Good afternoon, and thanks everyone for joining our panel on AAV manufacturing. I'm Geulah Livshits, a biotech analyst at Chardan, and the gene therapy space is in an interesting time or transition point, with a number of programs receiving FDA approval over the past year, including most recently, Beqvez last week, and a growing number of late-stage programs that are in development. However, the financial climate has remained uncertain for small biotech, and companies are looking to avoid setbacks and delays, many of which have related to CMC in this space historically. So in this panel, we'll talk about the state of AAV gene therapy manufacturing, the challenges, innovations, and the other key issues in this space that are facing the companies. So today, with us on the panel, we have from Kriya, Britt Petty, Chief Manufacturing Officer, and from REGENXBIO, Craig Malzahn, SVP Technical Operations. The format for the session is roughly 35-45 minutes panel discussion, and if any of our audience attendees have questions during the session, feel free to type them into the question box, and we'll try to ask them on your behalf. So thank you both for joining us today. To begin with, can you each briefly introduce yourselves and your companies and talk about their manufacturing capabilities? I'll start with Britt, because you're on the left in my view. Okay, sure. So hello, everyone. I'm Britt Petty. I'm the Chief Manufacturing Officer at Kriya Therapeutics. I've been with the company, just over four years, so from the very beginning, when it was just an idea. Previously, I was, I was at AveXis, and prior to that, a long career in technical operations in, in big biotech. You know, at Kriya, our manufacturing capabilities are really, built upon the macro strategy of the company. You know, Kriya is, is a gene therapy company attacking, prevalent diseases in larger patient populations. So, in building out our internal capabilities, we really put proportionally more emphasis on, on CMC capabilities, from the very beginning. We have a 51,000 sq ft GMP facility here in Research Triangle Park in North Carolina. It's fully integrated capabilities from early research all the way through manufacturing into drug product filling. We have bioreactor scale from 50-liter all the way to 3,000-liter, and our platform process utilizes. It's a scalable process and utilizes downstream chromatography techniques for purification. Great. Cool. Craig? Yeah. Craig Malzahn here, SVP of Tech Ops, so that's manufacturing, sort of process development, supply chain, engineering. I've also been with REGENXBIO for about four years now. Just started just prior to COVID. Prior to that, I led one of GSK's large-scale biopharm manufacturing sites. So I spent a lot of time in mAbs and, prior to that, about 10 years in vaccine manufacturing. A little bit on REGENXBIO, it's a company that's founded in 2009 out of IP from UPenn. Really the first five, six years of the company, small team, focused on sub-licensing that IP. In 2015, we started our own drug development, and that's when our CMC capability build started as well. You know, for about seven years, we primarily outsourced manufacturing. Today, we've got two programs that are in late-stage pivotal, you know, headed towards- hopefully headed towards BLAs in the next couple of years. A third program where we're hoping to move to pivotal fairly shortly. So pretty mature pipeline, a number of research programs as well. We've now, during COVID, we constructed our GMP facility, our in-house manufacturing facility. We've got two drug substance production trains. We also have a fill finish area, and the facility is designed to meet clinical needs as well as meet global commercial requirements. From a scale standpoint, we operate in our GMP space, anywhere from 500- to 2,000-liter scale bioreactors. Great. So with those overviews, maybe to frame the discussion, I'd like to understand how each of you would characterize the state of gene therapy manufacturing today. You know, again, as I said in my little intro before, there's been setbacks in the past. There's a lot of programs that are in late-stage development now. There's a number of programs in early stage that have been entering the clinic. Where are we at in terms of figuring out the process? Have we kind of ironed out the kinks here, or is there continued challenges that we're facing? And maybe we can start with Craig and then go to Britt. Sure. You know, I think the industry has come a long, long way in the last 10 years, certainly around AAV manufacturing. You know, where we all started with very academic lab-scale processes, highly manual, low yielding, you know, really hard, difficult to scale up. I would say a lot of companies have now transitioned to suspension-based manufacturing. I would say mostly on HEK platforms, but of course, there are some VE, SF9 platforms, a few others out there as well. So the production systems are coming along quite nicely, but we're still relatively immature, so there's a lot to learn, a lot of experiences to gain. We tend not to make, in this industry, a ton of batches, so that somewhat limits us, just given the fact that it's, you know, single dose type of strategy per patient. You don't often need to make a lot of batches. And I think the analytical methods have also come along by it. It's really important that along with the manufacturing processes, the analytical methods are coming along just as well. Still, a good bit there that needs to be optimized yet and explored. Mm-hmm. Britt? Yeah, I mean, I think Craig hit it spot on. You know, I step back and I see the industry really in a pursuit of optimization. You know, how pure is the thing that we made and the relationship between purity and yield? And, you know, I think the industry is having to come to grips to some extent with standard definitions of what we call these things. And, you know, needing to get to the point where we come to grips with where we can talk about these things in terms of standard units of capacity, that you can then fit to particular product. You know, being able to talk about productivity and yield in a common way across the industry, I think is something everybody's working on. We need to really harmonize our view of that. You know, have an ontology for capacity and productivity, you know, or some agreed-on metrics that sort of harmonize a view across the industry. And I think, you know, to me, that pursuit of optimization really starts with, you know, it's got to tie back to the macro strategy at each company, and, you know, what are the weaknesses in the existing or the early processes that you want to attack? You know, is it in terms of productivity? Is it volume? You know, is it purity, potency, or related substances? You know, what is it exactly? And then do you stay within the envelope of existing science, or do you try to reach out to a new technology? You know, I think that's something that the industry is sort of wrestling with right now and working through. You know, if I think about it in terms of timelines in other modalities, some of the big step changes, you know, take 10, 15 years to really play out. And I think for this modality, you know, using sort of current technologies, I think the first thing that has to happen is to continue to demonstrate the economic value of what we're doing now. You know, I think it's a long cycle problem, and that people are working on, and there's things out there to make it a lot better, but you got to balance that with you know, the need to get to the clinic and move things forward and create value for your company. So I think the whole industry, you know, I look at it, it's sort of in this pursuit of optimization right now, and I think it's a little different as companies, you know, how it ties back to their macro strategy. Well, that's interesting. I guess, maybe I'll just follow up on a couple of things that both of you mentioned. So, you know, in terms of, kind of, how should I put this? Leveraging the learnings that have been made in other industries, you both have spent time elsewhere in biotech. How can you incorporate what you've learned, and how can the field incorporate what you've learned in these other spaces into gene therapy manufacturing to kind of avoid some of the challenges? Not necessarily with respect to specific manufacturing protocols, because obviously everything is going to have a specific process requirement, but more generally, what learnings do you take from your time elsewhere in biotech? Yeah, maybe I'll, I'll start on this one too. Mm-hmm. I think it's really important because I would say most of the experience that we've certainly hired in here at REGENXBIO is coming out of mostly Mab manufacturing expertise. And that's, that's largely by design. I mean, I think first thing I'd think of is in Mabs, they're really good at scaling processes. You know, we know how to grow cells, we know how to scale that up. A lot of Mab processes are running at 20,000-liter scale on a scale that, you know, Britt and I are talking about is 2,000 or 3,000-liter scale. So, learnings can be pulled across, you know, these industries from a scalability standpoint. The unit operations, most of them are pretty similar between, you know, biopharm and gene therapy manufacturing. So, many of the failure modes with your equipment, with your procedures, with your personnel, you can learn and draw on prior experiences making other, you know, biologics and apply them to what we're doing here. The last thing I would mention is, and we've gone through now a couple of rounds and a couple programs of process characterization. Very similar approach for process characterization, getting ready for process validation. It's, you know, taking a lot of the concepts, a lot of the strategies, a lot of the risk assessment type of approaches that you would take in the MAFB world and applying it to gene therapy. So I think, I think there's a ton there that's that we learned from, from prior experiences elsewhere. Yeah, that's really good. I think for us, because we're going after larger patient population, prevalent diseases, it's very similar. We see manufacturing at scale as a critical success factor for us. So that's what we view every product that we have in our pipeline, is we have to be able to make it at scale so that it's commercially viable. There were really... I'm trying to think. There were really, like, three learnings that I had from a previous experience in gene therapy with one of the early approved products, and it was, you know, you need to consider very early on what we call manufacturability. 'Cause if you don't look at how your transgene is going to package and work, you know, you can find yourself trapped in a process that you can only scale up and you can't scale out. It just doesn't work. And I think Craig hit on the other one, which is you need to get a solid analytics package to really understand and measure, you know, these, these impurities and, and productivity early, so that you can understand, identify process capability issues early enough to do something about them, and make a repeatable process. And then the final thing would be, you know, the scalability aspect, that you know, if you want to be able to have a commercially viable process, you need to be able to make larger scale batches and to drive down the cost of goods. And so those were kind of the three learnings that I've had previously, and then how we think about it. Kriya? Mm-hmm. So in this space, does it seem like the AAV gene therapy space is kind of converging on, you know, a single or particular process? You guys mentioned HEK 293s or some folks that are using SF9. Does it seem like there's convergence, or even within that, is there still a lot of nitty-gritty detail that can differentiate processes there? You wanna take that one? Whichever you- Yeah. I mean, it does seem to me, you know, my competitive intelligence by no means is perfect. I think, I think this segment of the industry is still in its early days, and it's not an especially collaborative group. There's a lot of, this is my, my stuff, and, and it's all secret approaches kinda going on here, and I understand that. But it does seem like there's, you know, it's, it's sort of a suspension-based cell culture, you know, more of like what Craig was saying, the biologics approach, suspension-based cell culture, scalable downstream purification, and that sort of thing. And I think the challenge that, that people are solving now is, to me, the biggest challenge in gene therapy has been, you know, how do you purify at scale? And trying to, you know, how do you separate out, you know, partially encapsulated impurities? And I think that's, you know, what the industry is kind of, sort of working on right now and figuring out how to do that, so you can manufacture these things at scale. But it does seem like, you know, when you look at what people are publishing, that it's sort of scalable, suspension-based, sort of biologics platforms that are being deployed now. Yeah. I would definitely agree with everything said there. I'd probably just add to it, and we're with you as well. You know, being able to separate out empty and full partials, and that technology is just super critical to get that right. The other one, as we scale up, and I think many of us have run into challenges with, if you're doing triple transfection processes, that type of thing, scaling that to, you know, 500- 2,000-liter scale. I think we're learning that, we're figuring that out. I think we've had some good success, but there's still a lot to be solved there from an engineering, science, and certainly an optimization standpoint to get us the most out of these cells as we can from a capsid, a full capsid standpoint. So companies often talk about manufacturing platform, having a manufacturing platform. So in practice, what does that mean from your perspective? How much is, you know, it gets standardized, how much optimization needs to be done for each new product, and what does that mean from a regulatory standpoint? We can start with Craig. Okay. Yeah, I mean, we love talking about our platform. The first thing I will say is our whole platform has evolved tremendously over the last, you know, five, six, seven years. And effectively, we've now landed on what I would call a standard platform. We refer to this as NAV Express as our standard suspension platform, which we've, you know, we've demonstrated it's scalable, it's portable between manufacturing sites. And it can be applied, the basics of it, the framework of it can be applied, in our experience, between different serotypes, different transgenes. But it's, you know, it's not 100% plug-and-play. There's always gonna be some optimization work there, some tweaking that needs to happen. Hopefully, you know, it's but for the most part, we've seen it's all the same unit operations, but maybe some different conditions to be able to handle differences in capsid serotypes and that type of thing. I think it avoid, you know, it really affords a ton of benefits. We've seen it now, especially. I would say, in later stage development, where we can do process characterization off of one program and apply, you know, a large percentage of that to the next program. And you don't have to repeat all those studies again, and it also helps inform what do you want to evaluate on that next program. So there's a lot of process and product understanding that can be learned across manufacturing platforms. And Britt mentioned analytics. I think the analytics and the analytical methods go hand in hand with the process, right? So you don't have to redevelop all your assays either. Maybe you get to focus on a couple, like potency, and a couple of others, but you don't have to reinvent 20 different methods, as well. So that's a big advantage from a platform standpoint. Yeah, yeah, and I'd say that's spot on. I think, you know, for us, the whole thing about our manufacturing platform is that it's adaptive to an array of potential products. So, you know, we have a standard set of operations that we deploy, but we optimize for each individual product, you know. So, and we're still trying to do continuous improvement on it for different products. But we're not having to go back and replace, you know, our TFF with some other unit operation. You know, so it's really the platform for us is all about being able to quickly repurpose our unit operations for the array of products that we have at Kriya. I think the other thing I would add to this is, so far, at least our experience has been, is that regulators seem pretty open to, you know, cross-referencing programs and using data from one program to the next. Obviously, if there's a strong scientific rationale there, we'd always encourage the agency to, to you know, work with us on those types of approaches, especially in the rare disease space, where, you know, you have to be, you know, we've got to be very efficient with the drug development costs. We can't be always repeating studies and, you know, work on every single program. We've got to get more and more efficient as we move from one program to the next. Being able to, you know, share those datasets between programs is really critical to the success of rare disease. That makes a lot of sense. So it seems like the agencies seem to be pretty open to cross-referencing, as you've demonstrated, and being able to make product in one program. It makes sense. I imagine there's also supply chain overlap and things like that if the unit ops are similar. Yeah. Yeah, absolutely. I mean, having essentially a standard raw material kit or component kit, or 90% of your materials can apply from one program to the next, it's lower inventory costs, it's fewer vendors you have to manage. You can definitely, you know, reduce the risks, collectively across the programs. Yeah, I guess I would just, you know, add one other thing. I think for us, one of the learnings we've had is to make the platform concept work, you know, there's this teamwork between research and product development and early manufacturing that has to be, you know, connected and synergized. I mean, they... research has to know why we're asking them to make things a certain way, you know, so that, you know, we're not flaying, flailing away, just looking, you know, to make it, you know, a custom process for every new product that they know what the strike zone is that they're pitching to. And, I think that understanding and collaboration early with our colleagues in research is really critical, has been, for us, really critical to make the platform concept work. Otherwise, you're just catching whatever they throw over the wall to you, and you can end up with all these bespoke processes, and every time trying to, you know, customize and figure out how you do all this stuff. And that is a really important aspect of our workflow at Kriya. Yeah, it is, it is a great point. We actually, one of the things, the way we've organized here at REGENXBIO is the team that we call them Vector Core, that manufactures for research at the bench scale. Pretty much whatever research can dream up, they're doing the bench-scale production. Most of that team, a lot of that team is out of process development, and they've moved to that, and, they actually report into, into my organization. So they report into the CMC side of things. So we're always emphasizing, as Britt said, manufacturability, let's get an early read on this, especially as you start narrowing down candidates, you know, starting to come towards candidate nomination. We wanna see, you know, what, what challenges might be there when it, when it does move over to process development. How long is it gonna take to, you know, tweak the process, put it on the platform? You know, what are the known risks? Sometimes that can inform even, you know, what candidate we move forward, you know, in the pipeline. ... Right, right. So I want to move on to something that we touched on a little bit in the early part of the session, and that is the technologies that can help meet demand for some of the larger indications that gene therapy is starting to be applied for. Maybe you can talk about things that you're excited about that might improve yield or ways to reduce COGS, which again is top of mind for many in this space. And so, can start with- Did you say, did you start with me? Yes. Yes. Okay. Yeah, I think, I think there were, you know, a couple levers that we went after, going after these, these prevalent large patient population diseases. Early on, it was a must-have that whatever we do has to be scalable, right? So that was one lever that really drives, productivity and drives down cost of goods, is being able to scale this to large-scale bioreactors. You know, the second piece was, really the downstream purification. You know, how do we get those, those yields and, and purification from, you know, where they were a few years ago in the 20s, maybe 30%, and trying to get those up to, you know, 40 or above 40% downstream purification. And then I think the third thing is really on the packaging efficiency and really understanding that early in research where you're building in, you know, kind of the old concept of quality by design, you know, in biologics, that you start with. You know, I think it's just a really simple assertion, right? That you, the best way not to have this challenge in trying to purify out these partially encapsulated impurities is to not make them in the first place. And I think we focused on really those three levers that are driving us to, you know, we believe, you know, commercially viable therapies for large patient indications. Yeah, I would say, certainly, certainly similar, probably again, you know, like, Britt mentioned, the first one definitely getting to a suspension-based scalable platform with stirred-tank bioreactors. I would say our next target, and where we've seen some significant productivity gains is really around, better clonal selection, and, and, you know, engineered, cell lines so that we're, you know, we're making more capsid per cell. Getting to those higher-producing clones has been a, had a really big impact on, on what we're able to produce, and then coupled with, optimizing our plasmids. So like I mentioned before, we run a triple transfection plasmid system. We've done a lot of work around evaluating optimization of those plasmids, the right ratios and transfection, but also in the plasmid design. We've seen large, large productivity gains come from that as well. Got it. Makes sense. So you've mentioned a little bit regarding regulators' perception of platforms and allowing you to potentially cross-reference different products as you move down your pipeline. But in general, obviously, CMC has been a source of regulatory setbacks in the cell and gene therapy space for quite a while. Your companies are in different stages of development, but from each of your perspectives, how clear are the regulatory expectations regarding CMC? Let's start with Craig. They could always be clear, right? I think we'll never fully accept where they're at. You know, I think acknowledging... I, you know, haven't been in the field really just for four years. I've seen a lot of progress over four years of clarifying expectations. I think in particular, Peter Marks has been a fantastic advocate for the industry, kind of pushing the envelope on different approaches, recognizing, you know, some of the challenges we face, again, in particular in rare disease, but in gene therapy in general. Yeah, you know, we certainly give the agency feedback on trying to help further sort of clarify expectations. I think one of the examples of that is, I think from our perspective, maybe far too often in the CMC space, cell and gene therapy gets intermingled in some of the regulations. I think we'd like... I you know think gene therapy manufacturing is actually probably more similar to mAb manufacturing than cell therapy. So I think some clarity there around what expectations apply for gene therapy and what are more specific to cell therapy would be really, really helpful. Being in Rockville, Maryland, actually, it's kind of a nice thing. We've actually been able to host a couple times now, FDA educational sessions to bring them over to the manufacturing facility, kind of describe our challenges. Nothing specific to a product or a program, but just in general, take them through the facility. They get to experience sort of the GMP environment, having a gown up, go through that. I think it's, it's been a good way for us to kind of build a, build a relationship there, and maybe they get a different perspective on, you know, what it takes to make a GMP, product... So yeah, I mean, there's, there's still a lot of, I would say, trial and error, trying different approaches and seeing what kind of feedback you get. But I think we've made a lot of progress over the last, you know, five years. Yeah. Yeah, you know, this segment of the industry is certainly not completely immature by any means. So it would stand to reason that the regulatory authorities' understanding of this set of therapies is evolving, and the regulatory trajectory is sort of increasing, and you have to be thinking out in front of that. You know, so I think what has to happen here is, you know, we have to learn together. You know, we should be looking for opportunities for dialogue about, you know, "Here is what we've learned around this product, here's what we think it needs, you know, what questions do you have?" As opposed to, you know, a contest about who the biggest expert is. You know, it has to be a partnership, and, you know, we want these therapies to succeed, and they do too, because patients are waiting for them. So I think that partnership aspect is gonna be key to getting there. You know, certainly, you know, the analytical control strategy and the methods you're using to characterize, you know, the quantity, purity, and potency of both the capsid and the vector genome is critical and something that they're obviously looking hard at. So, that's sort of how I see that. Makes sense. Maybe to follow up on something that you mentioned earlier, Britt, you were talking about the challenges in this space, and I think you mentioned the downstream purification as one. Is that the main challenge that you see, or are there other kind of major challenges in the space that is facing the companies today, or are there other things that you'd point to? Yeah, I mean, from a CMC perspective, I think it's... I think that one is one of the biggest challenges to overcome, is how do you get the purity with scalable techniques, is primarily the biggest challenge that I can think of off the top of my head. Yeah, I don't know, Craig, if you have any other thoughts. Yeah. You know, we talked a lot about the platform, but, also, you know, it's taken us a long time to sort of evolve to a standard platform. So I would say one of the biggest challenges that we've experienced is just navigating process change, comparability. Yeah. You know, we all would love to only have to deal with analytical comparability and never have to, you know, go and do clinical bridging studies, things like that. Sometimes that's needed. And it hits on what Britt was saying around the analytical methods. I mean, the stronger your analytical methods, and you know, having orthogonal ways of looking at different- Yeah ... aspects of the product profile, building that data set, I think is critical to navigating through process changes. Especially when, you know, if you're lucky enough to have a program that's going through an accelerated pathway, that sounds fantastic, but at the same time, that means that everything CMC is gonna be accelerated, and you might not have the most time to, you know, develop the perfect process, or the best process you can have for commercialization. So, yeah, in summary, I think it's comparability is just super critical for- Mm-hmm ... for these programs. Well, coming back to something that I think both of you just mentioned, you know, how do you approach balancing that, you know, leveraging the platform that you have versus, you know, incorporating new things to make it better? Yeah, I'll jump on that one. Sure. I think it's a program-specific type of decision. You know, it sounds like Britt's team there, they're mostly focused on large market indications. We have a mix. We have some large market indications, we have some rare disease indications, and we also have different therapeutic areas, some in ocular, some in central nervous system, some in muscle, which really dictates how much vector you need to dose. So if you're dosing into the eye, it's not much vector. So in that case, you know, driving productivity up might not be a huge priority because you can make a lot of doses in a 500-liter bioreactor. Whereas if you have a muscle indication, it's a large dose per patient, and it might be super critical that you're focusing on driving productivity up, you know, improving your upstream titers, you know, improving your purification yields just to get to a commercially viable cost of goods for a program like that. So I think those are the types of things we think about, you know, where do we balance going after a new innovation versus what we have today is good enough for a certain program. Yeah, and I think, you know, it was a really good thought on the comparability, and what we try to do to mitigate that risk is get to scale early and run our early non-clinical studies as close to commercial scale as we can, depending on the program timelines and things. So we're trying to get there quickly in our platform. As I was saying earlier, I think the early work we do with research, we're finding that if they can pitch to the strike zone of our platform, we're having a lot of success scaling up and scaling up quickly, which allows us to get to scale for clinical material, for tests in humans very quickly to try to avoid trying to throw in, you know, process changes later that can jam you up. Mm-hmm. That makes sense. So, Craig, I think earlier you mentioned, you know, what it might take, what it might take to get to commercially viable COGS. How are you approaching what that means? That's a question for both of you. Either one of you can jump in. So what are reasonable COGS aspirations here? And obviously, I assume that's going to be program-specific as well. Yeah, yeah, it is. I mean, we look at that very early on. You know, we look at it, you know, even in the candidate nomination phase, just to say, "Okay, are we in the right ballpark, or is there going to need to be a significant investment in the manufacturing process such that we have a cost of goods that is commercially viable?" So I think making sure that we go in with our eyes wide open very, very early on, and that can help dictate, you know, how much time we're going to need to move this through, how much time we need with the process development team to get there. I'll say at this point, you know, our platform, you know, we've taken it to the point where we feel like there shouldn't be any constraints, from what we see. We think we're in a good place, and there's only opportunities to, you know, to further improve it. I can say four or five years ago, that wasn't the case. That we did have some manufacturing processes that were going to be very difficult to bring to the commercial market. But today, I think we feel great about even our late-stage programs, where they are, and certainly the new ones that are coming through- Mm-hmm ... through the pipeline. Great. Yeah, that's, you know, the same here. I think we look at COGS very early, and to understand, is this going to be even commercially viable or not? Can you make it especially for... You know, we're one of our portfolio products is type 1 diabetes, and, you know, there's already a standard of care out there. So we have to be able to get these at price points that are that these things can be commercially viable. So, you know, I think for what we're doing in our models, that we're, you know, we're very well positioned to be best in class in some of these COGS in the industry for gene therapy. You know, and I, you know, to throw out a number, you know, we think about trying to be in the $10,000 a dose, and there's a ton of variables that go into these models, all the way from what the dose is to, you know, the scale, all those type of things. So, but that's how we look really hard at that. Thanks. And so it does sound like the field has evolved quite a bit, even over the last four or five years. So looking five years into the future, what do you think it's going to look like? Britt, we can start with you. Yeah, I think this thing, you know, the industry is going to continue to evolve and get better and better. And I think the collaboration across companies and with the regulatory agencies is going to continue to get better, and there's going to be sharing and learnings across the industry, and, to me, it's gonna follow the biologics model that Craig mentioned earlier. I think we're going to continue to see productivity and yields and purity and these things continue to get better. And I think we're going to see a lot of gene therapies get approved in the next, you know, five to 10 years and, you know, I really see it following that biologics modality model. Yeah, I would just add to that. My dream here is to get CMC off of critical path and stay that way, and I think we can be there in five years for sure. Doesn't sound like much, but it's really important that, you know, we're able to be reliable with our manufacturing processes, even from a very early stage. You know, as Britt said earlier, you know, if you could start your phase 1 studies with close to your commercial scale or at your commercial scale with your commercial process, it's a huge de-risker. You really de-risk the development pathway. I don't see any reason why, you know, we can't be there within the next five years across the board. Makes sense. So we have a couple of minutes left, and I did want to give each of you an opportunity to ask each other a question. We can start with Craig, and then go to Britt. Oh, okay. All right, here's my question for you, Britt. If you had $10 million to invest anywhere in the manufacturing process, where are you putting your money? You know, I, you know, I think, I think I'm putting it in, you know, back in the molecular biology and the vector design. I think, I think really working there to, you know, understand what are the correlating factors are that, that really drive, not just the protein expression of your therapeutic molecule, but, you know, understanding in the, in the transgene design, what are the, what are the levers you can pull to, to really improve the manufacturability of it. I think, you know, instead of just trying to purify out all these impurities and come up with bespoke processes and technologies, if you can really understand what levers to pull and really correlate those back to different aspects in the molecular biology, I think it could really be a, you know, a game changer for the modality. Good question. Brent, your turn. Yeah, you know, I think, you know, just thinking about what Craig was talking about earlier and REGENXBIO's transition from, I guess, utilizing, you know, third-party laboratories and manufacturing capabilities to the investment in CMC capabilities in-house. How did you guys, you know, come to that conclusion and sort of arrive at, you know, you needed to make that transition? Yeah, that's a great question. Few different factors for us. I would say certainly getting to a point where we had a reasonably standard platform, you know, stirred-tank bioreactors prior to that. You know, we did adherent hyperstack manufacturing, that type of thing. But once we recognized, okay, we have cells, we can—we've adopted them to the suspension platform, we can do this. It's a little bit easier to design a facility now around a platform than, you know, trying to figure out all the different things that you might have to do in the future. So that was a big one for us. Another big one was feeling like we had enough of a pipeline as a company, that we weren't reliant on the success of one program to build a facility, right? So it's, "Hey, we got multiple shots on goal to get commercial products. They're gonna be on a similar platform," you know? "Okay, let's start doing this ourselves." You know, if you're only going after one- Yeah. - and then that one doesn't work out, you got an empty plant. What do you do with it? It's a lot of operating costs, it's a lot of capital you've spent. So that was a big factor. And then I'd say the last one was, I don't know, maybe we're just a bunch of control freaks, but you know, we wanted to control our own manufacturing destiny. You know, working with CMOs, I'm sure you're familiar, right? It's sometimes great, and other times not so much. So having direct access to a production schedule that we control, to the quality of product that we're putting out, as well as the cost, you know, it puts- Yeah. It puts manufacturing in our own hands. You know, that gives us a sense of control. Makes a lot of sense. Great. With that, I think we're just about at the end of the session. I'd like to thank you both for the fantastic discussion.