Hi. Good morning. Thank you all for joining our call to discuss ProQR's STELLAR data, which was, of course, announced yesterday and broadly the treatment landscape and unmet need in Usher syndrome and non-retinitis pigmentosa. I'm very pleased to have my colleague, Alethia Young, joining us this morning, as well as Dr. Byron Lam, who's a Professor at the Bascom Palmer Eye Institute at the University of Miami and, of course, an expert in inherited retinal diseases, including Usher syndrome. Before we get started, just a quick disclosure, the expert has confirmed that he will not discuss any material non-public information on this call, and we and the audience will frame our questions accordingly. To that point, if anyone does have questions that they would like to ask during the call, please feel free to shoot us an email at either emma.nealon@cantor.com or alethia.young@cantor.com, and we will try to work those in as well. With that, we'll jump into it. Dr Lam, thank you again for joining us this morning. Could you just introduce yourself briefly and then provide a quick overview of Usher syndrome and nsRP? Sure. I've been a clinical scientist for over 20 years. I'm a principal investigator in many of the gene therapy clinical trials. I think everybody who has been in this field with a good database will participate in studies have a good database of Usher's patients. In particular, Usher 2A is the most prevalent Usher syndrome. That provides a landscape of Usher syndrome. We can also say that Usher syndrome, this particular treatment is not for all Usher 2A, specifically those with a mutation on the exon 13. That is a very common region for Usher syndrome type 2A genotypes. Okay, great. What is the unmet need in this disease, and how quickly does vision loss progress, and what current treatment options are there? There are no treatment options currently. I think, in the past, people have done the study in DHA clinical trial a while back, and the results were certainly not impressive. There is no treatment progression. Certainly by the teenage years, most patients have decreased night blindness, some decrease in peripheral vision. Their visual acuity remains quite well, sometimes in the late teens, sometimes into their 20s. Eventually, the disease catches up with them, so it start to affect the central vision. By the time they're into their 20s and 30s, that's when the visual acuity decreases. In some cases, it can be rather severe. By the time they're middle-aged, some patient could definitely be very severely affected, and some patients may lose their vision altogether in later life. Got it. In clinical practice, do you think about advanced versus early to moderate patients as distinct subgroups? I know in this study, ProQR used 70 letters as sort of a cutoff. Does that threshold make sense to you? Well, I think the reason people want to put in different categories is because, what is really experienced by the patient in a disease, for example, in early and moderate group, is that they lose peripheral vision. Their central vision is still good. If you're following them for progression, clearly visual field is the way to go, and that's where you can document the progression. In a more severe group, when the visual acuity start to drop, that's when visual acuity becomes a better way of following the patients. I think people choose the criteria of 70 letters, which is equivalent to about 20/40 vision or so. It's the fact that, if you pick someone who has better vision, then you have a ceiling effect if you're doing a study. I mean, how do you include someone who's 20/25? Mm-hmm. When you talk about sort of losing peripheral vision versus the preserved central vision and this threshold of 20/40 vision, I guess what does that actually look like for a patient in terms of being able to maneuver in the world? It's sort of a slow process. In a young patient who, let's say, is a child, they're not going to have any symptoms. When their night vision, peripheral vision catches up with them, and when the peripheral vision encroaches on that central 20 degrees of central vision, that's when the activity of daily living is truly affected. That's when the patient bump into things. This early decrease in night vision may not be that apparent just because some patients don't realize they have decreased night vision. They don't realize what other people see. In looking back, they always have that symptom. Got it. What is the difference between Usher syndrome and non-syndromic retinitis pigmentosa? Why do we sort of group them together, and is there any difference between how the patients actually present in terms of symptoms? Well, I think Usher syndrome, when it's syndromic, it implies that there's also hearing loss to some degree. The rationale for that is because all the genotypes, genetic mutations in Usher syndrome affects the cell structure, this hair-like structure of the cell, which resides in the retina and also in the inner ear. When you have a Usher type 2A type of genotype, you can have syndromic RP, where you can have decreased vision from retinitis pigmentosa, as well as hearing loss. On the other hand, you may be nonsyndromic, you don't get the hearing loss. Conversely, there are even some patient with Usher syndrome that only have the hearing loss. Of course, we don't hear about it in our field. Okay, fair to say there's a good amount of heterogeneity in these patients and in the disease. How meaningful is stabilizing vision loss here in these patients? Has it been observed to be potentially reversible, or is really stabilizing the vision that is left the goal in treatment here? I think the first goal is to stabilize the vision and prevent progression or slow down progression. I think in terms of any efficacy, that would be one goal. Of course, I think a greater important goal in terms of demonstrating efficacy, is really to see improvement in some patients with some parameters. The rationale behind that is simply twofolds. One is that you can never conduct a study for a long time to look for differences to slow down the progression. Second of all, FDA also recognizes the importance that efficacy is always demonstrated with improvement in some patients with treatments in IRDs. Mm-hmm. That makes sense. I know in the STELLAR study of ProQR saw a six-letter benefit versus the untreated eye, in the combined populations of both advanced and early moderate. How meaningful is that in your view, and what is kind of the typical noise or variability in the BCVA measurement? Well, I think the variability typically is one line, which is five letters. Of course, that can fluctuate between four-five. I think six is meaningful. I believe when they look at the more severe group, where the vision was in fact 20/40 or worse, I think the number of letters was more like nine letters or something. I think that sort of helps in terms of demonstrating that for the severe group where visual acuity is already poor, that there is likely some efficacy signals. Mm-hmm. Why is it that visual acuity is not a good measure of disease progression in early moderate patients, just to walk through that again, and what exactly is the measure of static perimetry of the visual field? What does that tell us? How frequently is that used in clinical practice currently? The reason that the visual acuity isn't that good early on is because the visual acuity is really terrific. You can have a patient who is 20/25, and they can have a very small island of vision, and they're very debilitated in the early moderate time for the Usher type 2A. Static perimetry has been around since the 1980s. It is basically a computerized way of measuring your entire vision, what is called a visual field. The computer go ahead and test specific points in your peripheral retina and figure out the threshold. That particular threshold just means the minimum or the dimmest light that is presented at that location that a patient respond to. It's a very good way of assessing the peripheral vision. Of course, it is certainly much more meaningful in early and moderate stages. Toward the end stage, the visual field is so constricted that it's not really worthwhile doing a typical perimetry. How much variability or noise is there typically in static perimetry as a measurement? I think there is definitely some noise and also there is test and retest reliability. I think in general, people talk about 2 dB as a fluctuation in a person's reliability for short term. Of course, different studies are different, and sometimes for studies, I think you like to look at the untreated eye and see what the variability there is in the untreated eye. You can also do that with the treated eye. Definitely, if you have some sort of control of the variability, then that helps. Got it. No, that's helpful. Looking at the data that ProQR presented from this study on static perimetry, what do we actually mean when we're talking about the number of retinal loci with, I think they showed at least a 7 dB improvement, I guess. What does that actually mean, and how meaningful is that to you? I think that's meaningful and certainly, it's favorable certainly for an early analysis like this. They do have the 48 weeks out but I think they still need more data, and they have to have completed their data for all the subjects. I think for the differences between the treated eye and the untreated eye for the static perimetry, that amount of change, I think is significant. Got it. How, I guess, consistent is the deterioration of the visual field typically? Is that why you look at multiple loci, or would you expect it to be different at different loci in the retina? Well, I think you look at different loci simply because every patient is different, and sometimes people do analysis of the same loci, which means that in the beginning, if there are loci that are already non-seeing, then doing your analysis, that can become sort of tough. You can't really expect those loci to improve. The other thing is that the traditional FDA sort of bar for improving visual fields is like 7 dB and five loci. The reason for this is this dates back to glaucoma studies, which relies heavily on visual fields, and this was set a long time ago. The reason it was set that way is because that is sort of above the threshold of the test and retest reliability. That's why it's set. I think the FDA probably will continue to apply this paradigm, which is in existence for quite a long time. To your point, I guess, is there a precedent for using static perimetry as an endpoint in ophthalmology studies for registration? I think definitely. It certainly has done so in glaucoma. I think in inherited retinal disease, depending on which stage of the disease, and also, for example, at baseline, how severe the condition is. On the other hand, we also know that to do this test well, you definitely need to repeat it several times at baseline because you want to reduce the learning effect. I guess you can call the learning effect a placebo effect. A subject can, if they have enough practice, they can definitely do better. Typically, in studies for static perimetry, it's repeated three times at baseline. Has there been sort of a placebo effect observed with static perimetry? I guess how do you kind of avoid that by doing the test retest? Well, I think you try your best in terms of getting a good baseline. Subsequently, I think you compare the treated eye versus the untreated eye. In this case, I think you also have a sham group, which is almost like an additional control in addition to the untreated eye. That becomes sort of tricky. Definitely, I think this needs to be accounted for in terms of if you're looking for improvement in the perimetry, that what you like to do is you really want to have a comparison to the untreated eye in your control. That makes sense. I know another endpoint we looked at here was microperimetry, which is also looking at the visual field. How do those compare between static perimetry and microperimetry? A microperimetry is basically really looking at the very center part of your vision. The way it's tested is quite different from the typical static perimetry that is way out in the periphery. The microperimetry really hones in on the macular function. It's really on the central function. It's almost like a detailed visual map that is right in the center. Microperimetry at this point, the technology is so good that you can actually map it, and you can see where the loci are located on the retina. You can really follow the patients quite well, going back to the same points each time and see if the thresholds change. Mm-hmm. Got it. Okay. That's helpful. Then OCT imaging of the retina, how do you think about that in the context of also looking at the visual field and visual acuity? Well, I think that the OCT is such a good tool in terms of looking at the anatomy of the retina. It's almost like a in vivo biopsy of the retina, where you can look at different layers. I think it has several important advantages to be in these studies. First of all, you sort of will then know at baseline how advanced the disease is. Everybody hones in on this particular measurement called the EZ layer. The EZ layer, it's the outer segment of the photoreceptors. Photoreceptors converts light signals into nerve signals. This EZ layer is a measurement of this portion of the photoreceptor cells that actually converts the light signal. It's a good anatomical measurement. It can provide you with how advanced the patient is at baseline. It can also be a safety measure. It can also be an efficacy measure. What it is that for safety, you can make sure that, in fact, the EZ layer is not going down right after the treatment, because that would certainly be a concern, depending on what type of treatment you're looking at. You also want to look at the EZ layer to see if that can improve after the treatment. If you do see that, you want to compare to the untreated eye. The OCT is a very, very important tool. Okay, great. No, that's helpful. Looking at the phase I/II data, what were your impressions overall of the efficacy data at a high level? Also curious your thoughts on the differences we saw between the different patient genotypes as well as between doses? I would say on a high level, I think the data is very favorable because the safety profile is quite good. We do know that in Usher type 2A, there's a percentage of patients with cataracts and a percentage of patients with cystoid macular edema pre-existing, just as a part of the natural history of the disease. Patients, for example, they do develop cataracts much earlier, and depending on which group, it could be as high as 30%, 40%, 50% of Usher type 2A patients already have cataracts. Cystoid macular edema or CME. It's really an indication of inflammation and also some degeneration of the macular region. That's actually pretty prevalent in retinitis pigmentosa and certainly in Usher type 2A, and that can even be 20% or 30%, depending on which group you look at. I think the safety profile is good because one concern is that you have a gene therapy that has to be repeatedly injected into the eye. While the consequences or concern is inflammation or reaction to the treatment, that can make the cataract worse or make the cystoid macular edema worse. I think their data is pretty favorable. They have a patient with a preexisting CME that perhaps got worse, but that could be the natural history of the disease. I think they also have patient with cataract, but it seems like the safety profile is quite good. We do need to understand, though, the data so far involve one injection. It would be important to really wait when the study is further along into, let's say, phase II-III, where you have repeating injection. Right now, I think based on the data they have, that's actually quite safe. Looking at the efficacy, as we already discussed it, I think the BCVA is favorable. I think the static perimetry is favorable. The question then, of course, is some of the efficacy signals they are seeing, are they going to be sustained with repeat treatment, which is how this treatment is intended. On the safety, I know, here we saw the cystoid macular edema, which does occur as part of the natural history of the disease. I guess when we talk about inflammation in the eye, I know there's a lot of different various types, and I guess what can tend to occur with any sort of intravitreal injection, and then as we think about CME versus uveitis, vasculitis, other types of inflammation, how should we contextualize that? Well, I think we know that intravitreal injection is relatively safe, but you can certainly cause a retina tear. It can also cause a hemorrhage. Fortunately, we know repeated is safe. On the other hand, we also know that when you have injection, aside from cataracts and cystoid macular edema, which could worsen because of the inflammatory response, you can also have vitreous cells. Occasionally, you can also have some spillover cells into the anterior chamber. I think you also need to make sure that there is no changes on the retina on the OCT that would suggest inflammation. Not just the EZ layer, but looking at the other layer to see if there's any effect, any thickening, or unusual findings. When you have a treatment that's intravitreal, clearly one concern is the inflammatory or the immune response. Now you can treat this immune response, and if you do see it, if it's mild, you can definitely use topical steroid. Mild immune response is not a super concern, but if you have a persistent inflammatory response or a recurring one that's substantial, then I think that may require a little bit more prophylactic treatment to deal with that. You mentioned for CME, the typical kind of standard of care management for a side effect like that or if it occurs just as part of the disease is something like a topical steroid drop. For cataracts, what is the recourse like for that? For cataract surgery, basically, when it's bad enough, you go ahead and do a cataract surgery and put in an intraocular lens. That technique, fortunately, is very well established, even for younger patients, even in children. At this point, we could say that a cataract is something that's treatable, and that is probably not a consideration. On the other hand, if you have a worsening cataract and it could be related to the treatment, then that may suggest the inflammatory response, and then you need to look for CME and look for the other signs of inflammation I just mentioned. Got it. That's helpful. In the general Usher and nsRP population, what is the rate of cataract surgery? Well, I think it's relatively high as you go into the more advanced group as time goes on. Very early on, patients tend to get this posterior capsular cataract, which means that overall, the lens in the eye is fairly clear, but the back central portion gets a little bit speckled, a little change. At that point, this is not visually significant in terms of reading the chart or for your peripheral vision, but it really gives a patient some glare. This glare is particularly noticeable at night, for example, with a headlight or going into the sun. From there, the natural history is that it just gets worse, and then subsequently, you have the nucleus of the lens getting involved, the cortical area. Still, the posterior capsular area is preferentially involved, so that by the time a patient into their, let's say, 40s or 50s, they have cataract surgery rather than the typical normal population into their 60s, 70s, and 80s. Mm-hmm. Got it. Okay. In the context of the risk-benefit of a disease like Usher syndrome, I guess, how do you think about what an acceptable side effect profile might be as we see repeat dosing studies and longer-term follow-up. I guess, what rate of cataracts and CME would be acceptable in your view, and then how much follow-up would you need to see to be comfortable with that? Well, I think Usher type 2A is a devastating disease with a great unmet need. I think, when we talk about risk and benefit ratio, we always have to keep that in mind. This is a progressive disease that really makes patients legally blind and then subsequently affects their central vision as well. I think the bar for this is that you have an unmet need, patients are going to seek care, and that the complications or the side effects that you have for treatment, as long as they are treatable, as long as they are mild, then the benefit outweighs the risk by far. Fortunately, cataracts can be treated, inflammation can be treated. Again, I think, in this case, we still need more data. We don't have repeat injections. We need a little bit more repeat injections in, for example, every six months, for example, and then follow the patient for a couple of years and see if there's any other inflammation that may occur. I think based on the dosages they use, at least they didn't really find a dosing effect. Maybe the doses they are using at present are at a good level where it doesn't really cause much inflammation and cataract. I think that's good. Yeah, I think that's one important thing to be looking for when we get the patient-level data at some point here, is to see at what dose levels those events did occur. There may be a wide dose window here to select a lower dose to move forward into pivotal, it sounds like. I know you touched on the dose response, were you surprised to see that there was also no difference in efficacy between homozygous versus heterozygous patients? I think, first of all, homozygous is just so rare. When we talk about homozygous in Usher type 2A here, we're really talking about both copies of the Usher type 2A gene have genetic mutations in exon 13. Of course, we know this particular treatment targets that exon 13 in terms of the messenger RNA. Homozygous, theoretically, if you treat them, you will have much effect. I'm not surprised that the patients who only have one single copy or allele with the exon 13 change, in fact, did just as well, just because you don't necessarily have to block all the messenger RNAs. Clearly, if you just block the exon 13 and make some of the messenger RNA work, I think there's a good scientific rationale you could improve function. I think at least it's actually favorable that both are the same, because if one is a lot more, let's say the homozygous, there's just not that many patients out there. I think the heterozygous patients are far more common than the homozygous. Mm-hmm. Got it. No, that's helpful. Then, on the dosing regimen, it seems like based on the half-life of the drug and all of the data we saw here supports six-month dosing. How frequently are these patients coming in for checkups and visits? How much of a patient burden would every six-month intravitreal injections be? Well, I don't think it's much of a burden every six months. Every six months is still twice a year. It's not so frequent. I think it's something that is not difficult to do in terms of intravitreal injection because it can be done in the office. It's not quite a difficult procedure. I think in terms of the practical aspect, that if this treatment is efficacious and safe, I think in terms of giving the patient a treatment, that will work fine. Also, these patients are highly motivated. This is really a bad disease that for them to go to a doctor's office and really get this treated twice a year, I don't think that's a problem. Mm-hmm. Got it. That's helpful. As we do move into repeat dosing studies, I know the patients in this STELLAR study are rolling over into an open-label extension where they will be dosed again. I guess what would you be looking for in that long-term follow-up on both efficacy and safety? I guess what time point would be long enough for you to be comfortable with that? Is that 48 weeks with two injections or what do you think about there? I think having the two injections and the 48-week data is certainly going to be quite helpful because you can then compare to the one injection at 48 weeks, and then really look at your BCVA static perimetry, look at your EZ layer, and then to see if the efficacy that you do see in the single injection, whether that is improved by the second injection. Clearly, I think you'll be also good to have those patients get that third and fourth dose and continue to be followed because that's important, not just for the efficacy but for the safety. I think you sort of need data clearly beyond 48 weeks eventually and probably longer course, that would be up to what the FDA determines eventually. You do want to have some long-term, just because you want to know the safety a bit better. Mm-hmm. No, that's helpful. Looking at the designs for the two pivotal studies in advanced and then early moderate patients, which ProQR has outlined. I know the primary endpoint is at 18 months. Do you think does that time point make sense to you in terms of kind of the rate of disease progression and the efficacy we saw here? I know there's also potential for an interim analysis at 12 months. Just curious how you think about those time points for showing benefit? Well, I think that time point is reasonable. I think at 12 months, you have the two injections, and then at the end of the study at 18 months, you would already have three injections. I think based on after three injections, clearly you would have a lot more data on the safety and the efficacy. To really lengthen the study, I think then you have to really have a good rationale for that. I don't think any of the repeat treatments, even in AMD, for example, you can really do a study, let's say, greater than two, three years. I think just that's unreasonable. On the other hand, the original group of patients who gets treated, let's say, at the 18 months, could certainly be in a follow-up study if that's necessary to follow that particular cohort. That's what they did in RPE65 LUXTURNA, where you finish the phase III study, and then you still follow the patient for long-term durability. That clearly would be helpful. Mm-hmm. Got it. I know we've spent a little time on just the challenges of looking at a sham given the heterogeneity in this patient population. Can you just expand a little more on kind of why the best control here is the patient's untreated contralateral eye? When you look at data versus the contralateral eye versus the sham, I guess, how do you think about those in terms of clinical meaningfulness and which is more meaningful to you? Well, I think at this point, there's really no evidence of any possibility for an antisense RNA to travel to the other eye. You're talking about very small molecule, and it sort of gradually, based on the pharmacokinetics, we know that the presence gradually decreases. This is a little bit unlike the AAV gene therapy, which there are some demonstration that in non-human primates, there is some AAV virus going to the other eye. In this case, I think the untreated eye definitely is a reasonable control because it's really based on the patient's sort of progression and stage of the disease. When you do a sham, I think that they did include them here, which I think is quite helpful. It also does show that the sham eyes, both of them, as well as the untreated eye, are sort of similar. I think it's just very hard to match in terms of severity. You have to with sham and also with the treated eyes. I don't think that's something that's easy to do. You have to try to have criteria matching the static perimetry, matching the visual acuity. I think that becomes very difficult. I don't think that's really necessary in this case. Mm-hmm. No, that makes sense. I know often in ophthalmology studies for registration, the FDA likes to look at, in addition to sham, a lower dose as well for masking purposes. Can you just explain that? Well, I think that's probably expected. First of all, the two doses you can learn a lot, and I don't think there will be any difference here. First, the doses sort of tells you two things. One is that the lower dose works just as for the higher dose, why do the higher dose? The other thing is that by having two doses and having different data, you sort of get a sense of the sort of a dose response. Clearly, you're not going to have a third dose. You can definitely learn. You also learn more about the safety and the efficacy. I think that's the traditional paradigm. I don't think there's any deviations that will occur from that in general for clinical trials in IRDs. I think in this particular case, probably it'll be interesting which doses are sort of chosen since there doesn't seem to be any differences between the doses. We'll just have to see what happens. If QR-421a is ultimately approved and the efficacy and safety that we saw here in STELLAR is maintained in phase III studies, what percentage of Ushers and nsRP patients would you expect to receive this therapy? I think it's going to be a high percentage. I think certainly the early and moderate and the advanced patients that are sort of included in this study will want the treatment. There are going to be patients also who are count fingers, who are off the chart, who are worse than 20/800, and who had Usher's type 2A a long time. They're older, and they're in count fingers hand motions, and even LP. I think some of the patients definitely would want this treatment as well. This particular low vision group. Unfortunately, for LUXTURNA, for example, there's sometimes concern about treating patients with very advanced disease with a subretinal injection because you have the potential of, in fact, making the patient's vision worse because of the surgical trauma. In this case, I think that's less of a concern because the intravitreal injection is less invasive in that maybe it's also possible to treat that group of patients. We also have to recognize the fact that that group of patients may be difficult to identify just because they may already have given up on the system, because they've been told there's no treatment for a number of years. Some of them are still followed. Some of them have been genotyped, but many of the patients with advanced visual loss from RP simply may not be seeking care regularly or getting a genetic testing. I do think that the vast majority of patients who are in the early moderate severe group will be. The issue here is then is the genetic testing, is where they are, and there's really quite a big regional differences between how a patient with Inherited Retinal Disease seek care and who's seeing them, and who is more aware of the genetic testing. I think that's sort of down the line. Of course, young children, there will be a question then about younger children, let's say less than 12, just because they have a sibling who has Ushers, and then the parents decide also to see if the other younger sibling may have it, how young the FDA would recommend that. Clearly, the younger you treat a patient, the better off they're likely to be. What is the diagnosis rate, do you think, in Usher syndrome? Is it that once patients have symptom onset, it's usually diagnosed pretty quickly as this being the genetic cause of the disease, or are there patients who are under care of kind of a general ophthalmologist for vision loss without a real diagnosis of the underlying problem here? Well, I think there's really a significant percentage that may not be diagnosed. Even though this condition is recessive, there's often no family history because you can have a sporadic case. Only one person is affected. There are no siblings affected, or there are no siblings, period. Sporadic cases are not uncommon with Usher's type 2A. In fact, it's unusual. We do have families where multiple siblings are affected, and in sporadic RP, it's quite common. The hearing loss is a tip-off. Usher's type 2A, they do have congenital hearing loss, but the severity is variable. There could be an Usher's type 2A who walks in who have really not much hearing loss. They have sporadic RP. The doctor will see something, there's sporadic RP. I think it's really, hopefully, the doctor gets the genetic testing, that's not always the case. Some of the eye care providers are very much into knowing how to do genetic testing, some of them really don't do it and don't really, are not educated in this field. That make it difficult. The hearing sometimes could make the patient just have a little bit of a babbled speech, they can still hear. The hearing loss can, in fact, be missed sometimes. Got it. It sounds like it's really about finding these patients early enough where you can preserve vision at a very high level. Are there other Inherited Retinal Diseases where there have successfully been genetic testing programs established? Well, I think the prime example is LUXTURNA by Spark. I think they put out an IRD program. They certainly has expanded that into now 200 genes or more. They have worked with other genetic labs to really try to do that in different countries as well. I think you sort of need a genetic program like that to help to identify patients. There are also Foundation Fighting Blindness, My Retina Tracker, where one can get no-cost genetic testing. There are also many places that provide commercial genetic testing at a reasonable price now, some of them are actually covered by insurance. I think this is an expanding field, and it's sort of like a revolution since the approval of LUXTURNA in December 2017. I think this aspect is improving. Mm-hmm. No, it sounds like a lot of progress has been made. Are there any other therapies in development for Usher syndrome and nsRP that are promising in your view? There's nothing at this point that's into human trials. I think Usher's type 2A is a sort of a big gene that you cannot really fit onto the conventional AAV platform. If you want to do gene replacement therapy, that becomes a challenge. You need to try to do a dual AAV or there's no clinical trials that has shown that the lentivirus may be a good platform for Usher's type 2A, for example. I don't think that. I think, however, this treatment, highly effective, only covers those with exon 13. I think it's really important to get the genetic testing because this treatment would not be suitable for somebody with 2A and doesn't have exon 13 mutation. Mm-hmm. Got it. No, that makes sense. I know you touched on this briefly just with the potential risks of a subretinal procedure when you talk about gene therapies and doing that in a very advanced population that might have limited benefit. Can you just walk us through the differences between an intravitreal injection for an RNA therapy versus a subretinal procedure that's often used with gene therapies? I think, the traditional subretinal gene therapy is that you go ahead and you have a viral vector, which then carries a normal copy of the gene of interest, and then this then is injected under the retina. The reason you inject it under the retina is simply we don't have the technology to do an intravitreal injection of gene replacement therapy in human beings. There is so much barrier, including internal limiting membrane, for the AAV virus to then reach the photoreceptors, which is, of course, deep in the retina, in the outer retina. Typically, at this point in time, the traditional gene replacement therapy remains a subretinal injection. Subretinal injection, of course, can only treat part of the retina. You cannot inject under the entire retina. At most, maybe you can treat 20% of the retina or 25% of the retina. Much of the retina may remain really untreated. There are many advantages to gene replacement therapy as well is because if the treatment is successful, like in RPE65, it is one treatment. You don't need to repeat it. The normal DNA is then introduced into the photoreceptors, the RPE, and then the normal DNA then continue to be transcribed. The intravitreal injection with antisense, you are not really trying to replace the DNA. You're trying to sort of target the messenger RNA so that the messenger RNA product is actually unaffected, and you block that with the antisense RNA. You definitely have to repeat it. The good news for that is that you can treat the entire retina. You can also much non-invasive, so you don't have the risk of doing a subretinal injection. On the other hand, for this particular disease, for a particular inherited retinal disease, you probably have to have inherited retinal disease where there is a segment of the gene that where most of your genetic mutations occur. If you don't, if they're scattered around, then it's harder to do an antisense treatment. Mm-hmm. Got it. I know while exciting, these gene therapy and editing approaches are not yet in the clinic for Usher syndrome, but I guess once you reach that point, how much follow-up would you want to see in order to be comfortable with both the durability and the safety of a gene therapy? I think in terms of safety, I think at least you need 18-24 months of good safety data. I think the durability always becomes a little bit tricky because there's also a long-term durability. I mean, is the treatment truly, how much is it decreasing the progression? It's difficult to say at this point in time that gene therapy can indeed stop progression altogether. I don't think we know that. There are RPE65-treated patients who have been treated for a long time who still have better vision than baseline. In terms of years later, given the fact that the retinal degeneration will still continue, then what's the durability? We don't know that. I think we also have to look at that based on this is such an unmet need that even if you decrease the progression some, even by 50%, by 70%, by 80%, is worthwhile. You're talking about in terms of quality of life years to extend that, and that's certainly quite meaningful. Mm-hmm. Yep. Again, just given kind of the unmet need here and there being absolutely nothing, I think it's clear that even a chronic therapy that could stabilize vision would be a huge advancement in the Usher's field. I know we sort of talked around this, but just in terms of the percentage of patients that you think would be put on a therapy like QR-421a if it were approved today with the existing profile, what number would you kind of throw out there? Oh, in terms of the patients who would be treated? Well, I think of the identified patients, can't discuss unidentified, but of the patients who are identified with the exon 13 mutations, I think probably something like high number, 80%, would be treated or would want to be treated. Some of it may depend on the FDA criteria also. I think FDA criteria will likely be more somewhat lenient. They were certainly lenient even for RPE65. You can treat patients as young as 12 months, and the oldest patient or the youngest patient in the phase III study was four years. I think the FDA will likely, if it feels a treatment is efficacious and safe, will probably have a broader criteria so that the vast majority of the patients will be treated in this case. Great. Well, I think that is the extent of the questions we had, Dr. Lam. This is really helpful. We appreciate your time, and we're looking forward to seeing more data as this program progresses. Great. Well, thank you. Thank you. Thanks. Bye.