Taysha Gene Therapies, Inc. (TSHA) Earnings Call Transcript & Summary

[Audio Gap] here at RBC Capital Markets. And today, it's a great privilege to have R.A. Session, II, President, Founder and CEO of Taysha; and Suyash Prasad, CMO of Taysha as well for a fireside chat. Guys, how is it going today?

Good, good. And we really appreciate being on with you today, Luca. So thank you to you and then the rest of the RBC team. It's been a great day.

That's fantastic. Great to hear that. So maybe, R.A., let's start with the GAN opportunity here. Obviously, you've done a recent transactions there. You in-licensed the product there. Can you maybe just briefly recap the clinical data that you have seen today? And maybe bigger picture, if you can talk about why that transaction can be transformational for you and shareholders?

No, it's a great question. So maybe I'll take the second question first, and then I'll pass it off to Suyash to talk about some of the breadth of the clinical data, which is really extensive and only which we just shared a portion of it with The Street, which was focused on the MFM32. But really from a strategic fit, this transaction made a lot of sense. This is a program that we were fortunate the beneficiaries of a little bit of proprietary deal flow. We've known about this program for quite some time. The giant axonal neuropathy program is a groundbreaking program in itself, it's the first intrathecally dosed gene therapy in history. The study initiated in 2015 when the first patients were dosed. They've dosed 14 patients to date at 4 different dosing levels. And we're seeing really nice safety, durability, clear halting of disease progression at therapeutic doses, a pretty extensive frequencies as well as BeiGene analysis -- statistical analysis was done. And just overall, the totality of the data really points to a drug that is having a disease-modifying benefit and one that we're excited to be talking to regulators about. Not only did we have just a huge breadth of data across the clinical data set, but also the preclinical data set. But there's a huge natural history data set associated with it, of which we shared about 45 patients worth of natural history data. That study kicked off in 2013 and really acts as a nice comparator arm to the interventional study. And so Suyash will go through some of that. But really, the program not only, I think, accelerated the company from a kind of a clinical stage company to a pivotal stage company, one that is now going to be embarking on discussions with regulators around an approval pathway, both here in the U.S. and ex-U.S., but also would help validate our scientific approach. This program, from afar, helped really guide the way that we're thinking about the rest of our portfolio, what kind of our focus on endothelial delivery, AAV9, HEK293 suspension manufacturing. This giant axonal neuropathy really helps validate that. But also this notion around being able to intervene using gene replacement therapy and kind of those key variables into a monogenic disease at a CNS and one that's fairly progressive. And so we think it provides a lot of read-through to the rest of our portfolio. There's a lot of similarities between giant axonal neuropathy and GM2 and CLN1, that they -- this issue around the clearance of some sort of substrate that access a damaging agent to neurons and you kind of have this ongoing neuronal damage that ultimately leads to death, very similar to the other indications that we're going after. And we think it provides some nice read-through. So I'll stop there. Suyash, maybe you want to talk -- give your thoughts on the program as a whole and maybe you want to talk about the clinical data.

Sure. I think you've covered a lot of it, R.A. But I think the main point is that this is a study that was started at the NIH in 2015. And it's a really awful disease, giant axonal neuropathy. The children are born, they look fine at birth. Right around the age of 3 or 4, you start to see an issue around sensory ataxia. Meaning the children can't quite feel the floor underneath their feet. They lose sensory capability. So they walk with a wide base gate. They start to sway, they may fall over. This progresses over time. There's a [indiscernible] that develops, resulting in the children needing a wheelchair by around the age 10. They'll often [indiscernible] ventilate first time at 15 because of the progressive ongoing nature and muscular weakness. And then they're often passing away around the age of 19, 20, 21 years of age. So it really is an awful devastating disease for relentless progression. And this program, the TSHA-102 program, while with gene replace the gene that's missing, the gigaxonin gene is missing, this involving the metabolism of a neurofilament protein in the cells, in neuronal cells. TSHA-120 shows clear arrest of disease stabilization. The NIA -- the natural extrusion [indiscernible] shows a decline of 8 points per year on the MFM32, a 4-point change is clinically relevant. And this 8-point change becomes flat, it stops at the [ 1.1814 ] dose. And we have remaining the higher dose yet to come. In addition to MFM32, there's a whole array of other endpoints that have been looked at, in particular, in the conduction studies, in biopsies, a whole host of sensory examinations, all of which show improvement in the same direction. Thus resulting in an improvement in the totality of data for this particular trial. 14 patients dosed. We'll be starting the regulatory discussions this -- later this year. But it really checks all the boxes from a regulatory perspective, great natural history data, dose response, stabilization of disease at high dose, 3 years of long-term safety, efficacy and importantly, durability data, which is important for gene power. I'll stop there. There's lots more to talk about it, but let me stop there.

Yes. I know this is actually super helpful. And maybe, Suyash, since you just mentioned the high dose coming later in the year, I think the 3.5x1014 is coming later in the year. What's the bogey for success for that dose? How should we think about the data in the context of this higher dose? Any thoughts there?

Yes, I can make a comment and R.A. jump in as well. I mean basically, the 1.1814 dose, we're showing success. We're absolutely showing success. A 4-point change in the MFM32 is clinically meaningful. We're seeing an 8-point change in the trajectory at the [ 1.8x1014 ]. My guess is we're going to see at least that perhaps even a little bit of -- a little greater improvement. The safety profile in all 3 doses thus far online in a suppressive regime, which includes rapamycin as well as prednisolone. And we can spend a few minutes talking about that if you're interested. On that particular regime, we've seen no T cell-mediated inflammation, no complement activation, none of the other immunological consequences that are often associated with the gene therapy. So I don't anticipate any safety issues. If anything, we're going to see a slight, slight increase in efficacy over [ 3.5x1014 ] dose.

Fantastic. And maybe you mentioned it earlier, I want to circle back on that one. I think you're planning to meet the regulators later in the year for an end of Phase II meeting. Walk me through what's on the agenda for that meeting. What are some of the key topics of discussions? And most importantly, how are you thinking about the Phase III? What's the design of the Phase III? What's the key endpoint to get this molecule over the finish line?

Yes. So let me start and then Suyash can make a comment around the clinical development plan. I think you almost have to look at the U.S. and ex-U.S. in 2 distinct pathways. So in Europe and with the EMA and the MHRA in the U.K., there is a very clear distinct conditional approval pathway. I've been fortunate to get a number of programs through that pathway. Suyash has done that as well as in his past. And essentially, this data set is tailor-made for that. There's no therapeutic alternative, fatal disease, progressive, time is of the essence. It's essentially time is neurons, right? And so when we go and have initial discussions with the EMA, it's really going to be around meeting the criteria for that conditional approval pathway in the fastest way we can go there. If they tell us to go file for full approval and it's under that same time period, then obviously, we'll do this. But that will be our go and end strategy. And hopefully, if successful, that makes for a 2023 launch timing ex-U.S. And as you know, if you're able to secure an approval or a conditional approval either way in Europe, you're then able to leverage that ex-U.S. and to do reimburse name patient programs in Turkey, Middle East, Israel, Brazil, Colombia, all these great markets for rare diseases where you're able to secure meaningful reimbursement. So that's our ex-U.S. pathway. In the U.S., we see it kind of laying out in 3 scenarios. The first is exactly what Suyash said. There's not very much that you could go in and change from this study. It was done really, really well, and it's been ongoing for a long period of time. You got dose-dependent data, you've got good safety, you've got good durability. You've got clear halting of disease progression. And so that's what we're going to go in and have a discussion with the agency on. The first scenario would be -- and this would ultimately be the one that we would like to propose, but who knows what they would say is, go do analytical comparability with your commercial grade material. We're fortunate that the clinical material was done manufactured using HEK293 suspension. We're going to manufacture using the same process with the same CDMO partner that manufacture the clinical trial material. So essentially a like-for-like process. So the hope is that the FDA come back and say, "Hey, just do analytical comparability, use that as your module 3, file the BLA, go." That would make for an early 2023 launch or first half 2023 launch. The second scenario would be if they come back and say, look, the data looked great, all the things from the first scenario, but we like for you guys to dose call it, 2 patients, 3 patients using the commercial-grade material, just to make sure it's comparable. Give us 2 months' worth of dosing, use that as clinical comparability and update your Model 3 filed with BLA. That's your second scenario. That would make for a late 2023 launch, most likely. The third scenario is unlikely. We see it as unlikely, but obviously, we got to have a conversation with the regulators. They are the ultimate decision makers here. And that would be -- you present the data set to them. They say, "Look, everything is great, but we'd like for you to do a confirmatory study." Based off the guidance that the agency issued at the beginning of this year for the development of neurodegenerative diseases in gene therapy, this study checks all the boxes to pursue accelerated approval. So we're going to go in using that guidance to guide the discussion. But ultimately, that's the way we kind of see it working out in those 3 scenarios. Suyash, maybe you want to comment?

Sure. I think you hit the nail on the head there, R.A. I think we feel very positive about the interactions with the regulators. And the reason is, you're probably well aware of this guidance that came out by the FDA earlier this year on red -- gene therapy development for neurodegenerative diseases. And it talks about TMC matters, preclinical matters and clinical matters. And essentially, we were very, very reassured by that guidance because it actually really fitted in very nice with our approach across our whole portfolio of programs. But importantly, it definitely meets all the needs for GAM. For example, one of the questions we often get asked about GAN, is well, it's not a randomized controlled study. You've got an artery study, how is that going to work? FDA likes control studies. But the FAA guidance specifically talked about if there's an unmet medical need, if it's unethical or impractical in the clinical control, if the disease course is well-documented and if the expected treatment effect is large, you don't need a controlled study. You can do it based on solidly collected prospective natural history data, which is what we have. So we feel very confident going through these discussions with the regulators. And as R.A. says, it's going to be a little bit different from the U.S. to Europe to the U.K. and Japan. There's different approaches in Europe and the U.K. And actually, probably things will go even smoother than there than in the U.S. based on the different paradigms, the different kind of procedures that they have there. The ultimate thing is, is the drug safe? Is it effective? Safety looks impeccable particular -- in this particular program. As I say, no immunological consequences on the prednisolone/rapamycin combination. And efficacy is [indiscernible]. Got clear, clinically relevant change in every single patient, this is confirmed both of the standard frequent analysis and also with a more robust intricate based on analysis, which has been applied throughout the whole study as well. And most importantly, an area that companies often get tripped up on is the longer-term safety durability. And we've got at least 3 years' worth of data in the scan program. I mean the study started in 2015, and take their time, they do things very, very robustly. So we have been able to really benefit on the back of that robustness and diligence that the [indiscernible] have taken.

Got it. Got it. Super helpful. And I know I'm probably pushing my luck here, but I'll give it a shot anyway. So I think you laid out 3 really good scenarios, right? Scenario A and B seems like the ideal. Scenario 3, obviously, will require a little more time and energy. And maybe I think you possibly hinted at the Scenario 4, where maybe you need to run a control study for Phase III. Should that be the case, in the unlikely scenario that's the case, how are you thinking about potentially designing that trial? Again, big picture, will it be a control arm that makes sense? What will be the size of that trial? How should we think about that?

Yes. Luca, I actually think that's highly unlikely. I know you want us to answer the question. But I think it's one of those scenarios that it's like less than a 10% probability that it happens. Now watch it happen. But I just don't think that, that the regulators, based on the guidance, would go that route. What I would also say is, keep in mind, and Suyash brought this up, the robustness of the natural history. In the interventional study that we've been running, each one of the patients have their own natural history cohort because they were followed for at least a year before they went into the study. So you actually see their pretreatment rate of decline, and then you see their post-treatment stabilization. So they're all their own control arm really. So you don't have to necessarily do -- well, you didn't have to do age maps control because each one of the patients are their own control. You saw their decline before they got treated. Some patients were actually -- you have to have 2 years worth of data so you get a pretty consistent rate of decline and then a clear stabilization. And then for most of these patients, we have more than a year, and a lot of them we have 2 years or 3 years worth of data. And so you kind of get that long-term durability. So I think when you start to look at the totality of the data that's out there, I think it's going to be highly unlikely. I think what we would probably try to do, if we were asked to do more on the scenario 3, if you were asked to do a confirmatory study, it would probably be more of a run-in study where you would let the patients serve as their own control. But Suyash, maybe you want to comment.

Yes. Even if they asked us to do this, which as R.A. say is very unlikely, it's clear what the endpoint should be. It's clear what the population should be, to study based on the data we have thus far. We have such a wealth of data in this study. It would be a pretty simple study. I mean, if I had to have a randomized 3:1 drug to placebo, you would then get placebo, the open-label delayed treatment comparator [ fair ] observation grid of about [indiscernible] months. It's really very, very straightforward given the wealth of data we have. But it's highly unlikely. I think the other important piece to why we think this is unlikely is the breadth of assessments. We've got MFM32 with a whole bunch of other motor measures, i.e., the time to 4-stack climate, the time 10-minute walk, the time to get up off the ground and stand up straight. We've got sensory assessments, the sensory impairment score. We've got neurophysiological assessments, no conduction studies. Not just of 1 nerve, but of the 3 nerves in the body, perineal nerve, the nerve innervating the thumb and the nerve in the shoulder. We've got, importantly, pathology data as well, which is really, really hard to get in the clinical trial. We've actually taken -- or our colleagues at the [indiscernible] have taken nerve biopsies of the sural nerve and of the nerve in the wrist, there's a superficial radial nerve. So look at change over time before treatment and after treatment and we've shown the pathological improvement that mirrors the clinical improvement. So -- and actually, there will be a paper coming out from that history at some point in the next -- in the near future that looks at all these endpoints in some detail. So we're really quite confident that it's going to be either scenario A or scenario B.

Got it. Got it. Super helpful. And then maybe I want to circle back at something you guys mentioned earlier. In addition to steroids, they're obviously using rapamycin as part of prophylactic treatment here. Again, a T-cell depleter. Can you remind us the rationale for why that's important and potentially differentiating versus others?

Yes, absolutely. Suyash?

Sure. Yes. So it's interesting that the use of immunosuppressive agents in gene therapies evolved over time. In the very first hemophilia studies, they didn't use any immunosuppressive therapy. They notice just a few weeks after dosing, patients would spike up at their LRDs. They treat that reactively, because it looks like an [indiscernible] hepatitis with steroids and the arteries would come down again. So that move to pre-dosing with steroid for a short period of time, 4, 5 weeks. They realize they have to increase that longer to maybe 3 months, maybe even longer than that. And then they still got some breakthrough inflammatory elevations of liver function. And so additional immunosuppressives have also been trial. Now with the GAN study, initially, it was 2015 when they dosed the first patient. And they started off with just a short course of prednisone [indiscernible] because that's how the weather field was at the moment. After the first 1 or 2 patients, on the subsequent CSF samples that were taken at 1 month and 3 months after dosing, a very mild white sock count elevation was noted in the CSF. This had no clinical sequel line. Because of that, the principal investigator, in conjunction with the DMC, decided to extend the steroid course. And then that wasn't enough so they added in rapamycin as a T-cell modulator, which is quite a benign drug. I mean it's been used now for about 20 years in helping to manage patients about solid organ transplant. Lots of affairs in pediatrics and children. I prescribe it myself. The only side effect that's really troubling is occasional mouth ulcers. But apart from that, it's really a very safe, very well-tolerated drug and was very effective. And so ultimately, after the first 2 or 3 patients with [indiscernible], the way the field was moving cost and the cost [indiscernible] and the PI decided to implement a regime of 6 months of steroid, 4 months of 1 milligram per kilogram, followed by taper off, plus 1 year's worth of rapamycin, 10 months full of [indiscernible]. And that [indiscernible] sort of not or heard any immunological consequence. So there's no white cell [indiscernible] in the CSF. There's no T-cell-mediated inflammation of the liver, troponin elevations. And that seems working very, very well. And in fact, we've applied that now for the whole of our portfolio of programs at Taysha.

Yes. Luca, the only point that I would make is, in addition to what Suyash mentioned is adding in rapamycin not only helps with the kind of breakthrough immune response, but it also helps from a transduction efficiency perspective as well. And I think we're starting to kind of now realize this in gene therapy. And I always joke with Steve. Really, this program not only helped inform the development pathway for our portfolio, but it actually helped inform the development pathway for a lot of monogenic CNS gene therapies out there at current development because they're now starting to apply some of these learnings from this particular program in Carston and Steve, which they pioneered, this is way back in -- when they probably initiated this immune regimen, it's probably 2016, 2017. So we're going on probably 4 years now.

Got it. Got it. That's super helpful. I didn't know about the increased transduction efficiency there. We could have maybe a detailed discussion on the biology maybe a separate time. Maybe another question when I make sure I ask. Obviously, you guys are going AAV9, primarily intrathecal. I think it makes a lot of sense as [indiscernible] is approved with a similar approach and displaying critique, a lot of it, news for SPINRAZA and some of the others. However, I think the Zolgensma is still on clinical hold for adults using the intrathecal delivery. And we are seeing companies become more creative using interest external and prothalamic and whatnot. What are your thoughts on that? And what are the implications for your program here?

Yes. I don't want to comment too much on another company's program with the exception of saying, "I think if we could hope to have a clinical program, any clinical program in our development, it would be that Zolgensma clinical program." The data coming out of the intrathecal trial was phenomenal. They presented in March of last year. They saw twofold increase in the Hammersmith of what would be considered clinically meaningful. It was sustained, it was durable. There were over 1,000 patients being treated in the commercial setting at this point and it's performing just like it did in the clinical setting in our hands back in our former life. And so I think that program in itself really helped validate what the potential of gene therapy could be. With that being said, I think the intrathecal approach, from a route of administration perspective, is not a one-size-fits-all, but it's a one-size fits many. And it really helps inform the diseases that we're actually going after. So if you look at our portfolio, it's a big portfolio, but there's a lot of commonalities between each program to the next. The central commonality is these are broad CNS disorders. What that means, there's a movement disorder. There's a cognitive disorder or delay. There's a developmental issue, there's seizures. There's -- they have -- there's maybe some sort of other issue that kind of comes up, that kind of blinks one program to the next. And basically, what it means is we're trying to hit the CNS broadly versus hit one area of the brain over another. And if we need to get more drug into one area, we'll be -- essentially what we'll do is push the dose a little bit. We'll kind of manage it through dosing. But essentially, what we're trying to do is hit the CNS broadly. And what we know about biodistribution of intrathecal is about 70% to 50% of the drug stays within the CNS, circulates through the CNS. And about 30% to 50% of the drug actually gets dumped out into the peripheral nervous system or the systemic system. And this gives us a little bit more bang for your buck. Not only are you hitting the CNS and you're able to evade neutralizing antibodies and you're treating the disease exactly where it's happening, but you also cover some of the peripheral or systemic issues of a lot of these diseases, which a lot of these diseases have that go along with it. And so it's a little bit more bang for your buck, without having the issue around systemic dosing, which what we're seeing complement activation, elevated liver ALT, these things like that. We don't have those risks because essentially, we're giving a very targeted dose directly to the area that we're looking to treat. I'll stop there. Suyash, I know you had some thoughts.

Sure. Yes, I think the other important thing about intrathecal delivery is it's been used in clinical practice for decades. I mean, I myself have given intrathecal drug in my pediatric practice, not gene therapy, but anesthetic drug in oncology drug. And you have to hit the whole of the brain and the spinal core to the drug and it works beautifully. And there's now clinical precedent. We've got the GAN study data we've just shared with you. It's really stunning data in 14 patients, all intrathecally dosed. CLN6 and CLN3 from Amicus also intrathecally dosed, very nice data there, Zolgensma IT. So the clinical presence of this route of delivery targeting global brain and spinal cord disease and more so than any other route of administration. Having said that, if we felt this strongly an ICM route or an ICV route will be better for a particular disease, then we would, of course, do that. In fact, in much of our preclinical work, we do actually compare IT and ICM and ICV. And we don't really see that much difference. So by default, we go to IT because it's simpler, it's safer and there's more clinical precedent. But we absolutely would, if we needed to do an ICV or ICM route, if the disease and the therapy warranted it.

Yes. And Luca, I think Suyash brings up a really good point. I think as you guys are having discussions with other companies developing gene therapies for neurological diseases, I think the key question that kind of asked people is when are they dosing their translational studies. Because most people, they want to give you a best case scenario. This is something that we don't do. We typically dose our animal models at 4 weeks of age, 5 weeks of age or, at least, when the disease has had some accumulation in the animal model and the animal is actually symptomatic. Because we want to be able to make something that's a little bit more translational to what you would see in the clinical setting. And unless you have newborn screening, which most of these diseases don't, you're not going to pick up a patient until they already are exhibiting symptoms and may have some high degree of symptoms that they're exhibiting. And so what's happening is because a lot of companies like to do their studies, with best-case scenario, they dose at P0, right? They dose a newborn mouse or newborn animal model, trying to demonstrate ultimate rescue of that model. But it's not necessarily translatable because, again, unless you have newborn screening in the clinic, you're not going to pick up a patient at birth. And you're not able to do intrathecal dosing of a newborn mouse. The mouse is too small. So the only way that you're able to do CSF delivery is with ICV. So typically, what happens is they'll do their preclinical studies at P0, do an ICV and they'll carry that through. And so that's why I think we're seeing a little bit more of direct dosing with its ICV or ICM because that's a little bit more translatable to what they're dosing -- it's a little bit more translatable to how they're dosing their translational studies at P0 and P1.

Super, super helpful. I know we're already way out of time. We did start a few minutes late, unfortunately, had some technical difficulties. So maybe I'll just pose the last question. And RA, there'll be a 2-part question. So the first part of the question is, when you think about your pipeline, forget about GAN and GM2, which I think get a lot of visibility already. When you think about the earlier pipeline like Rett, Lafora, [indiscernible], what do you think is the one asset or indication that will play a more central role in Taysha's story going forward? That's question one. And then second question, obviously impressive track record of business development here. How you're thinking about business development going forward?

No, really, really good question. So to answer your first, and so you definitely need the time in here. For me, it's Rett. And I think for a number of reasons. One, it validates our miRARE platform, the self-regulatory feedback loop that essentially caps expression at wild-type levels and guards against overexpression-associated toxicity. That preclinical data set was just published in Brain last week, and we've gotten phenomenal feedback on that. Not only is it a huge patient population, but it's a patient population that is now been demonstrated to have a high level of reversibility. And so just by kind of this narrow, what we would say, Goldilocks kind of expression pattern has really been the thing that have hung up gene therapy development with Rett. And now we've kind of demonstrated quantitatively that we can control expression in a genotypic manner, essentially meaning, in wild-type cells, because these Rett patients are mosaic. Essentially, half of their cells are normal, half of their cells are Knoll or disease. But in the normal wild-type cells, we have minimal MECP2 expression. In the disease cells, we get back to normal levels of expression and cap it there. And now we've been able to demonstrate that quantitatively. And the data is in the paper. I suggest -- we're happy to send it to you, but for the people watching, we suggest they go and download that. It's a phenomenal platform that can now be applied to a number of diseases where you have dose-sensitive genes. Think Hopkins, FOXG1. Angelman is another one where you could potentially apply this approach to. Not only did it to validate that construct and provide the basis for filing the IND, but it also, again, provided a number of significant platform value that will have ramifications in other disease sets. Suyash, maybe you want to chime in here. Maybe I covered it, or I don't know.

Well, I would just emphasize, I agree. I think Rett is going to be the one saying that -- I'm certainly most personally excited about. I've looked after a number of children with Rett, it's a really devastating complication.And for decades, doctors have been trying to find ways of treating it. And with the advent of gene therapy, people thought there was a shot now. And then suddenly, it was realized, "Okay, you've got to somehow limit MECP2 expression on a cell-by-cell basis." And the people are struggling with that for a long time. And Steve Gray has -- he's been working on this problem for 15 years. Yes, he's actually vocally said to myself, "RA, if I knew how hard it was going to be, I would have never started." Lucky, he did. Luckily, he was useful and naive, but he brought it forward to the point where we now have a very, very nice way of moderating MECP2 expression, controlling on a cell-by-cell basis. And I think this speaks -- as R.A. mentioned, this speak to a wider portfolio and really where the brilliance of Steven Gray comes in. Because we stick with AAV9 HK293 immuno-cell suspension and ITAA because we know that all our work, and we now have to do it very, very well. The real innovation and the brilliance comes in, in the payload and [ one in that ]. We do standard gene replacement. We do it many gene replacement. We do this miRARE platform where we incorporate the miRARE platform into the untranslated region of the construct. We bring in short half in RNA to help sinus genes. So this is really, I think, where the real innovation comes in a Taysha. We control what we can control, and do well what we in the field can do well and then bring in Steve Gray's phenomenal creativity and innovation in that payload design. And we've proven it preclinically now with the issuers, let's move it into clinic and let's make it work in human.

And looking to your last question, I'll answer it quickly, knowing we're coming out of time. The way that we're thinking about BD is we're going to continue to be opportunistic on the buy side. I come from a BD background. I've been fortunate to have some success in putting together portfolios, and we're super excited about the portfolio we put here -- we've put together here. And we're kind of at the point to where we feel pretty good about where we are. So we're not necessarily going to look for deals. But if a deal finds us and it's a good one, we'll pursue it. On the sell side, I couldn't see -- it's -- you almost think about it from a dilution perspective. And I want to make sure that we're not diluting value within the company itself. And so I think with the portfolio as it stands today, we're pretty capable of moving these programs forward in the economies of scale that we're able to get. Because they're all, as Suyash mentioned, HEK293, they're all AAV9. They're all intrathecal, 90% of them are pediatric diseases. So we feel pretty good about that. One program that I think that we may decide to do some partnering on could be tauopathies, just because it's such a -- the broad potential of what that could be is just maybe bigger than what we would be able to do at Taysha, right, on our own, both on a development perspective, but also commercially when you start to think about the number of patients that are out there. So if I had to pick one, that would be one that I think we may consider thinking about some time. But right now, we're focused on moving these programs into the clinic and hitting value inflection points.

Fantastic. Maybe tauopathies, you need a big boy or some cooperation with a big voice here. So no, makes a lot of sense. R.A., really appreciate your time. Rest of the team, thanks so much for joining us. We're glad that you guys who are joining to be our RBC conference and look forward to speaking at the future time. So thanks so much for joining us.

Absolutely. Thanks, Luca. Thanks for having us.

Thanks.

Thanks so much. Talk to you again. Bye-bye. Bye now.