contralateral eye and a natural risk of 22% of exudations over the course of 2 years. Bear in mind, of course, that it was a 1-year study, right? Now the reason why I mentioned that is because the nature of the exudations that you observe also matters a lot. To put it simplistically, you can have exudations all the way from what we call classical CNV, where you have very fast leakage that rapidly leads to blindness all the way to very small leakages that arguably don't really need to be treated and can be observed. In the FILLY trial, of all the cases of exudation that we saw, there was not a single case, 0 of the classical CNV variant. So everything that we saw fell into that mild category of exudations that are not associated with significant vision loss, which is why in the Phase II clinical trial, we could continue the study and which why in Phase III, this is -- we had to change nothing about our patient population that we enrolled. Now getting back to what I would suspect your next question to be, which is how does -- how do we expect or why did we see then this imbalance in the Phase II clinical trial? Because that's really what stood out. I mean the monthly much more than every other month much more than the sham control, which automatically would ensure that this is drug induced. We think that there will be, in all likelihood, a bit of an imbalance between the monthly versus every other month than versus sham control. But we think that, that particular phenomenon was vastly exacerbated in the Phase II clinical trial by investigator bias. So we had a couple of sites, and it's important to point out that in the Phase II clinical trial, the investigating physician, the physician that administered the drug product was not masked because that is actually a real injection versus the sham, which is not a real injection. The readers read the images, those were masked. But these treating physicians knew who was getting drug product. And when they started seeing exudations, they started specifically looking for those, and that investigator bias was for sure they're in the Phase II clinical trial. In the Phase III clinical trial, where we do double masking, that bias will be taken out of the equation. So that's an important correction that we believe will lead to a reduction in the number of exudations or the imbalance that we saw between the 3 groups in the Phase III. The second aspect, and this is very important, is going back to that number of 38% of patients that came in to FILLY with wet AMD already in 1 eye and a very high predisposition to developing wet AMD. That 38% compares to an overall patient demographic of approximately 20%. So we had approximately 2x as many patients in our study that came from this high propensity category compared to what you would naturally expect. And the reason for that is that when we ran FILLY, we were competing with Genentech/Roche who excluded those patients. Now in the Phase III trial, we still enroll these patients because they need treatment. And the physicians gave us the feedback that they want to treat these patients, but we will probably have far fewer and probably much closer to 20% than the 38% that we had in Phase I.

Okay. So based on these changes that you've made, you're expecting to see lower incidence of exudative events in DERBY versus FILLY. Is that right?

That's correct.

What is sort of a reason you end up seeing similar events? What has doctor feedback been thus far on the overall profile of the drug just based on FILLY? Would it matter if it's closer to what FILLY shows?

So the answer is it would not matter, right? Because -- and there we go down to the nuance. One thing is to talk about it in the context, of course, with investors or with relatively uninformed KOLs. But the rubber hits the road with investigators, right? I mean I always think of clinical trials as a pretty good harbinger or predictor of what you can expect in the real world. Our Phase III clinical trial, A, as I already mentioned, did not exclude patients that come from this high-risk category and physicians are just as eager to treat them as they are patients with pure geographic atrophy that gives you a bit of context. Number two, the physicians that we have in the Phase III clinical trial have -- sorry, I have a little bit of a disruption. Sorry, yes, thank you. Can you hear me?

Yes, we can hear you.

Okay. Sorry, can you repeat the question, please? I got distracted by my screen that went off.

Yes, yes, no worries. So just in terms of what would the impact be if you ended up observing actually events that were similar to FILLY and the DERBY trial, even though your expectation is that you would have fewer such events? Like what have doctors told you about whether or not that's still something that's attractive?

Yes. Sorry. So I was saying sort of the investigators who have seen the data, A, wanted to enroll these patients, including the patients that are at high risk of this; and secondly, would have enrolled this trial faster than any Phase III clinical trial before that it hadn't been [indiscernible]. So the appetite in the Phase III with the investigators having seen the data, that is what gives us confidence that even if it were to be the same exudation rates, which we do not expect, it would have been acceptable product profile for patients.

Okay. Great. So nearly all of my companies have talked about some potential impact from COVID in their studies, either at the enrollment phase and more importantly that it caused some people to maybe -- if they're taking daily doses of something to miss a dose or if they're supposed to go in to be monitored at sites to miss visits. But as it relates specifically to your trial in GA, is there any concern about patients missing doses and this would be something that you'd be able to know in real time assuming -- I'm assuming relative to, let's say, a company that was just monitoring to see if patients were taking pills every day? So built into your powering assumptions, did you make a certain estimate of how many patients would be missing doses? And if so, how has that turned out relative to that estimate?

Yes. So we -- so needless to say, we, of course, had a lot of missed injections in the Phase III clinical trial because of COVID. But we also had a lot of missed injections in the Phase II clinical trial in FILLY. So -- and a twist of irony, if you want, right, in the Phase II clinical trial, when patients developed exudations, we actually stopped giving the drug product to these patients. And so we had a lot of missed injections in the Phase II clinical trial, just like we had missed injections because of COVID in the Phase III clinical trial. And the pattern with which these missed injections occur can be studied in multiple matrices. In other words, if you have 10 missed injections, it's different when it's in 1 patient compared to multiple patients, it's different whether they are sequential or interrupted with more injections, et cetera. So we have done that matrix analysis very thoroughly, and we feel very confident about the readout of this Phase III clinical trial in spite of the missed injections that we have had. And without spending too much time on the detail behind that, one thing that is very easy to wrap your head around is that we had statistical significance in a much smaller study, right, 246 patients compared to the more than 600 patients in each of OAKS and DERBY where every other month was already statistically significant as well with missed injections there as well on the every other month, which is half of the injections. So again, it would take me a full hour to kind of take you through all the detail that we do. We track this very closely. We were very fortunate in the sense that most of our sites are private sites, they're not in the hospital setting. So patients may have missed 1 or 2 injections, but then they essentially went back to these clinics with what we at the time called white plastic gloves service to make sure they felt comfortable going to the clinic and going home. And once they were back in the routine, it was actually very good. So we have no concerns around the missed injections based on COVID.

Okay. Before we leave the topic of exudation that we started a few minutes ago, I wanted to ask your thoughts about the potential theory that the reason you're seeing exudation is because of the PEGylation of your compound. Is there any science behind that? And what is your view on whether or not PEGylation is involved in causing these side effects?

Yes. So it's important to note here that the preclinical research because these are all animal experiments, right, or in vitro, in vivo experiments that are conducted around the role of PEG in neovascularization and exudation are all in models that are completely irrelevant to macular degeneration. So you have artificial setups such as what we call laser-induced neovascularization where you shine a laser and then you get the formation of new blood vessels, or you can have something where you have kind of a model of retinitis of prematurity with high oxygen levels, et cetera. Those are not relevant to the real world. What's important in the real world and in patients is that this is not by any stretch, the first PEGylated molecule to be injected in the eye. I mean, we can go back as far as Macugen, which was PEGylated -- other molecules that were PEGylated where there may be a nominal increase of exudation, but definitely not something that emerged. And the last piece that I will mention there is that, ironically, the mechanism by which these exudations from PEGylation are described in this research is through a complement, which we actually control. So this is not a mechanism that we believe is relevant here. What we do believe, however, is that going back to the hypothesis of these pre-existing neovascular complexes in the retina that become more leaky, we believe that, that is part of a wound-healing response. And wound healing is associated with the secretion of VEGF, which could account for this increase in exudation. It all comes down to the point is this clinically relevant and important or not. And it is not considered a safety concern from the physicians in our trials or from the FDA's perspective.

Okay. Now DERBY and OAKS are, I believe, double-masked, right, while FILLY was single masked. How are you thinking that this could have any -- if that all impact on exudation events in the studies relative to FILLY?

Yes. Thank you, Tazeen. So this is what I touched a little bit upon earlier as well. It is -- again, if you are the treating physician and you know which patients are receiving the real drug product, it is the thing to go and look more at the OCT images of the patients that are actually getting the real drug product compared to the controls. It's a natural thing to do. And it's important to note here for those people not familiar with SD-OCT, so spectral domain OCT imaging in the retina, is this is an amazing technology, right, where at a micron level, you can actually see small exudates occur in the retina. And when you treat a patient with anti-VEGF, even if it's a small exudate that exudate will actually go away, right? So now we know from the concentration in certain sites that there was an investigator bias. And we also know from the retrospective analysis that, A, many of the reported exudates that we had in Phase II, we could actually not find back in our imaging data base. And conversely, many patients, including in the sham control group, had exudates that were not reported as exudates and that on subsequent images were no longer visible. So again, going back to the point that the presence of small exudations in the retina is a natural phenomenon that is especially more pronounced than patients that have already wet AMD in their contralateral eye.

Okay. Now you've talked about the advantages of focusing on C3 versus the other companies, Novartis, MorphoSys, Roche, et cetera, that focus on other parts of the complement cascade. There is another company that's also trying to focus on C3 in GA. And I'm interpreting that as a validated -- validating the focus on C3 when there's more than one approach on the same target. But at the same time, just based on what you know now, how do you think that Apellis's approach would be differentiated from this other company or anyone else that might also want to focus on C3?

Thank you, Tazeen. So there's, of course, we're talking about NGM with an antibody against complement C3. Look, the -- for the benefit of patients, what we need is a therapy, whether it comes from us or anybody else. So if ours doesn't work, I certainly hope that theirs will work. So I'll just focus on kind of an important difference between our approach and the approach with an antibody. What we believe is probably essential to the power of pegcetacoplan is that even though we call it a C3 inhibitor, it is really a C3 controller. When -- and let me briefly elaborate on that. The complement pathway is more than 30 interacting proteases. So serum proteases are predominantly that work with each other and that enzymatically enhance each other. When you try to take one factor in there and control all of the effects, it is very difficult to do that. So what we have with pegcetacoplan is a molecule that, yes, it binds to C3, but it binds, and this is very important, specifically to the C3b unit, which is the activation product of C3 on that C3 protein. And by doing that, what you do is, on one hand, prevent C3 by binding to that pocket from binding to cell surfaces, very good, right, which you would expect an antibody to do as well. But then once it is actually bound in the form of C3b and where it wants to assimilate or assemble the convertases, it prevents assembly of the convertases, both the C3 convertase and both of the C5 convertases. So you get kind of a 4-pronged point of attack where you have an enzymatic dampening of complement activation that happens rather than trying to remove all of the C3 product. And why is this so fascinating because in PNH, for example, we see our maximum pharmacology when patients have still 10% to 20% of free-floating C3 circulation. So it's still there, but it's under control, which may account for kind of the remarkable safety that we have seen in PNH with a systemic control of C3. Now NGM's antibody was specifically engineered to not bind to C3b. So it kind of takes 1 of the 4 pieces of activity of pegcetacoplan rather than all 4. And we'll have to see if that's going to be [indiscernible].

Okay. So now if both of the Phase IIIs are positive, and we're keeping our fingers crossed that they will be, what would be the approach to applying for approval after that?

If both monthly and every other month, you mean or...

Yes.

Yes. Okay.

Both -- I'm sorry -- yes, if both studies are positive?

Both studies are positive, okay. Yes.

Yes. So how would you think about the approach to applying for approval? And what kind of label you would like to ask for?

Yes. So that is something we're not going to give details yet on the type of label that we are pursuing. Our first objective right now is to get the first therapy for this terrible disease to market. What we need to do is show that we can reduce the growth of atrophy, which equates to further receptor cell loss in the monthly dosed individuals. That is in a nutshell what we need to show first and foremost. In the United States, with the FDA, we know that, that is sufficient for approval after 1 year of investigation. So if in September, we were to show that with the monthly dosed individuals, we have statistically significant reduction in the growth of GA, we will have sufficient grounds, we believe, to file for approval. In Europe, it is a little bit more nuanced. In Europe, there's an expectation to not statistically, but to show directionally that you also have an impact on the function. So in these studies, we measure reading speed in patients. And we also, in 1 of the 2 studies in OAKS, look at microperimetry, which is to essentially with a laser find out if there is real light sensitivity around the fovea where typically this phenomenon of GA occurs. Importantly, again, we don't have to show something statistical. It needs to be directionally in the right direction. But the European authorities and just with all the work that we're doing now, this is increasingly becoming clear is also very well cognizant of kind of the evidence or the evidence point here, which is that you lose photoreceptor cells every time your area of GA expands. And that will ultimately affect your vision.

So do you think then that the applications for Europe and the U.S. would be staggered with maybe the U.S. first and then Europe a little bit later?

That's an appropriate expectation. Again, we will look at the data and as it gets generated, and then make a determination as to whether we want to go to Europe with 1-year data at hand already or not or whether we want to wait for the 2-year functional readouts for Europe, while applying in the U.S.

Okay. A minute ago, you said September. So is that one we should expect to see the studies read out, I think your previous guidance might have been 3Q?

Yes. I was going to let that cat out of the bag in 2 weeks. But yes, September, it is.

Okay. So what level of information should we expect then at the top line?

So that is something that, in 2 weeks, we will talk a little bit more about. So that's one of the important things that we'll discuss at our R&D Day is the type of information that investigators or investors should expect, not investigators, from the top line.

Okay. So what do you think is the most -- I mean, what part of this program is the riskiest to you? Like you've done everything you can to design the study as well as possible, pick the right patients, make the right changes from Phase II to Phase III. Is there anything at all that's still in your mind, something that you worry even a little about, is there anything like that?

What I'm particularly proud about with this program is that if these trials fail, it will be because the drug does not work.

Yes.

And that is -- I mean, it's always supremely sad, but it would be especially sad in this very high unmet need that you would get derailed by the logistics of running these very highly complex trials by making design mistakes, et cetera, or manufacturing mistakes. We've seen all of that before in the past, right? I mean now with only -- with less than 3 months to go, we're like feeling very good that if the trial fails, it will be because the drug didn't work. And we already know that it works really well in PNH, so even there, the pharmacological checkbooks has been made.

Okay. Excellent. Well, we're obviously looking forward to that. We will be excited to learn about what other indications you want to pursue at your Investor Day. Thanks so much for spending the last 50 minutes with us. And we'll look forward to seeing updates from your earnings call as well as now in September for the GA study. So thank you, Cedric, Tim and Adam, for your time. Really appreciate it.

Thank you, Tazeen. It was excellent.

Thank you, Tazeen.

Thanks, Tazeen.