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

Good morning, everyone, and welcome to the Apellis Pharmaceuticals 24-month Phase III DERBY and OAKS Results Conference Call. Today's call is being recorded. At this time, I'd like to turn the call over to Meredith Kaya, Senior Vice President of Investor Relations and Strategic Finance at Apellis.

Good morning, and thank you for joining us to discuss 24-month data from the DERBY and OAKS Phase III studies evaluating pegcetacoplan in patients with geographic atrophy or GA. For those participating via conference call, we have made the slides available via webcast. A replay of this call will also be available on our website following the call. Before we begin, I would like to point out that we will be making forward-looking statements that are based on our current expectations and beliefs. These statements are subject to certain risks and uncertainties, and our actual results may differ materially. I encourage you to consult the risk factors discussed in our SEC filings for additional detail. On today's call, I am joined by our Chief Executive Officer, Dr. Cedric Francois, our Chief Medical Officer, Dr. Federico Grossi; and Dr. Nora Lad, Associate Professor of Ophthalmology and Director of Ophthalmology Clinical Research Unit at Duke University Medical Center and lead principal investigator for the OAKS Study. Adam Townsend, our Chief Commercial Officer, will also join us for the Q&A session. Tim Sullivan, our Chief Financial Officer, is unable to join today due to travel conflicts. Now I'll turn the call over to Cedric.

Thank you, Meredith, and good morning, everyone. Today marks another important milestone for us at Apellis as we seek to bring the first-ever treatment to patients living with geographic atrophy, or GA. On this call, we will be reviewing the 24-month results from our Phase III DERBY and OAKS studies evaluating pegcetacoplan, our targeted C3 therapy for the treatment of GA secondary to age-related macular generation or AMD. Fede will review the data in more detail momentarily, and we are fortunate to have Dr. Nora Lad here today to share her perspectives. But first, let me provide a few introductory comments. GA leads to the destruction of the retina and irreversible blindness, and we are thrilled with what these 24-month results could mean for patients. Building on our previous data, the 24-month data showed that treatment with pegcetacoplan group line resulted in increasing effects over time. Specifically, between month 18 and 24 in the combined studies, we saw an acceleration of the treatment effect with pegcetacoplan, reducing GA leading growth by 30% and 24% in the monthly and every-other-month arms respectively. But simply, the longer patients who are treated, the better they responded to the drug. For a drug to demonstrate not just durable but increasing effects over the study period is incredibly rare. This means that patients with GA could save more and more of their photoreceptor cells each year resulting in substantial slowing of their GA lesion growth and the potential to preserve their vision for longer. Importantly, we also continue to see a favorable safety profile in line with what we saw at 12 and 18 months. There were no new serious adverse events of ischemic optic neuropathy or new cases of endophthalmitis between 18 and 24 months and new onset exudations were consistent with longer term exposure. The key secondary functional endpoints were also assessed in the past 24 months. As expected, there were no clinically meaningful differences observed. As we have mentioned before, and as you'll see in the data, we believe this is due to the challenges with measuring functional vision in GA patients as well as the relatively short period of time over which they were assessed. We anticipate a separation to occur over longer periods of time. The 24-month results will serve as the basis for our European application for pegcetacoplan, which is on track for submission by the end of this year. In the U.S., we are looking forward to the upcoming PDUFA date of November 26. If approved, pegcetacoplan will be the first and only treatment available for patients with GA, setting a new standard for GA moving forward and providing Apellis with the opportunity to establish the market where nothing has existed previously. The potential product profile for pegcetacoplan is compelling, including first, robust efficacy and safety data in more than 1,200 patients over 24 months; second, flexible dosing options with both monthly and every-other-month treatment; and finally, increasing and clinically meaningful effects across a broad, highly representative patient population including both extrafoveal and foveal lesions. In fact, between months 18 and 24, foveal lesion growth reductions were comparable to reductions in extrafoveal lesions in the combined studies. More than ever, we believe pegcetacoplan represents a breakthrough for patients globally who are living with GA. We look forward to sharing more detailed data at upcoming medical meetings, including at the American Academy of Ophthalmology next month. Let me now turn the call over to Fede to walk through the data.

Thanks, Cedric. Before we review the data, let me first quickly remind you of the study design. DERBY and OAKS are double mask randomized control global Phase III studies comparing the efficacy and safety of monthly and every-other-month intravital pegcetacoplan with some treatment in more than 1,200 GA patients across more than 200 sites worldwide. The primary efficacy endpoint of those studies was the reduction in growth of GA lesions as assessed by fundus autofluorescence. The studies continue for a total of 24 months with patients remaining in their randomized mass treatment groups. In addition to our earlier 18-month analysis, lesion growth was assessed at month 24 as a prespecified secondary endpoint and key secondary functional endpoints were also assessed. Let me now review the data. I will start with the effects of pegcetacoplan at the 6-month intervals. These are the most important data in this analysis as they show how the treatment benefits evolves over time. As we know, GA is not a 1-year disease. So it's critically important to understand the effects of pegcetacoplan at various time points over the full 24 months. Particularly, the last 6 months interval is important because the treatment effect during this period tell us about how the treatment is benefiting patients today and what can we expect as the base benefit moving forward. As you can see on this slide, treatment with pegcetacoplan resulted in increasing effects over time in both DERBY and OAKS with an acceleration in the reduction in lesion growth, specifically between month 18 and 24 as compared to previous periods. This increase in effect becomes even more clear when looking at the combined data from DERBY and OAKS as shown on this slide. With rate between months 18 and 24, increasing to 30% and 24% with monthly and every-other-month treatment, respectively. The potential magnitude of the benefit to patients could be substantial. If this increase in effect continue, it could mean that the curves for the treament arms have the potential to flatten over time. The acceleration in month 18 to 24 was driven by slower rates lesion growth in the pegcetacoplan treatment arms and not by faster growth rates in the sham arm. Sham was linear for the duration of the studies with highly consistent growth rates of approximately 1 square millimeter for 6 months over each of the 4, 6 months intervals. When we look at the effects in both extrafoveal and foveal lesions within this 18- to 24-month period, we found that pegcetacoplan appears to be equally affected across both extrafoveal and foveal lesions, not only in the monthly treatment arm, but also in the every-other-month treatment arm. These are very encouraging results, further underscoring that pegcetacoplan works in a broad patient population regardless of lesion location and reinforces the importance of early and consistent treatment. We will share more detail in extrafoveal and foveal data at an upcoming congress. The increased effects over the 24-month period was also translated into greater reductions in lesion growth from baseline out to month 24, with all nominal p-values less than or equal to 0.003. What is more striking about these graphs is the treatment curves for both DERBY and OAKS. We now have 2 independent studies that are behaving similar in terms of overall effect size at 24 months in the monthly arm as well in the every-other-month arms. In DERBY, treatment with pegcetacoplan resulted in a 19% and a 16% reduction in lesion growth in the monthly and every-other-month arms, respectively. In Oaks, there was a 22% and 18% reduction in lesion growth with monthly and every-other-month treatment, respectively. This means that a growing amount of retina tissue is being preserved across both studies over the 24-month period. If this effect continue over longer periods of time, the amount of retina saved could have a substantial impact for patients. As part of this analysis, we also assessed the key secondary functional endpoints. Because 2 years is a relatively short time to assess functional benefits, as expected, that was not clinically meaningful or a statistically significant difference between pegcetacoplan and sham at 24 months on the key secondary facial function endpoints. Today, I will summarize the best corrected visual acuity or BCVA results from the combined studies. More on functional endpoints will be presented at an upcoming congress. As you can see, there's variability across all 3 arms over the 24-month period. The monthly and every-other-month treatment arms performed relatively in line with sham at various time points, with sham bouncing back in month 22 and 24. This variability further underscores that 1 single touch point may not be adequate to evaluate functional benefit in GA due to the high variability. Due to this change in the last 2 months, we observed a loss of 1 liter in the monthly arm and less than 2 liters in the every-other-month or at 24 months as compared to sham. This difference is not clinically meaningful. We believe we will see a benefit over longer periods of time as the reduction in vision growth translates into visual function. We look forward to following our GALE extension study where patients are treated with pegcetacoplan for an additional 3 years. Turning to safety. Pegcetacoplan continued to demonstrate a favorable safety profile, consistent with safety data with and longer-term exposure to intravitreal injections. The combined rate of new onset exudations over 24 months was 11.9%, 6.7% and 3.1% in the pegcetacoplan monthly, every-other-month and sham groups, respectively. No cases of endophthalmitis were reported between month 18 and 24. Over the 24 months, the rate of infection and endophthalmitis was 0.034% per injection, and the rate of intraoral inflammation was 0.24% per injection, both of which continue to be generally in line with reported rates in studies of other [indiscernible] studies. Additionally, no serious adverse events of ischemic optic neuropathy were reported between month 18 and 24 and no events of occlusive vasculitis or retinitis were observed across the entirety of both studies. With that, I will now turn the call over to Dr. Lad.

Thanks, Fede, and good morning, everyone. I am pleased to join the Apellis team today to share the full 24-month results from the DERBY and OAKS studies. This is an important readout for the retina community, providing us with even more data around the effect of pegcetacoplan in geographic atrophy. GA is a devastating disease that causes patients to lose more and more of their vision every year, resulting in irreversible blindness. Patients with GA lose their independence and they can no longer take on tasks, which are crucial in their daily lives. The results today are remarkable and give these patients hope. It is rare for drugs to show increased effects over the study period. Yet this is what we're observing with pegcetacoplan and its ability to slow down the progression of GA. It is fascinating to think what may be possible with even longer treatment. When looking at the 24-month data in the DERBY and OAKS study side by side, a few things really stand out to me. First, despite the variability that characterizes geographic atrophy, the sham control arms are virtually identical and linear in both studies. And consistent with previous large studies in GA most notably the lampalizumab chroma factory. This means that these results were clearly driven by pegcetacoplan and not by an inconsistent sham. Second, the efficacy and safety profile for pegcetacoplan were robust across both the monthly and every-other-month dosing and in both extrafoveal and foveal lesions. Well, we would like to have seen an impact on visual function. This did not come as a surprise. And given the increasing rate of lesion size reduction, we can expect to start seeing functional impacts emerge over the next few years. Finally, based on these data? And should the increasing effects continue beyond year 2 it's now possible to believe that treatment with pegcetacoplan may ultimately halt the progression of the disease over several years of treatment. Now I'd like to turn the call back over to Cedric for his closing remarks.

Thank you, Dr. Lad. In summary, pegcetacoplan's transformative potential in GA to further reinforced by the 24-month of data, demonstrating the potentially increased benefit for the millions of people with GA who are at risk for vision loss with no treatment options. DERBY and OAKS are some of the most comprehensive and meaningful studies in this business. We have a potential treatment that we believe is increasingly effective with a favorable safety profile and multiple dosing options. All of us at Apellis remain committed to bringing pegcetacoplan to GA patients, eagerly awaiting the FDA decision in November of this year and advancing our ambition to be #1 in the retina. I would like to close by again expressing my sincere gratitude to patients, physicians and scientists participating in the DERBY and OAKS studies. It is thanks to all of you that this milestone was made possible. And to our amazing team at Apellis, we are so grateful for your unwavering commitment to advancing care for people living with GA and other complement-driven diseases. With that, operator, please open the call for questions.

[Operator Instructions] Our first question comes from the line of Jonathan Miller with Evercore ISI.

Congrats on the update. It looks very interesting. I have a couple of little questions. First, on the regulatory front, you mentioned 24-month data being the basis submission from the EU, but you didn't talk about whether you're going to submit that to the FDA and what your expectations are for their response to the availability of this data, if any, on the PDUFA date. And then secondly, maybe more housekeeping. Can you explain some of the numerical changes to the data since the 18-month update? It looks like a couple of buyers have shifted around by 1% or 2%. And likewise, the every-other-month arm when you say it has 28% efficacy in both foveal and extrafoveal patients, but then on Slide 8, it's still only 24%. So I just want to get a sense for where those minor discrepancies are coming from.

Thank you so much. So first of all, for the first question with FDA. So when we submitted the18-month data to the FDA, we also included, of course, whatever we had available beyond 18 months. So in many ways, this is basically a confirmation of what we suspected already at 18 months and which is now confirmed. So for the FDA, there will be no major amendments, and we will simply provide them with the updates as we have to you. For the numerical changes, so when you do the modeling with the MMRM and the way in which the data are when recalculated, there can be these small changes. I mean they're not meaningful, but it's depending on how the model goes through the different segments of the timeline. That's really what it is.

Okay. Makes sense. And then maybe on the functional side, are there any other functional signals that might be better for -- than BCVA for getting some sense of functional benefit at an earlier time point? Is there any visible effect on like PROs or anything like that, that we could expect to see at AAO.

Yes. Thanks, Jon. So at this point in time, so after 2 years, we don't really have anything yet on the functional endpoints that stands out. We will talk more about functional endpoints at AAO and later as well as we do more analysis. But the key thing is that with what we see with this increased effect over time, what is really centralizing is the possibility that this disease ultimately may slow down completely, right, in many of these patients. That will inevitably result in functional changes, we believe, and that's something that we will also be able to evaluate, of course, in GALE. But at this point, this is the end line, if you want, for the DERBY and OAKS studies and that we couldn't be more pleased with the results.

Our next question comes from the line of Madhu Kumar with Goldman Sachs.

This is Omar on for Madhu. So we have a few questions. First, given these data, how should we think about the particularly patient retention? And then what gives you confidence based on prior interaction -- the functional data to support your care goal I ask the last question in my notes.

Thank you so much. Well, I will hand the first question over to Nora, who's actually -- to Dr. Lad who was an investigator in DERBY and OAKS and has also enrolled patients in GALE. Nora, would you like to talk a brief?

Certainly, I'm delighted to be here. So the patient experience has been very positive with this. Unlike other prior studies, the retention in GALE, the extension study has been excellent. I understand 85% as a whole. At our particular site is 100%. And I'll tell you if you anecdotes from some of my patients. And keep in mind that elderly individuals with comorbidities during a pandemic. They had a hard time continuing the visits, and they said, I think Dr. Lad, I think I've done with the study. And then they went -- one of them, especially went to an optometrist and got the prescription checked and non-study eye deteriorated substantially, whereas the fellow eye stayed stable. He came back and said, "Dr. Lad, I think I need to stay in this extension." And then there are others with similar subjective and objective experiences. So everybody was retained, which is pretty remarkable, I think, I'm telling also the retention in OAKS and DERBY was great, and I think similar only and in line with anti-VEGF studies.

Thank you so much, Dr. Lad. Then for the EMEA submission. So as we've always mentioned, the 24-month data will be included in the submission which is scheduled to happen before the end of the year. That submission or the request from the EMEA is to understand the relationship between the functional endpoints and the lesion sizes, right? And to have an idea what it means over the longer term, meaning beyond 2 years for the function of these patients. Again, it is important to note that the variability on the functional endpoint and the fact that we really have no good way of measuring visual function is the reason why at 2 years, it is too early to really be able to see that. But with the lease size reduction, because remember, what we measure is dying for the receptor cells, right, that should naturally translate into functional benefits over a longer period action.

And for the last question for Dr. Lad. Given this 24-month profile for PEG, what kind of GA patients would you think are best suited for this drug? And what would you need to see from competing agents to select them owner?

That's an excellent question. So I have a hierarchy in my mind already, but these data actually make my conclusion even stronger. So who to treat? I think all patients will benefit to prevent loss of retinal tissue according to these data and to prevent new areas of blind spots, which would decrease peripheral vision as well as deepening their central vision loss. So my #1 choice remains the patients that lost vision in the other eye. So in any GA in the study, I would -- or the treated eye. Now if both eyes have useful vision, I would recommend still that the patients that are extrafoveal are treated next to prevent the foveal involvement because this is the most valuable real estate that we have in the retina responsible for central vision. And number 3 would be foveal patients because I think they still benefit long term and visual function is difficult in GA to assess in a 2-year interval. We need long-term data. But over time, we'll prevent enlargement of the blind spot, loss of the peripheral vision, per central and peripheral and decrease central vision in the long term. So I will treat everybody in that order. But the data is getting better and better and gives me more hope and optimism as a retina clinician and for my patients.

And then could you let us know what would you like to see from competing agents to let them over PEG?

I'm sorry, can you repeat that for me?

Sure. What would you need to see from competing agents to select them over PEG?

Yes. So I think I probably would like to be able classify patients in terms of lesion characteristics to see who are best candidate for a particular drug. That's what I do currently with anti-VEGF agents, and we have a few tools in our armamentarium. We used individualized imaging to guide treatment decisions. So I have to see more data from the competitors right now to draw conclusions, but I could tell you right now, all GA will be a good candidate for pegcetacoplan and then down the road refine my treatment according to the profile that I see from competitors as well.

Our next question comes from the line of Anupam Rama with JPMorgan.

Just a couple of quick ones for me. So at the AAO presentation, what new analyses will we be getting? Second question, Cedric, I think what we're trying to understand here, and I've gotten a couple of inbounds on this is around the visual acuity data and if there is any regulatory rich based on what we know today around the EU filing based on visual acuity data? And then I guess final question, Dr. Lad -- how do you think about every-other-month dosing? And does that matter to you as a potential optionality for pegcetacoplan?

Thank you, good hearing yourself start with the first 2 questions and then hand it over to Dr. Lad. So at AAO, the new analysis, we are running the analysis. So we're going to analysis. So we'll see what we will be able to present at that point in time, but it will be in line with our history over the past year of providing very good insights into the data that we have. The VA data and the regulatory risk in Europe to go back to the question that I got from Jon, we believe that this case further strengthens our submission with the European Union, I'd say when you look at -- when you look at DERBY and OAKS next to each other at 24 months. And when you consider these increased effect over time, and how similar after 24 months, DERBY and OAKS have become, it's very hard to imagine -- and remember, you're measuring bang for receptor cells, right, with the sham that is absolutely exclusive. It is very hard to imagine this drug not getting approved. And I have no doubt that we're able to build a story around the functional lesion-based correlations. Dr. Lad, would you like to continue on questions?

Sure, definitely. But first, I'd like to finish up on the UI. I think it's great that the 24-month data will be submitted there. I think our tools has also improved. On the key secondary endpoints, you have the data as they are and it came -- there's no surprise to us. However, our technology has improved on structure function correlation in terms of algorithms and ways to look in more detail of this disease. So I think this will all help longitudinal data will be critical, as was mentioned a few times now and very important in this disease. So I think we'll have that and that will decrease the risk of the AAO submission. In regards to the last question, also very important for our patients, how often do you treat. So first of all, today's data suggest the long-term treatment is beneficial to achieve a maximum benefit in presenting the generation. Maximum benefit is overall with monthly and consistently with the data, but every-other-month really seems to give the patients according to these results, 80% to 90% of the effect depending on what time point and group you analyze in both studies. So it's also a very good option, very valuable in patients that will be unable to return for monthly treatment. As a clinician, we know down the road, the sustained delivery will be beneficial as down the road, and there will be next step for any approved intravitreal drug for GA.

Our next question comes from the line of Tazeen Ahmad with Bank of America.

Congrats on the good updated data. A couple of questions from me. Cedric, did you specifically submit the visual acuity data when you provided the 18 months and beyond update to FDA? And then secondly, can you just remind us when the mid-cycle review meeting with FDA will be -- is that going to be a live meeting or is this just going to be a commentary -- written commentary that you exchange with the agency. And then I'll have one question for Dr. Lad, if I could.

And so the VA data was included with the 18-month submission and as well beyond all the data that we had available at the time was part of the package. So again, that will not be surprised at all. Then for the mid-cycle review, we're not going to comment on that, but we feel, of course, very good about that interaction. We haven't had it yet, but -- and we're not commenting on exactly when we will have it.

Okay. And then Dr. Lad, as it relates to the specific data of -- I think it was one letter loss in the every-month arm versus just under 2 letters in every-other-month arm. To follow up maybe on Anupam's question about your view of every month and every-other-month. Taking that data into account, would you feel as comfortable dosing every-other-month as you do every month?

I think monthly is better. Overall data from studies that show that the maximum benefit is with monthly treatment consistently. And we have a similar scenario with anti-VEGF. There's on the treatment in the community still the best, results are done with frequent treatment per everybody's individual disease. However, and keep in mind, I think visual acuity data right now is very difficult to gauge after only 2 years. It's a nonlinear function, unlike lesion growth in GA. So it's difficult because the vision stays stable for extrafoveal lesions until foveal involvement, and then it decreases slowly over time. So we will see more in GALE, I think, Longitude. So right now, the 1 to 2 letters, I would not make too much of these data right now because it is too early to gauge I would still make the same recommendations based on prevention of lesion growth and the retinal degeneration with pegcetacoplan because we know with increased GA size, the blind spot, which is gotamine largest and that could be central peripheral and it really affects their activities of daily living for our patients. So the prevention is GA growth, and that will still make the same recommendations based on the data.

And do you think that based on what you just said, there should be a meaningful difference depending on what part of the complement cascade, one is using, whether it's C1Q or C3 or C5 as it relates to the impact on visual acuity and times impact.

That's an excellent, very insightful question. I don't know yet, but I'll tell you after these data, I have done quite a bit of reading on immunosenescence, immunomodulation and different aspects of the complement cascade, the mite [indiscernible] function and how. I've been working on visual functions for 10, 11 years now or more in dry AMD. And I think there's a lot to learn. So time will tell, but that's a great question, and I'll take it back to the lab and start to think about how to best address.

So I think it's maybe one little clarification there as well for people that may not be that familiar with it. Visual acuity is really a measurement of central vision alone, right? So there's 1 spot in the retina centrally, whereas the periphery is not measured in visual acuity. So somebody can have significant geographic atrophy affecting the peripherally meaningfully in a highly functionally impeding way. And when that person reads the Snellen chart, that person will still have 20/20 or 20/40 vision that entirely happens very often. So that's why visual acuity is a poor measure in geographic atrophy because it typically starts outside of the fovea. And then at the end, only starts overtaking the fovea at which point it will affect the acuity.

That's correct.

Our next question comes from the line of Colleen Kusy with Baird.

Congrats on the updates today. One on safety, if I could, nonexecutive CNV events come up in some of my conversations and obviously, there's another competitor in this space. Can you just remind us are nonexecutive CNV events being measured in DERBY and OAKS? And how would those patients be managed? And what your thoughts are on what that nonexudative signal might mean?

Yes. Thank you so much, Colleen. I will briefly comment on how we measure these lesions, and then I will let Dr. Lad comment as well. So first of all, and I think this is really, really important. Nonexudative CNV does not get treated with anti-VEGF, right? So I think that is really important. And it's also really important to mention that you can measure CNV at the time of an exenatide event. So basically, patient comes in, there is liquid in the retina and you want to treat that patient with anti-VEGF, which is an exudative period. After the patient is then on anti-VEGF treatment, if you look at 1 year or at the end of the study, that, of course, those lesions will be dry. And you could say they're not exudates, but that is because they are on treatment with anti-VEGF. So typically, the way it happens in our protocol is that every patient who has exudation, gets treated with anti-VEGF and goes on standard therapy until the end of DERBY and OAKS. So that I think is really important. There is not a single patient in this study that was detected with extrafoveal lesions that was not treated with anti-VEGF. We only had 2 cases outside of that for other things that were treated with anti-VEGF, but it's a complete picture. So Dr. Lad, I don't know if you would like to comment on that as well.

I think you covered it really well. And I don't have a lot to add other than I treat exclusive CNV in my clinic, not the nonexertive disease with anti-VEGF.

That's helpful. And then as a follow-up, in the GALE study, can you just confirm, are you measuring visual acuity there? And based on the data you've seen so far for DERBY and OAKS, do you have a sense on how much fault you might need to see separation in terms of those functional endpoints?

Yes. So we haven't run these analyses yet, but we, of course, are measuring visual acuity. We will also continue to measure lesion sales growth not as frequently as we did in DERBY and OAKS. So this will be every 6 months update. And we also continue to do macro parameter in the patient that continues out of OAKS into GALE. So it'll be very comprehensive. It is approximately 800 patients that we'll be able to follow for the next 3 years.

Great. And last one, for the 24-month update for DERBY and OAKS, did you look at visual duty based on the foveal versus extrafoveal subset of patients? And did you see similar trends or anything different in those populations?

Yes. So that is still being subjected to further analysis. But so far, we don't see differences.

Our next question comes from the line of Lyla Youssef with Cowen.

Congrats on the progress. maybe really quickly, could you elaborate on the [indiscernible] acceleration that you saw on the reduction in lesion size for this month between 18 and 24. So typically in the DERBY trial, it seemed like there is a chunk there. And then I have a follow-up question for safety for Dr. Lad.

Thanks, Lyla. So this is something that is incredibly exciting to us, right? I mean -- and it is worth noting we don't see it in DERBY and not in OAKS. Remember in every-other-month in OAKS, we see it as well. It is only in the monthly arms in OAKS, where we also see a compounding effect, so an acceleration, but they're smaller than in the other arms. And that was the arm that performed the best at the start of the study as well, of course. It's important to note also that when you look at the area of retina saved, this continues to happen. I mean in DERBY, in the second year, there is twice as much retina saved as was the case in the first year. So it makes a really tremendous difference for these patients. As it relates to the mechanism behind it and what could be going on, look, at the end of the day, and this is something that we've postulated from the very beginning is that when you control complements over a long period of time, we believe that you make an adjustment in the immune microenvironment that drives this disease. Geographic accuracy, we believe is similar to climate change, right, homeostatic system that goes out of control. And for a homeostatic system for retina to be healthy, there are many buffering systems that keep that retina in a healthy state, and there needs to be a tipping point after which the disease occurs. And at that point in time, the same buffering mechanisms will keep that pathological state in a very tenacious spot. So in order to come back, you probably have to work up against that same sensing point and then get over to the other side, which is why again, we are really intrigued by the possibility that ultimately, over several years of treatment, this disease could be stopped growing altogether.

Got you. That's very helpful. And then maybe for Dr. Lad. Could you maybe go over maybe in a little more detail your impressions of the current safety profile may specifically the actual -- the rate of the nuance at exudation that we've seen a little bit of an increase in the monthly arm with the update. And then while there hasn't been any new cases of ION, there were some previously. How are you thinking about maybe those elements of the safety profile in your mind as you think about pegcetacoplan.

So I think the safety profile is quite favorable. And again, there have been no additional cases of endoptomitis between months 18 and 24. And the rate is over 24 months is low 0.034% per injection. The rate of intra formation 0.2 or so per injection in line with prior other studies of intravitreal therapies, -- and of course, we always pay attention to any occlusive vasculitis events or retinitis, which were none, and no optic neuropathy in between 18 and 24 months. So the combined rate of new onset excitation we have listed there 24 months, you have the percentages nearly 12%, 6.7% and 3.1% in the monthly, every-other-month in Sham group consistent with prior data. So it appears to be a dose response. The anti-VEGF injections have been tolerated in excluded CNV in our patients. We've had no problems of what I understand contention of the regulators as well. So I think the safety profile has been favorable, and I'm not honestly particular concern. It does deserve through conversation with the patients about the risk of the potential exit of CNV at this rate, depending on the frequency of injections. And I always explain what's going to happen with any treatment to my patient. So that is no surprise there. It has not deterred patients from -- at all from the receiving the injections and they've done really well with the protocols of the ones that have had the dual injections.

That's very helpful. .

But I will add to the excalation treatment, I was very surprised like you are as well in a positive way as a scientist. And we don't know, but I have a few hypothesis. Cedric explained the one about dysfunction and aging of the inflammatory system in takes years to decades to occur. So it's no surprise that reversing some of the dysfunction may take time. There appears to be the tipping point that's necessary. Apellis has done, I would add a beautiful job in looking at some chance in baseline balances between the 2 studies that explain some of the differences in data and they've been at congresses. And some of these may take effect now. So the interaction between the drug and lesions changes after 18 months. And because of some of these changing balances, it might be a little stronger from 18 months on into DERBY. And we don't know why, but we'll keep looking into this, and I'm confident that we'll have more and more data on this. It's fascinating in a positive way.