Thank you for being here. My name is Sam Semenko. I'm one of the biotech analysts here at Citi, and today it's my pleasure to be hosting Steve Harr for a fireside chat. Steve, thank you so much for being here. Thank you for having me, Sam, and thanks to Citi for having us. I will always say the one thing that just in case people are listening more, you know, online and things, we will make forward-looking statements. We have a whole bunch of risk factors in our most recent 10-Q. We spent a lot of time writing those, and take a look at them. There is always good information in those, so please read them. Great. Thank you, Steve. Why don't we just kick it off at the top and tell us about Sana and where you are as a company today? Sure. We're now about six and a half years old. We're not that old of a company, but we're old enough to have done a few things, made a few mistakes, and made some real progress. We started the company with the idea that over the next several decades, one of the most important advances in medicines would be the ability to modify cells and use them as medicines. That's not a popular thing right now, but I think there's nothing that we've seen that shows that to be anything but likely to be true. With that, we wanted to tackle two scientific challenges. One was, if you think about replacing a cell or transplanting a cell, you want to be able to make a cell at scale that will engraft, function, and persist. The biggest challenge in that equation has been persistence, and in particular, overcoming allogeneic rejection. People have gone about that in two different ways. One is by severe, or significant immunosuppression, like an organ transplant. The other has been with autologous cells, neither of which is very scalable or addressable for the broad population. That's the first thing we went after. The second is we want to be able to deliver payloads in vivo. The goal was to be able to deliver any payload to any cell in a specific and repeatable way. The technology we started with allowed us to do any payload, DNA, RNA, protein, and be very specific in the cells we went after. We've actually made really good progress in both of them. I think we have best-in-class platforms in both of those. In particular, that's left us with three real areas of products or platforms, let's call them. One is a drug called SC-451, which is a gene-modified stem cell-derived pancreatic islet. It is a potentially curative, one-time curative therapy for people who are living with type 1 diabetes. It's a massive unmet need. There are 9 million people in the world that have this. It's growing. If you have this disease, if you have the best possible care right now, you probably live about a decade less than someone without diabetes. Most people are probably more like 20 years less. During that time, you have risk for amputations, blindness, heart attacks, stroke, kidney failure, and the other problem, not too high blood sugar, but too low, which is coma and/or death. It's a disease where there hasn't really been a meaningful change in the standard of care in 100 years, and where I think we have a real opportunity to make a very important medicine. The second area is, and we're going to talk a lot about that, I'm guessing, because that's probably the area most investor interest. The second area is our in vivo delivery capability. In particular, right now, we're focused on in vivo CAR T cells. I think we've shown really best-in-class data in the non-human primate. That's not in people. It's an area where strategic activity has picked up a lot lately, in particular for programs that have shown at least one or two patients' worth of human data. We are ready to move towards human data. We have to figure out how we come up with the money to do that. I think that's something that we're really working towards. It's a program we're very excited about. The third is allogeneic CAR T cells. We have a couple of different drugs in development for allogeneic CAR T cells. One of the things we've said in the past is what we've shown is, first of all, with allogeneic cells, you want to avoid immune rejection. You put someone else's cells into you, you will kill them. You'll recognize the forum to kill them. We've shown we can avoid immune detection. Actually, we just had a paper published with data in that last week. It was one of the Cell journals, again, showing that we avoid immune detection. We're quite confident that we have a drug that works here. We've shown you that a CD19 CAR T cell that we make will lead to deep B-cell depletion in patients with cancer, and that's the goal in the autoimmune setting. We have to define if this drug is OK, good, or great. That's part one. Part two is we have to come up with capital to move it forward. I think it's pretty clear right now. I think there are only two standalone CAR T companies in the world that have a positive enterprise value. One of them has a couple billion dollars of sales, and the other one has a drug that may be thought of as best in class. What we have to figure out is can we come up with enough data that investors are going to want to continue to invest in this, or is this a program we have to either partner or move on from? We will not starve our type 1 diabetes asset. It is the most transformative asset in the company. That gives you just a sense of these three areas. Probably the fourth area that everybody likes to talk about, which is capital, is a hugely important part of our future. That's where we are and what we're up to. Thank you for that overview. You're correct. I do have a bunch of questions on type 1 diabetes. OK. Let's start there. You recently had the FDA Interact meeting with some feedback that you framed as positive. I guess what gives you confidence out of that meeting that you can take your lead GMP cell line forward for the type 1 diabetes program? Yeah, I'm going to take two steps back and walk us through that. I'm going to answer your question because I think the reason we're confident we can do this is because we have alignment on what we need to do. That's the simplest answer. We can do it. We've done it. As you take a step back, type 1 diabetes is actually a relatively simple disease to understand. It's a missing pancreatic beta cell. The patient's immune system has killed the beta cell. I'm going to talk about islets. A pancreatic islet is simply just pancreatic alpha, beta, and delta cells. You could also think about it as pancreatic beta cells in a support structure, right? About 25 years ago, James Shapiro and others began to show that you can transplant cadaverically derived pancreatic islets and that would actually be curative for a period for a patient with type 1 diabetes. Some people are out now 15, 20 years. Cadaveric islets aren't a very scalable or replicable supply source. These patients, the people who get the cells, have to be on lifelong immunosuppression. It has been important for some people, but it hasn't transformed the field. It just turns out there aren't that many people who are on lifelong immunosuppression; better on lifelong insulin. Over the last few years, several different groups have shown that you can take pluripotent stem cells and make them into islets. That may be a more scalable source. It's certainly more replicable. You still have this challenge, which is the immunosuppression. Very recently, the New England Journal of Medicine just published this study for us. We showed that we could get rid of this immunosuppression. Now, really, a cure is kind of inevitable, right? I hope we're the ones to do it. We seem to be the most advanced in this field. Someone's going to do this. There are really kind of a handful of scientific challenges at turning that vision into a reality. They relate at this point, because we've already shown you can overcome allogeneic and autoimmune rejection, to a couple of things related to manufacturing. The most important and the most challenging is what you're getting at, making a gene-modified pluripotent stem cell master cell bank. What that really means is you're making a single cell where you made all the gene modifications. From that single cell forever, you are going to grow out all of your product. It's going to divide and divide and divide. Then you're going to differentiate it from stem cells into a pancreatic islet. The challenge that we faced over many years is that we would see, as we went through many, many divisions, the emergence of some types of mutations. I always give the example of something like a P53 mutation. I think most people would recognize that's a pretty bad mutation to have. That leads to a high risk of cancer. We don't want to transplant those types of cells into a person. It's taken us a long time to develop a gene-modified stem cell bank where we didn't see emergence of mutations. We've done it. We then need to align with the FDA around all the release criteria so that this could be our source for good, right? I think, you know, we have alignment with them. We can do what we need to do. We've done what we need to do. It's a relatively straightforward path. Now we need to do two things to get to an IND, which I'm sure you're going to ask me about. The master cell bank is something that you should feel very good we have in our hands. How long have you had that master cell bank? How long have you looked for mutations as those cells divide? We've been looking, so this single cell we've had for several years, it's been through many, many, many divisions. It's actually made it to where we made islets from it using more or less our current process and where we transplanted them into mice. They continue to function for 15 months with no evidence—we'll talk about this a little bit—of any tumors or histologic abnormalities. We're quite, this has been really, really well studied. It's been through many, many, many divisions. You know, enough to make enough product for decades. This is something that I think is a pretty special cell. Like I always, people ask how you get it. I think it's a lot about the quality of the cell you start with. It's somewhat about, it's a lot about the process you use to make it. It's somewhat about luck, right? Even in the run we made, we got, I don't know, you get like 30 cells that have all the right edits in them and all the right locations where you're very confident. Only one didn't have mutations happen, right? I think there is an element of luck to it too. OK, what? We looked at probably a thousand, hundred and some lines. This was not an overnight process or a little bit of like, oh, we tried a couple of times and it turned out one time it worked. This was a lot of hard work. This is the cell line that's been into the mice, and it's been into the NHPs as well. Never been in non-human primates. Non-human. OK, it's a different cell line that was in the non-human primates. Really, transplant lines across species. You can do it in a human immune system in a mouse, right? You can knock out the immune system in a mouse, but you couldn't do it in a non-human primate. It would It would reject a human cell. Got it. You have different oscillation patterns on our proteins. OK. What do you need to do before you can file an IND? Yeah, so there are two main bodies of activities that we still have to do. One is to finish up and do all and finish all of the non-clinical package. The non-clinical package are things like GLP toxicology studies. The main risk in a gene-modified stem cell-derived product people will be worried about is tumors. There are other things to worry about, but that's the main risk, right? Within that non-clinical package, we also have to do efficacy studies and some other things. The second thing we have to do is complete our GMP manufacturing. That really, at this point, is final process lock, tech transfer to the manufacturing site, and then actually doing the runs and then releasing the drug product. Just think of it as two tracks: preclinical toxicology studies, GMP manufacturing. And those. All of which we've done, but all of which have to be redone in the right setting in the right way. They're happening in parallel? They're in parallel, yeah. OK. When do you think you'll be in a place to give maybe more narrow guidance, then, as early as 2026? After we've done it. Does that suggest that we could find out that the IND is filed once it's filed, or would you potentially tell us? I don't know. I don't see any reason to do it beforehand. The thing is about this, maybe we would, just because it was some reason to, if there was a delay, we would. We've got buffer built into this. You do multiple engineering runs, for example, at a manufacturing site. Are you going to have a 100% success rate? Maybe. Every time you have a failed run, let's say it takes you four to get three good runs, that's another month, month and a half, maybe two months to fix it. Who knows? We need to make sure that we get, I'd rather be accurate than precise. Right now, the accuracy is next year. I think we've got some buffer built into that. There may not be. I think to be more precise in that right now is probably not correct. OK. Would you need to engage with the FDA ahead of filing for the IND, or do you have everything you need from? Talk to them some more. You'll talk to them some more? Yeah. Is there anything specific that you're able to share along the process? Is it just locking in that manufacturing process and the release assays? We did an Interact meeting. Generally, there's a pre-IND meeting. We may have more than one. We'll likely have more than one interaction with the FDA. The things that you're most, you want to make sure that they are aligned with you on exactly what you're doing in your GLP tox study. That actually cannot, if you look on the website, be the topic in your Interact meeting. It's not allowed to be there. That's part one. Part two is your final release criteria. You want to have that as late in the process as you can, as you know more and more about where you are in your analytical package. It's really not the manufacturing process. It's the analytical package that you need to align with them on. In particular, you can see this in both what we and competitors do. This is a little bit of a different drug. You manufacture, we manufacture the cells. We then, you know, kind of, you're almost done. You cryopreserve them. You can then hold them and leave them frozen for a long time. The drug is then ordered. The cells are thawed. At that point, they are reclustered into a, like, an islet. It's called a cluster or a pseudo islet. Right? They have the final release. They're sent to the manufacturing site. They need to be utilized relatively quickly after that. One of the things we need to figure out is what tests will we do at that cryopreservation step where we have, you know, all kinds of time. Right? What steps do we need to do at the time they've been, you know, kind of clustered where we have very limited time, right, to do a lot of, you know, analyses. We need to do as much as we can up front. In order to do as much as we can up front, we need to prove to ourselves and then to regulators that what we do up front directly correlates to what happens at the final release. Does that make sense? It does, yes. You'll have to do some of those. You'll have to do a bioequivalence between the two. I don't know if that's quite the right word, bioequivalence, but that's a fair way to think about it. Yeah. Understood. OK. How quickly, once you have IND clearance, could you move this into a phase one study? Very quickly. I mean, this is a very competitive, in a good way for us. I think there are a lot of sites that would like to be a part of this. There are a lot of patients that would really like to be a part of this study. Sites that take three to six months to get their process started won't be part of the early phase one study. We just won't do it. There are enough places that are willing to do contracting and other work at risk. I think it can be done very quickly. I say that within the context of I'm always surprised at how long it takes these medical centers and CROs and companies to align on the simple elements of contracting. You know, it has to go through an IRB post this. You can have an IRB that just doesn't meet for a month. It just happens to be that they can't get a quorum, and then you're delayed a month. Those types of things happen, but it should be very quick. Once it's cleared, I would think the first patient treated is a matter of days or hours. It's not going to take very long. Should we expect that first patient to be your initial disclosure, like you did with the investigator-sponsored trial? You're not sure. It's a little early. I know I'm pushing you. I think we filed an IND, which is probably material. We cleared an IND, which is almost certainly material for us. We dosed the first patient. At some point, you guys will have a little fatigue of press releases. You have to, yeah. Yeah, we'll see. OK. It sounds like the path to getting to that phase one trial is pretty well outlined in your mind and as you've described. What do you think the, or how do you envision the market opportunity once you establish proof of concept and you have registrational data? Our goal is to make this as broadly accessible to patients with type 1 diabetes as possible. That doesn't mean that all the data will be available to everybody, right? As an example, a lot of patients with type 1 diabetes, particularly those that have lived with it for decades and decades, are at very high risk for adverse cardiovascular outcomes. We're unlikely going to want to have a patient who has a high risk of a heart attack in month two, right, in our study. There will be elements that we restrict around pre-existing conditions that I think will go away over time. That just is so that we have a clearer idea of any safety signal. If it's caused by our drug or it's caused by the disease, we're going to want to kind of be narrower. We almost certainly will start in adults. Can't imagine that not to be true, and then gradually work our way back into adolescents and then younger and younger. I would presume the initial population is probably pretty similar to the initial enrollment criteria you see in the New England Journal of Medicine paper that we did for the cadaveric islets. I would also presume that over the course of even a phase one development plan, we can grow that, but it will not, even at the launch, it will not be all 9 million people with type 1 diabetes. I hope it's the vast majority. Again, things that we've done as much as we can to do that. We started by really looking at almost every O negative cell line we could find on the planet. Blood type matters, and so you can see even in other programs that if you have a blood type, you probably, I'd like to say an A blood type, you may only be able to treat 40% of the population, right? Or if you have a B blood type, it's going to be even smaller than that. By having an O negative, we can hopefully treat 100% of addressable patients. Again, I think we'll get to a very large, large percentage of the 9 million people pretty quickly. What do you need to do on the manufacturing side in order to scale? Yeah, so maybe just, first, I want to just kind of put some parameters around the problem. Let's assume somehow, like, you know, your most fanciful vision of cell therapy scale. We're making enough drugs to treat 100,000 people per year. If you do that and you think about, you know, by 2040, there's supposed to be over 15 million people with type 1 diabetes. That means it would take you 150 years, assuming no more growth in the market, to treat all the people with type 1 diabetes. That just gives you a sense of the work we have in front of us in terms of what we want to get at to meet and satiate demand. I haven't yet met a patient with type 1 diabetes who doesn't want this drug. I'm sure they exist, by the way. The idea of a single treatment that leads to them being able to get off insulin with no more monitoring, no injections, and no immunosuppression is pretty interesting. Right now, we can make enough drug, just to give you a sense, to run a phase one study. That's really about all we can do. We have work to do. I've become more and more confident over the last few weeks of our ability to get to, I kind of think of there will be three phases of this scale problem, or maybe of our manufacturing. The first will be what we do in phase one. That's it. It's just good enough. The second will be the scale we are able to reach for a registration study, which will have to be then the same process we use for early commercial launch. It will be meaningfully better than where we are today. I don't think we would take forward where we are today, just to be clear. I'm confident we'll get there. I'll come back to what we need to do. The third will be something that gets us to tens of thousands or maybe even hundreds of thousands of patients. That's going to take time to get to. You can gradually build, like any business, you don't start out on day one with peak demand. The big challenge is, yeah, so the first is you have to be able to make tons of pluripotent stem cells. You have to be able to differentiate them into islets. It turns out a couple of things are true. One, cells kind of like surfaces. Generally, that decreases shear stress. Shear stress is really hard on these cells. When you get shear stress on these things or other types of stress, you can actually end up with mutations you don't want, and/or you can affect viability of the cells. That's ultimately the challenge. People overcame that in biologic manufacturing over several decades. It used to be that they were done in pretty small bioreactors. Now they look like, you know, like your local brewery, right, when you go by. I presume the same thing will happen here over time. It may not look like your local brewery, but we'll get to some scale like that. There's work to be done. We've made a lot of progress very recently. I think the field's made some progress. I think we'll get there. I think we'll get there for launch in the thousands and thousands of patients per year. I don't think it will be 50,000 year one. I don't know if there's a distribution channel for that anyway. OK. Got it. I want to talk about the actual, the transplant itself. In the New England Journal of Medicine paper, it described, I think, 17 injections for the cell, which seems like a lot. Is there a way to optimize the surgery itself to make that a little bit more streamlined? Is that an unnecessary thing in your mind? The injections, let's just start. It was done in phase one as an open surgery just because of a desire to make this as controlled as possible. I'm confident this will be done as a series of injections that will be able to be done over time as an outpatient, assuming that the monitoring allows that to happen. The second relates to your question of what's the optimal way to do this. What you don't want to do if you just stick a whole bunch of cells into any space is you'll just create all kinds of pressure, right? You will end up with nutrients not being able to get in, and it's not a good outcome, right? If you look at the pictures in the New England Journal of Medicine, you'll see in the MRI or in the PET scan little, like tiny, almost like pellet-like things, circles. Those are intentional in that they allow perfusion, right? That allows nutrients to get there, including oxygen and things at the outset, and that's really, really important. I presume that that will always be the case. You'll want to have some way of putting cells into tracks, right, and not just injecting them. Whether that will end up being, you know, 17 little manipulations or some other number, we'll have to see. You can go into bigger muscles, and you could then end up, again, with a longer needle. You're going to need to find out a way. It's not going to be just injecting a bolus of cells in. You're putting in circa a billion cells, and they're unlikely, you know, the more you put in, if you put a bunch of them into a single region, it's almost like you create a compartment syndrome, right, in that little area. A little hypoxic environment, which we're looking to avoid. There is some potential optimization to be done there, but the process itself sounds like you have. I mean, this is simple. Just think about it like there are, every year, there are in the U.S. 15,000 thyroidectomies done. As part of every single thyroidectomy, they take out the parathyroid because if you lose your parathyroid, you probably die. Right? They ground it up, and they put it into the forearm of the arm, which means there are basically 15,000 parathyroid transplants done into the muscle of the forearm every year in the U.S. alone. It works, and it works pretty much every time. Again, because if it doesn't, the patient's unlikely to survive because they can't deal with the calcium flux, and you have cardiac arrhythmias. We can do this, right? It will end up being some modest procedure that's done that's not, again, when you think about a thyroidectomy, you don't think about the parathyroid as the big problem, but it's a big part of the surgery. Got it. OK, that's a good analogy. Yeah. Just talking about financing the company, I know that you've spoken on how you need some cash for those other two pillars of your company. What are your latest thoughts on potentially partnering SC-451 to help fund those or vice versa, partnering those to fund themselves and SC-451? I'll partner the other assets to, I think we would partner the other assets to fund themselves and to fund SC-451 very quickly. It may or may not happen. Those dialogues are happening. We will have to see how the world plays out. SC-451 is more complicated. I kind of think of it like if the only challenge you're trying to solve is capital, we probably can come up with capital in other ways to fund SC-451. I think it would be very valuable to our shareholders if we can own 100% of worldwide rights for as long as possible with this drug. That may not last forever. The flip side of that is I'd rather own 50% of something that happens than 100% of something that doesn't, right? If a partner can come to us and really kind of articulate a plan that both solves some of our capital challenges, which are real, and also really changes our probability of success, that's a very interesting transaction for us or a very interesting business development deal for us. That's kind of the dialogues that we're trying to have, which is both, you have to do, you know, we don't, we're not just going to solve it to improve the probability of success. It has to deal with the capital challenge as well. You really want to see both of those happen. If it turned out our shareholders said, no mas, we're not going to give you more money, and we couldn't find it in any other place, then just solving the capital problem would be something that we'd need to do. I don't think we're in that position right now. OK. You mentioned in the beginning the interest in the in vivo CAR T landscape right now from Strategics. What is the advantage of your fusogen program versus some of those other approaches? What is the strategic interest, if you could generally characterize it, for your programs? The most important kind of bet we made with this platform is that cell specificity and delivery matters, meaning that we go into the cell we want to go into, and we do not go into other cells. I do not think any other platform has shown you that they are very, very specific. Generally, they have pretty substantive uptake by antigen-presenting cells, which may lead to immunogenicity against the target. They also often have a lot that goes to the liver and some other places. We may not be right that cell specificity, I think cell specificity really matters for safety. I think it also matters for manufacturability. If 90% of your drug product is going into the liver, it means you have to make 10x as much drug as we do. That may not prove to be true. Because if it proves not to be true, we have made our lives more complicated because this has been a difficult platform to develop. The other bet we made in the CAR T space is that you want to integrate the DNA into the cell. Just sticking RNA into a cell that is going to go through logarithmic expansion will not be enough. If it turns out specificity does not matter, and you can just put RNA in an LNP with mRNA, it is a lot easier to make than a cell-specific VLP that has an integrating signal. I think if specificity matters, we have non-human primate data I have never seen from others that shows that really, actually, we do not see uptake in the liver as an example. It is clean outside of the cells we are targeting. That is at a very, that is using VCN, so PCR to try to find these cells, our signal in all these different tissues. We do not see it. Very specific and very sensitive test of that. We have shown that with this, we get really deep B-cell depletion with a single therapy, including clean lymph nodes if you target B cells. After when the B cells return, the drug goes away, B cells return, you get the B cell reset that everybody has been looking for in the autoimmune space. Single treatment, no lymphodepletion. Again, not something we have broadly seen. You have seen a little bit from people without the cell specificity. What do others have that we do not have? They have human data. I think that is the thing that we are lacking. I believe, based on what we've done, we've built what looks like a best-in-class platform, based on preclinical data. We don't have human data, and we need to get human data to really unlock the value that you've seen in these strategic transactions for other platforms. If we have it, I'm optimistic that we will have a best-in-class therapy. What needs to be done for, I think it's SC-451, before you could file an IND in that program? Depends on exactly what we do. Most likely, the long pole in the tent is GMP manufacturing run. From the time we say go, that's probably about a year. Right now, what we don't have is the money. We have not allocated the money to do that. That's kind of the long pole in the tent. Got it. That is on pause for the time being. Not on pause. No, no, we've continued to move forward. Actually, we've made a lot of progress in making this a better therapy. In fact, we have a very, very complex non-human primate series of experiments. It's about 20 different animals looking at different versions of ways to deliver this. I think we've learned a lot. It's a better therapy for it. It really needs to start getting going to meet the timelines we have. If we don't have the capital, it will end up being delayed day for day for the timelines we've articulated. Got it. It is just a decision for when to move forward with that GMP manufacturing. Yeah. I see. OK. Maybe just a little bit on your allogeneic CAR T cell therapies program. You know, you have data that you've guided to for this year. I guess what could we see in that data set? I'd love to hear a little bit about the paper you recently published where you showed the evasion of the immune system. Yeah. I'll start with, so the problem of allogeneic CAR T cells, autologous CAR T cells, is the difficulty in scaling them. I think people recognize that. The problem with allogeneic CAR T cells to date has been that typically, historically, what we've seen is very good initial reactions and not so great cellular persistence, likely because of immune rejection. There are other reasons why cells don't persist, but there's been clear immune rejection of these cells as soon as the immune system returns. What we have shown, we started showing this about a year and a half ago, is that when we put these cells into people, they're just not recognized by the immune system. It's actually a simpler system than in type 1 diabetes. In type 1 diabetes, you have a pre-existing immune response to the cell we're transplanting. Here, we're giving patients lymphodepleting chemotherapy. The drug itself is killing B cells or antibody-producing cells. It's not that surprising if we can do it in type 1 diabetes. It works in T cells. It does. We just showed patient data over a long period of time for many patients in this kind of cell stem cell paper that came out, I don't know, it was a week or two ago, walking through some of the data in humans in the allogeneic CAR T program. It was in one of the cell journals. That's what we have there. What we've shown you in the past is, again, we're moving this forward in autoimmune. We tried it previously in oncology. We didn't move forward in oncology. I just don't think there's that big of an unmet need right now. What we saw in the cancer setting was that we had a dose-dependent B cell depletion. That doesn't mean the B cell was deeper. Depletion was deeper. It means it was more predictable. The higher the dose, the more patients got it. You got the higher doses that pretty much happen in everybody, this deep B cell depletion. What we need to see in the autoimmune setting is can that happen safely in the autoimmune setting? Does that translate to clinical benefit ultimately? Those are the two questions. I kind of look at this. You kind of know the drug will work. What you're trying to figure out is, is it OK, good, or great? OK is it's not quite as good as an autologous CAR T cell. Pretty simple decision. Don't go forward. Good. It looks like an autologous CAR T cell, and it's a whole lot easier to make. It's, you know, available for the patient off the shelf, and it's a lot easier for the caregiver and the patient. That's probably a go, but it's not something our shareholders are likely going to pay for, right? It's both easier and it's better, right, than an autologous CAR T cell. Again, I'm not sure whether or not investors are going to pay for that. I'm more optimistic that Strategics will, but no guarantee. We're having dialogues around partnerships. I do think that there are actually a number of partners available to move forward with here, and that's the most likely path of what we do. I think at some point, given the company's capital constraints, we can't keep having three programs go. We've got to make sure we make type 1 diabetes work, and if that requires us to partner off the allogeneic CAR T cells or even not move them forward, that's just the price of the environment that we're in. Makes sense. I think we're at our time. Steve, why don't you just share a little bit of closing remarks, anything that we didn't touch on that you think is important for you to say? Maybe I'll just context that and why we're so excited about type 1 diabetes. It's really rare that you find these opportunities where, first of all, it's such a large market. Right? I mean, call it circa 10+ million patients where there's such a clear unmet need. Right? I mean, I talk to any person with type 1 diabetes and ask them about our profile, ask them how satisfied they are, and they will tell you that they are completely unsatisfied. They would love to have something like this, whether it's our drug or not. You have a disease where you haven't had any real progress in over 100 years. Patients don't live as long. They face all kinds of toxicities. It's hard to identify who our nearest competitor really is with a similar profile. I think that's a very unique opportunity that we need to make sure we go forward and execute on. There's a lot of hard work ahead of us. Optimistic, we can do it. All right. Thank you, Steve. Thank you. Really enjoyed the conversation.