REGENXBIO Inc. (RGNX) Earnings Call Transcript & Summary

Good morning, and welcome to the REGENXBIO conference call to provide updates from the RGX-314 wet AMD program. [Operator Instructions] As a reminder, this conference call is being recorded. [Operator Instructions] I would now like to turn the conference over to Mr. Patrick Christmas, Chief Legal Officer for REGENXBIO. You may begin.

Good morning, and thank you for joining us today. Earlier this morning, REGENXBIO provided an update via press release, which is available on our website at www.regenxbio.com. We will also be reviewing slides during this call, which you can access via the Investors Section of our website. Today's conference call will include forward-looking statements regarding our clinical development of RGX-314, in addition to regulatory and product development plans. These forward-looking statements are subject to risks and uncertainties that may cause actual results to differ from those forecasts and can be identified by words such as expect plan, will, may, anticipate, believe, should, intend and other words of similar meaning. Any such forward-looking statements are not guarantees of future performance and involve certain risks and uncertainties. These risks are described in the risk factors and the management's discussion and analysis sections of REGENXBIO's annual report on Form 10-K for the full year ended December 31, 2019, which is on file with the Securities and Exchange Commission and available on the SEC's website. Any information we provide on this conference call is provided only as of the date of this call, August 4, 2020, and we undertake no obligation to update any forward-looking statements we may make on this call, on account of new information, future events or otherwise. Please be advised that today's call is being recorded and webcast. I would now like to turn the call over to Ken Mills, President and Chief Executive Officer of REGENXBIO. Ken?

Thank you, Patrick. Good morning, everyone, and thanks for joining us on this call with an update on the RGX-314 wet AMD program. Importantly, I hope that everyone is healthy and staying safe. I'm pleased to welcome with us today, several well-known and well-respected retinal physicians to join Steve and me on today's call to share their perspectives and experience in treating patients with severe wet AMD. Dr. Robert Avery from California Retina Consultants and Research Foundation, and one of the study investigators will walk us through the 1-year data update from Cohorts 4 and 5. His colleague and another investigator, Dr. Dante Pieramici, and Dr. Peter Kaiser from the Cleveland Clinic will also join us for discussion and Q&A session. Today, we're going to highlight a few corporate updates related to our RGX-314 wet AMD program. Importantly, we remain on track to initiate the first ever pivotal program of a gene therapy for the treatment of wet AMD, based on the insights gained from our ongoing Phase I/IIA study. In a moment, we'll walk through the 1-year data from the 2 highest dose levels in our dose escalation study. After a single administration, RGX-314 was generally well tolerated across all doses and demonstrated stable to improved visual acuity and retinal thickness as well as a meaningful reduction in anti-VEGF injection burden. From a biomarker perspective, we've also seen dose-dependent anti-VEGF protein levels across all cohorts, which supports the important functional and anatomical effects we've seen. We've also announced this morning the advancement of AAV8 study towards the clinic. This study is a randomized, controlled Phase II study of RGX-314 delivered to the suprachoroidal space, using the SCS microinjector for the treatment of wet AMD. This targeted in-office delivery approach may provide additional options for wet AMD patients in all clinical care settings, and we anticipate providing interim update for the first cohort in late 2020. But first, I want to turn it over to Dr. Steve Pakola, our Chief Medical Officer, to give a quick overview of RGX-314 in the ongoing study. Steve?

Thanks, Ken. So just to level-set everyone on how the RGX-314 gene therapy is designed to work. It uses our proprietary NAV AAV vector to deliver a gene encoding an anti-VEGF antibody fragment into retinal cells. The Phase I/IIA dose escalation dose design is shown here. We have dosed 42 severe wet AMD patients across 5 dose cohorts. Patients received an anti-VEGF injection on study entry to evaluate their response to anti-VEGF via OCT scan. After response was confirmed, RGX-314 was administered via a commonly performed procedure of vitrectomy, followed by a subretinal injection. Patients did not receive any prophylactic steroids beyond what is normally administered after routine vitrectomy. The patients in this study are being evaluated for safety and efficacy end points including visual acuity, retinal thickness and anti-VEGF rescue injections. As reported in April, patients in Cohorts 1, 2 and 3 have completed the 2-year study with Cohort 3 showing a meaningful treatment effect with improved vision, stable retinal thickness and reduced injection burden. Today, we will be talking through the new 1-year data from Cohorts 4 and 5. As a reminder, since this was a first-in-human study for RGX-314, we provided the most liberal protocol specified anatomic and functional anti-VEGF retreatment criteria to ensure investigators had broad discretion to re-treat patients for any sign of disease. Now for an overview of the data from Cohorts 4 and 5, I'd like to turn it over to Dr. Bob Avery. Bob?

Thank you for having me, Steve. I'm very encouraged with the data that we're seeing from the study of gene therapy in wet AMD patients. I'd like to start the data review with some background on the patients coming into the trial. Across all 5 cohorts of the study, these patients enrolled had long-standing severe wet AMD, and were receiving frequent anti-VEGF injections over many years before enrolling in the RGX-314 trial. Today, we'll be focused on Cohorts 4 and 5, which enrolled patients with a high need for frequent anti-VEGF administration in the year prior to enrollment, with both cohorts receiving 10 injections in that year before coming into the trial. Patients were enrolled regardless of whether they had neutralizing antibodies against AAV8 and we'll show some protein data on this later in the slides, but it's important to note that the subretinal space is considered to be immune privileged, which allows for us not to have to screen patients out based on prior immunity. This slide provides a broad overview of the safety profile of RGX-314, which is generally well tolerated. As of July 13, 2020, there have been 17 SAEs that have occurred in patients and were not related to RGX-314, which is typical in a population of this age. For example, one of the more recent SAEs is from one of my patients who is hospitalized for a medical procedure, unrelated to wet AMD. In addition, there was one SAE that was possibly drug-related. This is one of my patients as well. She was an 85-year-old with chronic, dry and wet AMD, who had already undergone 94 intravitreal injections over the previous 12 years. Consistent with this long-standing history of AMD, she already had pre-existing pigmentary changes in her peripheral retina and other findings of advanced disease in her retina. In this patient, RGX-314 was administered by placing the blood superiorly above the macula. And during the study, she developed additional pigmentary changes peripherally below the blood inferiorly in the long-term and also in macula. At the week 50 visit, she experienced significant decrease in visual acuity of 25 letters. The degree of pigmentary change was not more prominent than other cases, but since the patient had significant visual decrease she was designated as severe. It's important to note that this patient's wet AMD has been well controlled over the year, meaning she has not had any fluid buildup, she has not received any anti-VEGF injections, and we continue to detect sustained anti-VEGF protein levels at 1 year. Many of the mild-to-moderate AEs listed here were transient and expected following routine vitrectomy such as subconjunctival hemorrhage. Many of these AEs resolved on their own within days to weeks after the procedure. Across several cohorts, patients had changes in retinal pigmentation, most of which were peripheral and inferior away from the macula. The consideration going forward is that blebs could be placed inferiorly below and far from the macula. Retinal hemorrhage is typical with older patients with comorbidities such as hypertension were picked up more than usual due to using widefield imaging in the study. There have been no reports of clinically determined immune responses or drug-related inflammation beyond what's expected following routine vitrectomy. This is important to me from a patient management perspective. I'm sure my colleagues here on the phone agree, treatment of long-term chronic inflammation can bring its own set of challenges for patients. So it's important that a wet AMD treatment not introduce new challenges for patients and physicians to manage in this older population that's predisposed to many of the complications related to steroid use. The next few slides relay efficacy data from Cohorts 3, 4 and 5. Cohort 3 patients have now been followed over 2 years, and the data has been previously presented. We can see here that the visual acuity over time from the patient's baseline is stable to improved in all 3 cohorts. Cohort 3 and 4 gained 5 and 4 letters, respectively. Cohort 5 is at minus 2 letters, which is largely driven by the one patient with visual decrease that I discussed earlier. Overall, we see stable to improved visual acuity across Cohorts 3 and 5 -- 3 through 5, which is the outcome we would hope for in previously treated severe wet AMD patients with long-standing disease. If we can address the underlying disease and reduce or eliminate their need for frequent anti-VEGF injections, while simply stabilizing their vision gains, that's meaningful for patients. If having foundational steady state anti-VEGF levels can lead to individual improvement, this would be an added bonus for these patients. Here we're showing the visual acuity changes in patients who have not received any anti-VEGF injections over 1 year after administration of RGX-314. So there are 3 patients from Cohort 3; 3 from Cohort 4; and 8 patients from Cohort 5. Cohort 3 and 4 had a gain of vision of 10 and 6 letters, respectively, and patients in Cohort 5 remained stable over 1 year. When we look at retinal thickness, we see Cohort 4 and 5 came into the study with more fluid at baseline than Cohort 3, which is closer to the normal retinal thickness. All 3 cohorts have a nice anatomic response. Cohort 3 demonstrates stability of retinal thickness and Cohorts 4 and 5 demonstrated an improvement of retinal thickness over 1 year from baseline. It's important to remember that these visual acuity and retinal thickness changes are in the backdrop of a reduction in anti-VEGF injections for the vast majority of patients in Cohort 3, 4 and 5, the needed frequent anti-VEGF injections to control their disease in the year prior to treatment with RGX-314. The first bar is the annualized injection rate for the year prior to enrolling in the study. As you can see, they're all reasonably similar across the 5 cohorts, showing how anti-VEGF needy these patients enrolled in the study were. And now looking at 1 year after administration of RGX-314, we see minimal to no change for Cohort 1 and 2, with a consistent reduction starting in Cohort 3, which had similar rate of reduction to Cohort 4 over 60%. And patients in Cohort 5 had a reduction of 85% at 1 year. The third bar, of course, we have 2-year data for Cohorts 1 and 3 -- 1 through 3, as reported previously, which show that the level of reduction in anti-VEGF injection burden stayed similar to the prior year, highlighting the long-term durability of gene therapy demonstrated in Cohort 3. Here's another way of looking at the data. First, laying out the anti-VEGF injections for all 42 patients prior to enrollment in the study. Again, we can see how frequently they receive anti-VEGF injections in that prior year to enrollment. Here, we see Cohorts 1 and 2 out of -- out to 2 years following the RGX-314 administration, where we can see there's no real change from the past, except for a few patients in Cohort 2. And now as the dose escalation continued to Cohort 3, we can see a meaningful reduction in anti-VEGF injection burden over 2 years. And finally, in Cohorts 4 and 5, now out to 1 year, we can also see a meaningful reduction in the anti-VEGF injection. Drilling down on the higher dose cohorts, we see clear reductions in injection treatment burden in Cohorts 3, 4 and 5, ranging from greater than 60% to up to 85%. We can also see a few missing visits towards the end in Cohorts 4 and 5 due to the COVID-19 pandemic precautions and for Cohort 3, these reductions have been sustained out to 2 years, which is particularly remarkable for wet AMD patients. And now turning to RGX-314 protein expression measurements, which is good evidence for production of anti-VEGF in the eye after the onetime administration, there's a nice dose-dependent increase across all 5 cohorts at 1 year. And here, we look at Cohorts 3, 4 and 5 over time. We observed that there was a dose-dependent increase in protein production, which levels off to protein levels higher than month 1 across all 3 cohorts at 1 year. Also, in Cohort 3, we've seen the benefit of 2-year data where we see that the protein level is stable between 1 and 2 years. And finally, we're sharing protein expression data in patients in the 3 cohorts by their baseline neutralizing antibody status to AAV8. As noted previously, there was no screening out of patients based on neutralizing antibody status. We see for each cohort that the patients with high and low neutralizing antibody titers produce RGX-314 protein consistently over a year after RGX-314 administration. So there's no indication that neutralizing antibodies affect the cell's ability to make the protein. Just to recap, we provided relevant data across Cohorts 3, 4 and 5, and I just want to say I'm really encouraged by the data shown here today, which has shown clinically meaningful and durable improvements across key parameters for wet AMD patients in the trial. I believe that RGX-314 has the potential to profoundly impact all aspects of clinical management for patients with wet AMD and could be a onetime gene therapy treatment option for a broad range of patients. Now I'll turn the call back over to Steve.

Thanks, Bob. We really appreciate your presentation and comments. I'd now like to take everyone back to the route of administration. As we've discussed, this first-in-human Phase I/IIA study was conducted using subretinal delivery, which is the gold standard for retina targeted gene therapy, we're very excited to move into the pivotal trial using this approach. We are also planning to evaluate RGX-314 delivered to the suprachoroidal space, using the SCS microinjector, a targeted in-office approach. We announced this morning that our plans for AVA, a Phase II study of RGX-314 delivered into the suprachoroidal space are moving forward, and we expect to dose patients in the third quarter here in the coming months -- weeks. We will evaluate the mean change in vision, safety and tolerability as well as effects on the retinal thickness and need for anti-VEGF injections compared to a control group of monthly ranibizumab treatments. Here, you see the design of the study, which will be a randomized controlled study with treatment arms of RGX-314 compared to monthly injections of ranibizumab. We're evaluating 2 doses starting with the 2.5E11 GC per eye, the same dose is used in Cohort 5 of the subretinal trial. I'll finish up here with our upcoming milestones for 2020. As mentioned, we have been really pleased with the data from the RGX-314 Phase I/IIA study and look forward to initiating a pivotal trial later this year. In addition, we expect to dose patients in the AAV8 trial in the coming weeks and plan to provide an interim update for the first cohort by the end of 2020. We also plan to initiate an additional Phase II trial of RGX-314 suprachoroidal delivery for treatment of diabetic retinopathy in the second half of 2020. So this is the end of the presentation component of today's call. So thank you, again, Bob, for providing your perspective on these results. And before we hand it over to the operator for any Q&A from anyone on the phone, I admit, I can't help but take advantage of this unique opportunity of having you as well as 2 of your colleagues who are also experts in the retina field, including treatment of wet AMD, Dante and Peter, here virtually to kick things off with a general question that I think we can utilize the benefits of all of your collective experience. And I think we'll start with you, Bob, since you went through the presentation, where we have a lot of results in here and I think it's good to step back and hear from your perspective, what are the important factors for you when you're thinking about potential new treatments, and in particular, gene therapy, as to how that would apply for the patients that you treat with wet AMD? And after you go, I'd love to hear what Dante and Peter say as well?

Great. Well as with any new therapy, safety and efficacy are paramount, and the RGX-314 was generally well tolerated in this trial. And with placement of the bleb of medicine inferiorly, I believe we can avoid the single complication that may have been drug-related in this study. The efficacy is outstanding, I think, and especially taking into account this very needy patient population, it is very difficult-to-treat folks who had so many injections, especially over the past year, but even over their history and demonstrates that if we can make this market reduction in injection frequency, and thickness and improvement in vision in this population, I think this is a very powerful treatment going forward and would be very helpful for folks. There's also the convenience factor, which is very important, this onetime treatment for a majority of patients is very appealing. We know the treatment burden of our standard therapy is really overwhelming and COVID has sort of shown us what happens when this population of patients is afraid to come in. They miss appointments. And when they do show up, they've often seemed to have lost vision. So it'd be great to have a very durable treatment option and not have to worry about these patients for an extended period of time.

Great. Thanks, Bob. Dante, you're a co-investigator in this study as well. What are your thoughts?

Yes. I think Bob summarized it fairly well. I think one of the things, anti-VEGF therapy has been immensely important for our patients. It's made a huge impact. As my colleagues, I've been involved in this space since before we had anti-VEGF therapy, and it's really changed everything for our patients. We get great outcomes. The problem really is over time. And if we look at real-world data, if we look at clinical trials after they end, we see that patients -- because of -- in large part, because of the burden of therapy, injection frequency drops off over time as does visual acuity. So if we look at trials like the AURA trial, which follow patients in a variety of European countries over time, patients do well initially. And over time, they eventually lose visual acuity sometimes back to baseline. So what we really need today that we have effective anti-VEGF therapy is we need treatment that's more durable, that's easier to do with equal or better efficacy. So what we're really proposing is increased durability, less treatment burden that will result in equal or better efficacy and then much better efficacy in the long run in these patients. So we also want a treatment that has a reasonable safety signal as well. And I think that what we're seeing so far with this trial is we're going to get what we need: increased durability, less treatment burden, and a relatively reasonable and safe and effective treatment for these patients in the long run.

Great. Peter, you aren't an investigator in this trial. But of course, you're familiar with not only all of the standard of care treatments, but everything that's out there, experimentally trying to address this big unmet need in terms of durability and other aspects. So we'd love to hear your general take on how you look at this data and what you're looking for when you think of gene therapy as a treatment option?

Yes. Thanks, Steve. To me, I've been on BLA teams for Regeneron, Novartis and recently Allergan and really, what we're starting to see, especially with the recent inflammatory events with brolucizumab, Beovu is that the FDA is really looking very closely at safety. So any time I look at a potential new treatment for macular degeneration, safety has now become paramount. And looking at the results of the Phase I/II study, the safety appears to be very good. Even the patients who Dr. Avery showed who had the SAE, and this -- it's hard to say if that's related to the treatment or if it's just progression of underlying dry macular degeneration, which we unfortunately see, especially in patients that have been treated for extensive periods of time like that patient has been treated. I think the other important thing is what Dr. Pieramici brought up, which is are real-world studies with the intravitreal injections have shown very, very poor results, nothing anywhere close to the fixed dosing, but nobody anywhere in the world does fixed dosing. So there's really no way for us to achieve the results in clinical trials. In contrast, when you use gene therapy, if you look at the amount of production of VEGF in these studies, it's well above what we need in any study we've ever looked at for the management of macular degeneration. And looking at the levels over time also makes me feel very comfortable that this one treatment seems to be holding steady in our patients. So the hope then is if we start to actually enroll patients that will be reasonable for this therapy. And I say that because whenever we do a Phase I study, as in this case, in general, we're enrolling patients who are train wrecks, for lack of a better term. And you kind of can see that in some of the slides that were presented today where the patients were receiving almost monthly injections right up until they joined the study. These are train wrecks. These are patients who, many times, can't even improve and to actually see some improvement on the OCT and even some evidence of efficacy in terms of visual acuity in these train wrecks, really made be optimistic with this treatment. To summarize, I think the ability to have long-term durability in terms of treatment well above what we would need in the eye, combined with excellent safety. And the final thing I would say is a lot of people say, well, this is a very difficult, specialized surgery that would require retina specialists to go out and get specialized training. And I would disagree with that statement. Vitrectomy is something that any retina specialist does as a bread-and-butter surgery to produce a subretinal bleb. It is also something that is very simple. So to me, this is not a complicated surgical procedure. It is something that is within the wheelhouse of anybody who's done a retina fellowship.

Wonderful. So thanks, all 3 of you for that perspective based on the results that we're seeing with this onetime treatment. That discussion, a lot of it centered around what we're seeing with subretinal delivery, we gave an update as well today, an announcement on our advancing suprachoroidal delivery first in wet AMD. And subsequently in DR. So before we hand it over to the operator, I just wanted to ask one follow-up question on this. How do you think of the potential utility and these variables that you've just been discussing when you think of subretinal and now the potential of moving into in-office suprachoroidal delivery.

Well.

This is Peter, sorry. I was part of Clearsight when they obtained this needle that allows us to inject very easily into the suprachoroidal space. And over the time, with developing this needle to make it a very -- this is a relatively simple in-office procedure. I was cautiously optimistic about the preclinical work with the needle and gene therapy with RGX-314. To me, it's actually very interesting, very exciting. And I'm looking forward to the results of this study because, certainly, this would be a considerably easier modality in terms of applicability across the country and even across the world. I think I would love to see the difference in terms of how this produces VEGF -- anti-VEGF proteins. So I'm excited about it. I'm sorry, I cut you off, Bob.

No, you said exactly what I was thinking. Thank you. I agree completely. And I think that it certainly would be more convenient for the patient if they didn't have to go to the operating room, but I completely agree with what you said about the surgery not being anything special. It's very routine in our practice. In the subretinal study, most of my patients were very happy to undergo the procedure in hopes of having a lower treatment burden and avoid future injections. And yet, if we could do something in the office without a trip to the OR in this population, that would be preferred. It may be an option just for those patients who don't have the neutralizing antibodies due to the lack of that immuno-privileged subretinal space. And so we may be looking at this in the future just for patients without the neutralizing antibodies. But on the other hand, the subretinal procedure looks like it's good to go for all comers, and that's also very exciting.

Yes. And I would just end by saying that I agree. I mean, I think that an in-office procedure is going to win out over a surgical procedure. But again, as my colleagues have mentioned, this is a relatively straightforward surgical procedure, the subretinal procedure that is, and it doesn't take extensive training to be able to do this. There are some potential advantages as far as immune privileges, potentially less inflammation with the subretinal considering compared to the suprachoroidal, but we don't know that exactly. I mean we always assume things based on basic science, but the suprachoroidal space may be quite fine. Good transduction with relatively low inflammation and the neutralizing antibody may not be a big issue in that space either. But I'm excited as well. I think that it will offer our patients some variety and potentially an in-office treatment.

I just want to add one additional point, Steve, before we open it up. People think of a suprachoroidal injection, again, is something very complicated. When we were starting to develop this needle, that was our biggest concern was that this is going to be a difficult procedure to perform as opposed to, say, an intravitreal injection where you just basically are sticking a needle in the eye as long as you're in the right place, it works. In this case, you have to be in a potential space. But as the clinical studies have borne out, basically, it's very easy. As you're going in, you're starting to feel the resistance and all of a sudden the resistance releases, and you know you're in a correct potential space. So all our sort of concerns that we had prior to human studies were basically alleviated by using this needle, and this is the same needle that's being used in the study. So I feel very comfortable saying that investigators in the study then hopefully, eventually, retina specialists will find this procedure very, very easy.

Yes. This is really a great feedback, Peter, and it's great how we get the benefit of just not all 3 of your global experience in terms of treating wet AMD in trials and your everyday experience. But even your hands-on experience with the different devices that have been tried in the past. So thanks again to Bob, Dante and Peter. Now operator, we're ready to hand it over to you. So our esteemed panel can also answer any questions that our audience has.

[Operator Instructions] The first question comes from Geoff Meacham of Bank of America.

This is Alec on for Jeff. My first is on the dosing. So given the encouraging 1 year follow-up from Cohorts 4 and 5, how are you thinking about the optimal dose in terms of safety and efficacy? And do you think it would be worth dosing even higher? Do you think you've pretty much reached saturation at this point? And then how are you thinking about powering of the pivotal study? And do you think noninferiority, say, to monthly Lucentis would be sufficient as a pivotal endpoint? And then I have a follow-up.

Great. Thanks, Alex. Steve here. I'll take the sponsor end of that first before handing over to a couple of our investigators on the call. So on the back end of your question on moving forward, noninferiority, superiority. We feel very comfortable with non-inferiority based on the precedent of other programs. And also thinking of the target profile that our guest experts eloquently walked through, that the ability to greatly decrease the treatment burden that currently exists while maintaining visual acuity, would be a tremendous advance with or without any further improvement in visual acuity. We're encouraged that we've actually seen some visual acuity going in the right direction. But even stability would be favorable. But that all supports the ability to go with a noninferiority design. Next on the dose selection. Certainly, what we've seen from Cohorts 3 through 5 makes us feel very confident with the results we're seeing across that range that we can pick and potentially go up to 2 doses anywhere within that range is what we'd be evaluating as of now. And we'll be continuing to evaluate that with our investigators, including Bob and Dante, and other advisers. So I'll hand it over to Bob and then Dante, to see how you look at what you're seeing in terms of benefit/risk profile coming out of the data in the doses that -- or a dose range that you think would be viable to move forward with.

Well this is Bob. I'll just comment that these 5 cohorts we have give us a good sort of range of doses. We know Cohort 1 and 2 did not seem to do hardly anything looking at the injection frequency and the amount of protein measured. And yet we have these high levels of protein measured in 3, 4 and 5 in the dose response curve, but all of them seem to give us a good clinical benefit, 3, 4 and 5. So I think selecting 2 doses for the trial within that range would be the way to go. Because we know that we've bracketed the best therapeutic range most likely given the potential side effects of going higher, without even knowing if it's tolerated. So I would think we would be going in that range in the 3, 4, 5 range for our 2 doses.

Yes, this is Dante. I agree. I mean, luckily, we do have some -- we have some flexibility here since 3, 4 and 5 have all shown very nice durability and effectiveness and good visual and anatomical outcomes. And again, I agree, I don't think I'd go any higher. We don't have any data on this. We can probably select a couple of dosages to go with and end up with a good outcome. I just emphasize again for folks on the line that, in our field right now, the push isn't that we're looking for more effective or more efficacious anti-VEGF agents. We're really at a sea change now to try to deal with this durability question. So the noninferiority study is what we really would be looking at, equivalent noninferiority study. That would be a home run to equal -- to end up with equal efficacy with visual outcomes and anatomical outcomes with a much reduced treatment burden because that would equate to, in the long run, much better outcomes for patients 3, 4, 5 years down the line. So as Steve mentioned, I do think the noninferiority study is the proper one for this. Based on the data so far, it's even surprising we're getting visual improvement. So we may end up not only with a home run, but with a grand slam. Perhaps we will have slightly better outcomes than we expected. But again, I think that the winner here is equivalent, visual outcomes, the equivalent anatomical outcomes with a much reduced durability, and we've seen that with Cohorts 3, 4 and 5.

Okay. Great. And one more, if I may, on patient selection, which hopefully can be answered by both you, Steve, as well as the panelists. Have you seen any defining patient characteristics in the individuals that didn't require rescue injections out to 1 year? And based on observations to date, how do you plan to structure the patient selection in the pivotal versus the Phase I/IIa?

So from a sponsor perspective, looking across the data, echoing what Bob and the others have said, we've seen very good response in Cohorts 3, 4 and 5. Overall, in the Cohort, if you dig a little deeper, there are some of these patients who have needed injections where it's not surprising they wouldn't need injection because as a first-in-human study, to quote Peter, you wind up with more of these very chronic train wreck kind of patients who, even if you dry out their retina, they may not get better visual acuity and may even deteriorate over time and then also even in terms of what we're looking for, having the wet AMD addressed without need for injections, this study had a high number of patients who were very difficult to treat, were not well-controlled, even after getting the many injections that Bob showed on many of his slides. So in a nutshell, a very easy aspect that we're looking at of adjusting inclusion, exclusion criteria is basically tweaking some of the criteria in terms of how well controlled patients are when they come in. In this first-in-human study, we had a low bar for what we called anti-VEGF response. So even a marginal response to the screening anti-VEGF Lucentis injection allowed the patient to come in. But we saw some of those patients who, yes, they had a marginal response but they weren't very well controlled. And therefore, it's not surprising they still needed rescue injections, and they still didn't do well in some of those cases afterwards, supporting that those aren't the types of patients that you'd want in the formal pivotal development. So maybe, Peter, since you didn't have a chance to comment on the last one, when you look at these results as they were presented and your thoughts on, as you've already elucidated, what you can see in our first-in-human study from your clinical expertise, how do you think of the potential to enrich for a pivotal trial?

Yes. Thanks, Steve. To me, looking at the results of these studies and knowing sort of the regulatory environment that we are working on it's important to understand that you use these studies to design your Phase III. So there's certainly going to be aspects of the patient selection. So for instance, for designing a Phase III, you're not going to want to enroll patients who've been on multiple injections for 3, 4, 5 years. The chance of that patient having a response is much lower. We know now that this is a safe procedure. So at the beginning, it's sometimes hard to get patients to the studies. In contrast, getting patients who were previously treated who would do anything, especially in the era of COVID, to reduce their injection burden, yet have a possibility of similar either a vision gain or I would just say, just keeping the same vision would allow us to get a lot of patients into this type of study. So to me, obviously, a lot of work needs to be done to really look at the data. But I would limit the amount of time the patients had received treatment. I would make sure that the patients had a very nice treatment response to that first injection, not like just minor change, really a good treatment response to show this patient will really benefit from this treatment. I just wanted to go back to the regulatory question about noninferiority. That's absolutely the best course of action here. The FDA doesn't care about reducing treatment burden. That's not an approvable endpoint. Visual acuity is, the vision acuity endpoint at 9 months. The noninferiority versus monthly Lucentis has been shown to be approvable based -- I mean not approvable, at least acceptable to the FDA in the recent Roche Archway results, which was, again, almost identical in terms of what was -- should be done here with the Phase III. So the pathway forward has already been forged for sort of a sustained delivery of anti-VEGF with Archway, and ranibizumab would definitely be the comparator of choice in this study.

The next questions comes from Gena Wang of Barclays.

I have a few. So the first question is regarding the one SAE and the Cohort 5 after 10 months or month 11. Could you give a little bit more color, again, how would you decide this is a drug-related or not since it's after 11 months of a surgical procedure? And also how the patient was resolved? If we look at the Slide 15, if you can tell us which patient actually is on that chart that had the visual loss. My second question is related to Cohort 4. So if we look at the Slide 15, I guess the tumor patients that need a rescue injection compared to 6 months update, I assume it is a patient 5 and 6. So could you give a little bit more color on these 2 patients regarding their protein level during that time, are there any decline regarding the protein level? And then last question is regarding the suprachoroidal delivery. So now you pick 2.5 E11 and 5 E11. 5 E11 is basically higher than Cohort 5. So first is the rationale to picking higher dose. And then more important is also how do you do the injection? Would that be 2 injections, one on each side of the eye? So that's my questions.

Okay. So if we can remember all 3, let's start chronologically. So on the single possibly related SAE, since that was your case why don't we let you start off? [

Sure. First of all, we're not sure this is drug-related, but this patient did have inferior pigmentary changes as was common in Cohort 5. What I mean by that is there's modeling of the pigment in the far periphery below the [ arcades ]. And this may be related to the drug itself, when we make the bleb and do a gas exchange in the eye. The fluid may migrate inferiorly and pool down there. And that may be what is the source of this pigment change, that's at least one theory. And as it tracks down from the superior bleb, the fluid tracks through the macula. And I think that by making the blebs inferiorly to start with, you won't have this -- the drug itself, the RGX-314 going through -- tracking through the macula, and then you wouldn't have the potential for pigment changes occurring there. So we're not sure that these pigment changes are related to the drug itself. This patient, as was mentioned, had 12 years of almost 100 injections and patients like this, as Peter mentioned, should be excluded from future trials, most likely because we all know from long-term follow-up studies, these patients can have loss of vision just from atrophy regardless of our treatment. But because this patient did have some pigment modeling in the macula, we thought that it would be safest to say that this was possibly related to the drug itself. Now other patients have had these inferior changes and have had no visual consequence. And this just happened to be one that had some mild pigment modeling centrally and that's the reason we said it was potentially drug-related. With respect to this one lady's injection status, as I mentioned, she never required an injection. She never had fluid after the treatment and her protein level was very good at 1 year. It was adequate. And so we stopped her wet disease. But at 11 months, she had this sudden visual loss. It did not seem to recover at the 1-year follow-up time point, it seems to be about the same. So I hope that answers the question. I don't know that it's drug-related for sure, but it's possible.

Thanks, Bob. And I think that you hit on a lot of how we look at it. There's always a conservative aspect of saying possibly related if you can't rule it out. But this really was a classic train wreck type patient in a Phase I/IIa type study with almost 100 injections, and evidence of advanced disease, where, yes, it seems a lot of confounding factors that, fortunately, we would not have in our larger pivotal studies for the reasons that we've all discussed. Your second question, Gena, I think, revolved around the protein levels and what variation or what we're seeing over time, I think it's important to reiterate what we've said in the past when we've given updates on protein before at earlier time points where we've always viewed protein as important to clearly demonstrate that we're getting transduction and an expression and that it's actually durable. And that's what we've seen, and we've seen a clear dose response. It's also important, though, to recognize that we're taking aqueous humor samples because that's the ethical least invasive way to be looking at protein and therefore, you're removed from ultimately what matters, which is anti-VEGF protein in the retina itself. And what's important is that we actually see protein and that we see that it's durable. But there's certainly a lot of variables at play. If we're looking at aqueous humor of it having to diffuse from the retina to the vitreous and then to the aqueous humor, and the reality is that before you get to some steady state, you might see some variability. So we think that is a very likely explanation for frankly, not too much variability that we're seeing. So we're quite pleased with what we've been able to demonstrate in terms of the protein measures. And especially the fact that for Cohort 3, where we have the benefit of 2-year data, we see that, that's extremely stable from year 1 all the way out to year 2, the longest follow-up that we have. And lastly, that we're seeing these protein levels are translating into the intended clinical benefit that has us encouraged and our advisers encouraged as well.

Yes. So my question is more like for -- I totally understand and then totally agree with your comments. They are very clear dose-dependent protein level. Just wondering for these particular 2 patients, if we look at it, they do have a few recurring repeated rescue injection. Just wondering, were these patients were already, say, borderline protein level? And over time, after 6 months, would their protein level have a slight decline, and therefore, they fall into the other category that they will require rescue injection.

Yes. So good question. There were 2 subjects, those later ones that you can see that then got an injection. Their protein levels were stable. They were not losing protein. Keep in mind, we had the very low bar for retreatment. So in these 2 cases, even a small amount of fluid, they could get retreated, and we see that those patients have done very well. Your last question in the interest of time, I'll take as well so that we can make sure to have other questions as well on the suprachoroidal also since that is specific to the study design. So we know from all the work that's been done before and experience that Peter has summarized that there's excellent clinical experience with 100-microliter single injection. So that's what we're starting out with in initial testing with the lower dose cohort of 100 microliters to achieve the dose level C5, 2.5 E11. You raised the higher dose. So there, we would actually give a second injection based on preclinical data, where we know that, that can actually be achieved safely and with good transduction around the entire globe in terms of retinal transduction based on work from Peter Campochiaro and others. So that's our current approach for the wet AMD AAV8 study.

The next question, sir, that will be from Matthew Harrison of Morgan Stanley.

This is Connor on for Matthew. So just 2 quick ones from us. Can you comment on the variability and need for rescue injections between, or I guess, among Cohorts 3, 4 and 5? And then so you touched on this briefly just a minute ago, but for the higher dose in the suprachoroidal plans, how do you expect the patients treated at the higher dose level to compare to patients in the subretinal from a safety and efficacy perspective?

Thanks, Connor. So Dante, why don't you take the first question on how you look at the variability in terms of treatment response within Cohorts 3, 4 and 5?

Yes. A couple of things. First, I would just emphasize the fact that you already mentioned, Steve, is that the bar for retreatment was very low here and much lower than we've seen in some other trials for extended delivery of anti-VEGF agents. So I think that overall, as we tightened up those a little bit more, we're probably treating some patients that don't really need to be treated at least in this trial. Also, I think you just have to remember there's relatively small numbers of patients in each of these cohorts. I mean there's more in Cohort 4 and 5, but Cohort 3. So some of it just may be variability of small numbers. And I think that this may be at play here as well. So again, I think that we're probably over-treating some of these patients based on the criteria. We wanted to be very flexible as well. And remember, when we have small numbers of patients, there's a lot of variability and we have patients that are included and the patients may have been a little bit more severe in Cohort 4 than Cohort 3, particularly if you looked at the fact that these are patients that needed lots of treatments and had thicker retinas in Cohort 4 compared to Cohort 3. So these may be ones that are somewhat not -- unresponsive to anti-VEGF therapy. And based on the low retreatment criteria, I think that they probably got treatments that may not have been necessary. So small numbers, more variability. The 2 groups may not have been exactly comparable and Cohort 4 being worse than Cohort 3, for instance.

Was there a follow-up, Mr. Harrison?

Yes. Sorry. So I just wanted to hear if you guys had any comments on what your expectation is for the higher dose level in the planned suprachoroidal study? And I guess, how you expect that to compare to patients in the subretinal study?

So Connor, one important point is it's certainly earlier days for suprachoroidal delivery. This is, in fact, our first-in-human study technically in terms of this route of administration. So we'll have to see. I think, though, we do have greater confidence based on the fact that we have the experience we have with 314 given subretinally with a clear dose response shown there. And given the data that we've accumulated after 2 years for Cohorts 1 through 3 out to greater than a year for the higher dose cohorts, all of this put us in a position to not have to reinvent the wheel or start from scratch in terms of dose ranging. So that's why we were comfortable and the protocol could move forward with already starting with the highest dose GC per eye that we were able to evaluate in the subretinal study to already start with that. And then we'll be doubling that in the second cohort. So too early to say what we'll see, but we'll be looking at the same types of endpoints and the same signs of efficacy that we've been discussing for the existing Phase I/II study.

Next question comes from Gbola Amusa of Chardan.

I got a couple of questions, regarding the SAE and I think that it may or may not be drug-related, and she'd likely be an exclusion for the pivotal. But to the extent it were drug-related, are you believing it's more on the anti-VEGF activity spectrum? Or I guess, that would be more likely than late onset inflammation or something of that nature. And then second related question is, I think that you said that to avoid any retinal pigmentary changes you could move the bleb. Are you saying that you would do this in the pivotal trial? Or since this could be an exclusion that you wouldn't have to?

On the first part of your question, Gbola, as far as etiology, I think one important point is it's not unsurprising to have pigmentary changes show up, particularly in a wet AMD populations, but also just in general, with gene therapy, where we know, for example, with LUXTURNA, there are numerous cases of large areas of pigmentary changes after gene therapy. So that's to that aspect. As far as Bob and the others' comments, yes, this is something we've discussed where in that particular case, whatever the underlying cause is certainly the ability of the fluid to potentially track through the macula, it just adds greater comfort and is a clear, easy -- relatively easy risk mitigation step to simply not inject over the macula. So I think Bob and the others are spot on that it's just as easy for us to design a study where we inject inferiorly. And that just adds greater comfort, no matter what the underlying etiology is. Since after all, we're delivering a therapeutic protein. So the key is to get it to some retinal tissue where that can be produced to supply anti-VEGF for the retina, it doesn't need to be in the macula. So we have the flexibility to inject farther from the macula.

So I'm just confirming your belief is that, that would not affect efficacy.

Certainly not. What we've seen -- we've seen no relationship between where the blebs were placed and the good results that we've seen in the Phase I/II study. Maybe I think since we have the benefit of Bob and Dante and Peter, do you see any issues with just injecting in a different spot?

No.

Go ahead, Bob.

No, I was just saying, I don't -- I mean I don't think we've seen any difference really with the production of the protein. So why not just eliminate a potential problem by injecting inferiorly we sit the patients up afterwards and the fluid goes into the periphery where it's less vital, but still can transfuse the virus well into the cells, and thus produce a protein. So I think it's just a -- we've seen that there's no difference really in the production of the protein, so why not just eliminate a potentially risky step.

I'll also say that there was certainly no inflammation. I mean one of the questions you had is could there have been inflammation in this patient at month 11, she never had any inflammation as none of the patients really showed any inflammation during the course of the trial other than just the typical vitrectomy inflammation in the first week or so.

Gbola, I think you have a follow-up.

No. I'll just say congrats on the progress. I'll leave it at that.

Next question comes from Luca Issi of RBC.

Luca Issi from RBC Capital. Congrats on all the progress here. Maybe one strategy question and one smaller question on the science. So the strategic question is, I think Roche/Genentech recently defined their subretinal delivery as a challenging first generation way of delivering gene therapy. Can you just talk about why you think starting a Phase III trial with a subretinal delivery is the right strategic decision instead of maybe waiting for the suprachoroidal approach given that you guys seem pretty excited about that? So that's the first question. And then the second is, are the patients with neutralizing antibody going to be included or excluded in the suprachoroidal study?

Yes. Thanks, Luca. Well your second question is part of the answer to the first, but not the only component. Subretinal, we have a burden in hand here with clear positive results across a range of doses that's really hitting our target profile. So we're at that stage with really what is the gold standard for how to deliver. We've heard Peter discuss how in practice, this isn't going to be an issue to scale up in terms of feasibility. So to have a onetime treatment that can achieve what the experts are saying is what the real unmet need is, we see this as a clear go. The other related aspect is we know with subretinal with AAV8 and with our RGX-314, in particular, that it doesn't matter if patients have neutralizing antibodies to AAV8. And we've shown that, and Bob has walked through that. That's what we expected to see. But now with this data, we know we have the opportunity to address unmet need in a broad population. We don't have the answer to that in terms of immune privilege for suprachoroidal. So to answer your question, we will be excluding patients in initial cohorts who have high titers to AAV8. But fortunately, that's minority of patients to AAV8. So we'd anticipate about 30% of patients being excluded based on that. But again, it's earlier days for suprachoroidal. So for us, it's very clear to move ahead with subretinal, while we learned more about suprachoroidal and how that treatment option may or may not fit in based on the results that we get. I'd love to hear -- yes, Peter, you mentioned your view on the subretinal. I just think strategically, probably the most important thing is your view from what you -- the patients you see, do you see the subretinal onetime treatment option with the kind of results we're seeing being a go for moving forward?

Well,I think the development plan that REGENXBIO has is a good one because we don't have the answers. If we need the results of the AAV8 study, then it'll be a much easier regulatory decision, moving forward with a pivotal with subretinal delivery is the smart thing to do right now, awaiting the results of suprachoroidal, we'll see. If it turns out that neutralizing antibodies have no relevance with suprachoroidal, and remember, the number of new antibody patients with AAV8 out there and the general public is an order of magnitude smaller than AAV2, which is what others have used. So to me, I think the smart thing to do is what you're doing, which is let's see what happens. Predict in the future at this point is a way to fraught with error. Hopefully, both work, and then it will be a nice situation to make a decision about which one to take forward or to continue, I should say. But I think right now, doing both developments, it makes the most regulatory sense.

The next question is from Mani Foroohar of SVB Leerink.

A couple of sort of trial operational questions. It looks at first blush to me, the administration of the suprachoroidal delivery versus controlled, well, say, ranibizumab active control would be challenging to blind, how operationally do you plan to blind assessment of CRT, visual acuity, et cetera, is there a plan to mask those assessments as you move forward in the Phase I/II, Phase III and beyond. So the second question, it sounds like the rescue criteria used here were more stringent than what you would use with the subretinal approach in the past studies, is that true? And if so or I guess not, how should we think about the resting criteria for AAV8 versus what we saw in the equivalent Phase I/II stage development for the subretinal approach?

Thanks, Mani, both great questions. The first question on masking, that's very critical for key endpoints. And yes, we standardly for these, even if there's procedural interventions that make it hard to mask. Generally, we still can always mask the best corrected visual acuity. Examiner, and we always do a reading center assessment where the central readers are masked to treatment allocation. On your -- can you remind me your second question?

Yes. Sure. My second question was around rescue criteria, and how should we think about the [Audio Gap] versus what we saw in subretinal and then similarly going into AAV8?

Got it. Yes. So that was our first-in-human study. We had no clinical evidence yet of durable effect. Now that's different, albeit with a different route of administration than what we're going forward with SCS. But certainly, the learnings from that are that we can go with tougher retreatment criteria even in our suprachoroidal delivery approach to assess and we still have the opportunity to look at all these different layers of data in terms of with tougher retreatment criteria, do the patients still do well? Or do they stabilize? Do they get better? So I think that will actually make it easier to get a determination in terms of a pharmacodynamic signal in terms of not only looking at retreatment, but how the underlying disease is doing in terms of the retinal thickness and also the visual acuity results.

This concludes our question-and-answer session. I would like to turn the conference back over to Ken Mills for any closing remarks.

Thanks, operator. Well thank you, Bob, Dante, Peter, for sharing your views and insights with us and with everyone today. Thanks, everyone, for joining. I think we can see that RGX-314 continues to make progress, and I'm really incredibly proud we're able to present this update that we're preparing to launch first pivotal program for gene therapy in wet AMD. And that will be leading the field in the use of a new device for AAV8 delivery in ophthalmology that is certainly, this progress is really a result of the tremendous work by all of our teams, partners, advisers and particularly the patients and investigators. So we look forward to providing everyone with updates on other parts of our pipeline on Thursday during our regular earnings call. That's it for today, and I hope everyone stays safe and well. Thanks.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.