Morning, everyone. Welcome to the Jefferies 2026 Global Healthcare Conference. My name is Roger Song, Senior Analyst covering SMID-cap biotech. It is my pleasure to have our next company, Sana Biotechnology. We have CEO Steve Harr here. Welcome, Steve. Thank you, Roger, and thanks, everybody, for joining both online and in person. Awesome. All right. Maybe, Steve, take a moment. What's the state of R for Sana? You have an exciting year ahead of you. I think everyone's looking forward to that. Maybe just give us some high level. Yeah. I'll start, since you asked me to look at the year ahead of us, by making sure everybody knows we'll make forward-looking statements. Take a look at our disclosures to read about our risk factors. We built the company around two platforms. One is an ability, we hope, to hide allogeneic cells from immune recognition when transplanted into patients. We call that our hypoimmune platform. We've done a lot of work to show that this works really across multiple species, whether that's non-human primates, mice, humanized mice, and we've even done some human work, which we'll get into. We're applying that in, I think, its most advanced and maybe most valuable form for a potential functional cure for patients with type 1 diabetes. I'll come back to that in one minute. The other platform was set up around in vivo delivery, so to be able to deliver DNA or RNA or other genetic material directly and specifically to cells in the body. The most advanced program around that is an in vivo CAR T platform with two different assets moving forward towards the clinic. That's a place where it's gained a lot of excitement over the course of the last several quarters. I think we have been in the forefront of moving in this space for a while, and we took our time and tried to make bells and whistles that we think will be really important in making a best-in-class therapy. All that being said, the most recent information is highly compelling from people in the field. That excites us. It gives us a high bar to stretch for. We think we can do it or do better, and we'll be able to tell you about that in the not-too-distant future. To take a step back in type 1 diabetes, and first I'll lay the frameworks. I think it's often a disease that's underestimated. There are about 10 million people in the world with type 1 diabetes, and it's growing at a mid-single-digit rate per year. People who are diagnosed. I always think about this in the context of my daughter. She's 22 years old. A 22-year-old woman who's diagnosed with type 1 diabetes has a shorter expected lifespan than a 22-year-old woman diagnosed with either HIV or breast cancer. It's pretty astounding when you think about that. That time is very difficult, right? You have a daily burden around insulin and food intake that you have to grapple with, exercise if you're feeling a little bit sick. That time is also spent with a number of comorbidities, such as blindness, amputation, heart attack, stroke, and hypoglycemia and coma risk and things like that. We have to do better. The field has made a lot of progress, and we stand on those shoulders. About 25 years ago, someone named James Shapiro in Canada started taking pancreases from someone who was recently deceased and isolating the islets. To take a step back, the etiology of type 1 diabetes is super simple. The immune system kills the pancreatic beta cell, and the pancreatic beta cell is the only cell in our body that makes insulin. Without insulin, sugars just float around in our blood. They're too high in the blood, but our cells starve to death because they can't get any sugar. 100 years ago, there was the invention of insulin, and what James Shapiro started taking was pancreatic islets, which think of as a beta cell plus its support structure, and transplanting them into patients with type 1 diabetes. What he's found is that patients can live for a long time off insulin, and they do quite well. There are two problems with that. Number one, cadavers aren't really a great and predictable source for islets. Number two is, like any transplant, these patients have to be on lifelong immunosuppression. There just aren't that many people for whom lifelong immunosuppression is better than lifelong insulin. The second step is that several in the field have now taken stem cells and you can differentiate them into islets. They've shown they can transplant these, and patients do quite well with much more predictable, probably much more scalable, but you still have to be on lifelong immunosuppression. What we did about 18 months ago was gene modify in a proof-of-concept experiment, cadaveric islets, and transplanted that into a single person in Sweden. It's great to say that this person is now living and making insulin in his body for the first time since 1987. These islets are living with no immunosuppression, and they are functioning, and they are persisting. You put that all together, gene-modified stem cell-derived islets, and you have a chance for a functional cure, a one-time treatment for type 1 diabetes. We've been working super hard. It's been a long slog to bring this into humans. Our goal is to have an IND this year, begin our trial this year, and I'm quite optimistic that in less than 12 months from now, we'll know whether or not this is really working. Similarly, on the in vivo CAR T, and I'll be much briefer there. We'll start the study this year, begin to get data this year. When and how we share it, we'll see, but again, within 12 months, we'll know if that platform is working. It is something where we have, I think, a pretty meaningful opportunity to really see how these drugs work in people. Once we see how these drugs work in people, to rapidly move them through clinical development and towards the market. Excellent. Yeah. It's meaningful and transformative trauma for both platforms. We focus right now for the type 1 and the other, the alloimmune, hypoimmune platform. For the type 1, I think you're guiding a couple months, a couple quarters, you say as early as 2026, you're going to file the IND. I think recently you changed, even updated that to start phase I in 2026. What's the remaining steps? I think we talked about this in terms of a master cell bank. Just walk us through that, a little bit of technical and what's remaining to be done. Yeah. One of the real challenges that we have accomplished is making a gene-modified master cell bank. A master cell bank is the single cell from which the product starts and grows forever. Right? When cells, and stem cells are no different, divide, they tend to make one or two mistakes, and it's really not a big deal. Mostly, it's in DNA that isn't really copied and expressed. When you're making trillions and trillions of cells, what you can see is that particularly when you're doing that in a growth media that you're trying to make cells grow fast in, right, which is what you're trying to do when you're making the full cells, you can actually select for DNA repair enzymes, mutations, something like p53, use that as an example. That really can be problematic. That took us years, but that's in the rear view mirror. We've made a gene-modified, genomically stable GMP master cell bank, and we have hundreds of vials of it. We take each one of those vials, and we make hundreds of vials of a working cell bank. You thaw one working cell bank vial, and that's how you start a manufacturing process run. That's basically in the rear view mirror. What's left to do is really two things. One is to complete a non-clinical package. That's generally just around things like GLP toxicology study, efficacy, biodistribution. Where does it go? Could this drug go anywhere else in the body? That's nearing completion. I don't think that's going to end up likely being a problem. It's one of those things, though, if it does become a problem, it becomes a real problem. We're optimistic it won't be. We're also in the middle of transferring a manufacturing process that we utilize in our own labs, that we developed in our own company with our own people, into something where it's in a GMP facility, utilizing an outsourced group of people. That process is ongoing. It's, again, one of those things that it will happen. It's not really an if. There is always a component of when to this. Things don't always go perfectly during technology transfer across sites. I'm optimistic. We have plenty of buffer, in the guidance we've given you. We've eaten into some of that buffer because now it's not like it's super simple. We still have buffer, and we are on track to do what we said we would do. Okay. That's an interesting comment. I know you have some buffering because that's why you would say as late as 2026. Well, I always say as early as, because I don't want to get into some debate with a lawyer or regulator. Is it really going to happen in 2026, or could it happen on January 1st? How do you change your disclosures? This was a lot easier to say as early as, because I know it's not going to be in 2025. It's true, right? Yeah. It's already in. Yeah. Got it. You did mention you do have a buffer, but you think you ease some or add some buffer. How much can you say about that, or it's just a comment? It's just a comment. It's just a comment to say this is hard. While we are optimistic we'll get it done, there aren't guarantees yet that it will get done, right? We still have some work to do. We still have some ability to kind of. Our team still has some leeway if they need to take a bit more time. Those are things we're working through, and again, optimistic we'll get this done. Got it. I think it's super straightforward. We can get into what it then takes to understand, does the drug work? Right? Is that worth jumping into? Yeah. Go ahead. Yeah. Because once you start this study, I think there is three simple questions you want to ask. As we move across to this new technology, do these cells engraft to evade immune detection and function, right. You should know that within, the last time, it was within about four weeks, we were very confident that these cells wouldn't be rejected, and they would live. I think similarly, if the cells engraft in the first patient, and they're not rejected by the immune system within three to four weeks, they're going to last a long time. Right. There's just not a component of the immune system that pops up later than that. That, I think, would be part one. At that point, you know this technology works. The second thing is we're not actually trying to evade the immune system. That's just part of our goal. What we're trying to do is make a curative therapy. Right? Are our cells potent enough that you get patients off insulin, and they're able to live a normal life? Again, that should take a few months longer, but not forever. At that point, you know you have a really viable drug. The third question is, how generalizable is the result? Right? That's one of those things where if we're six for six, you can be pretty confident it's very generalizable. Right? If we're three for six, you might say, "You know what? I want to see a little more." Right? I don't know how long that would take, but those first two will happen pretty quickly. The third one, I'm optimistic this will work in maybe not everybody, but most people. The catchment of people we're going to enroll in the study is more or less all comers. I'm sure we'll get into that. It's a very broad catchment of patients. Okay, got it. Yeah, you basically set the goal for this phase I to achieve in terms of the rejection and then the curative thing, the potent enough, and then obviously the end, right? We need to repeat that. Okay, got it. Yeah. Let's take a step back. Okay, you want to enroll the patient, what's the baseline characteristic you are trying to do? You say all-comer, any additional details we can say this is a no, all-comer. It's essentially all-comers over 18. We're not going to put this into really old people to start, and we're not going to put this into really young people to start, right? It's an experimental therapy, you want to have equipoise, which is likely benefit outweighing risk. The second is we won't put in someone who had a heart attack yesterday. That could be confusing, right? Some element of relative health. Beyond that, there are a few small exclusion criteria, but it's essentially going to be all comers. Assuming success, we would then look to expand that as part of a broader phase I program into 16+ and then 12+ because teenage years. Ultimately, we want to get down to two and 6+. That may take a little bit of time. We'll work on the over 65 population, as I think older people will also want this. I don't really feel like you're 65 that old anymore. It's just around the corner. Anyway, that's a little bit around what it is. It's most comers. It's not going to be brittle diabetics or people who have hemoglobin A1cs over some number. Frankly, if you do that in a study like this, people will just let their hemoglobin A1cs drift up, so they can hit the entry criteria, and then they'll be part of the study. That will be really not a good thing for a patient. We get a lot of calls to be in this study. There's a lot of demand, and so we don't need to have a whole bunch of patients going out of control just to try to hit entry criteria. Got it. Just want to also emphasize, you do have other company also doing the cells therapy for the type 1, seems the inclusion criteria are a lot more restrictive compared to Sana because they require some hypoglycemia event, even more severe event. That's not the inclusion criteria or exclusion criteria you have. It's not the patient population we're targeting, first off. We're really going after all comers. Generally, others in the field-- By the way, I don't consider this competition, and I don't say that arrogantly, I say that humbly. There are 10 million people with this disease. If we somehow just nail this, and with the best possible outcome you could think of, it's one-time treatment, it works in every person. We somehow scale this to 100,000 people per year, which would be spectacular for a cell therapy. We will take the global growth rate from 5% to 4%. There's so much space for so many different approaches here. We need help in conquering this for patients. The population that we're going after is pretty much all comers. Others, if you have immunosuppression on board, you may need to go after a sicker population. At the same time, that sicker population has a real need for a novel therapy, so it's great that you have other people going in that world. Yeah. Got it. A little bit more on the phase I design, and then how should we think about the would you do a dose escalation, or you start with potentially therapeutic dose, and then how many patients for those cohorts if you want to do multiple cohorts? I think you know what the right dose is, more or less, right? There's such extensive experience with cadaveric islet transplants. Our cells may be moderately different in potency, but it's not going to be meaningful. I do think we'll be at a therapeutic dose out of the gate. We could be wrong by one dose or something, but we should be. Really, if you think about the side effects here, they're not generally linked to dose. The first thing we worry about is severe hypoglycemia right after the transplant. That can happen because cells die, and they have insulin granules in them, and they release them. It's manageable very simply by just monitoring, seeing it, and if a patient develops it, giving them IV glucose, right? Not challenging, it will be gone within a matter of hours. Obviously, you don't want to miss it because that's something that could be catastrophic for a patient, but it's not difficult to manage. The other is, I think, more of a long-term challenge, which is when you're making stem cells into islets, there's a risk that some of those cells are off-target cells or the wrong cells, and those cells could continue to divide and create a tumor or growth or something else in the patient. That's going to take years, potentially, to see itself and develop. It's unlikely really that dose-dependent. I think we'll be in a dose range to start that we're optimistic could be therapeutic. The very first patient that was dosed in this trial from Vertex Pharmaceuticals was in The New York Times, just to give you a sense of it went very well for that person. I'd like to hope that similarly, we can offer a benefit for a patient right out of the gate. Yeah. You mentioned Vertex, and certainly they are pioneering this, but complete different setting because they are not hypoimmune, and then also different stem cell like you. I think in terms of data disclosure, right, so if once you start to enroll, maybe one is how quickly you expect the enrollment going to look like, and then two is how soon we can see the first data. You think you are guiding around 12 months we can see that? Yeah. I think enrollment will start very quickly. I'll start there. It's a pretty unique program, I think, in terms of just external interest. When we'll disclose data, we'll let you know when we know something. Right? It doesn't really help you to say we're still learning. We don't really know yet, but if we learn something, we're small enough, and this is important enough, that most early learnings will be material to the company. I think at that point, we'll have an obligation to let you know. We will do that as it comes about. Exactly predicting when that will happen or what will be material is difficult ahead of time. I think we have a relatively low bar understanding the importance and external interest of this. Got it. On the flip side of that, the long-term view and really I presume investors are aligned with this. We need to get these data presented at peer-reviewed scientific conferences, not just in press releases and things like that. We will need to kind of balance those two as we go forward. Got it. It's the first in human in this product for sure you will disclose data as you learn something material. How should we think about at what level of the maturity you can have the regulatory discussion and then thinking about even moving to the pivotal stage? I don't think it'll take much. I think it's about 12 patients will allow us to move forward. I think the harder part is not the clinical data. I always talk about the scientific challenges here, and we've done pretty well in tackling them so far. We're not done. One is overcoming autoimmune and allogeneic rejection. I think we've shown we can do that, right. We've done that in the papers published in the New England Journal of Medicine. It's done very well. The second is we need to make this gene-modified genomically stable master cell bank. That was super hard for us. We've done that. The third is make the drug at a purity, potency, and yield to allow us to run a phase I study. We've done that more or less. We barely did it. I mean, it's like we're hanging on. Right? This is not a product we would ever take to market as currently. Actually, the fourth is to make a process that is at a scale of purity, potency, and yield that's really commercially important, and we can serve an unmet need in a large population. We have work to do to do that, to be clear. I think the rate limiter to us moving into a pivotal study is much more likely to be having a locked commercial process than it is to kind of get 12 patients through an early-stage study, which should enroll very quickly. We may get that done in a timeline that there's no delay. We may get that done in a time that allows us, and I'll take the lemonade side of this, to explore other patient populations, like younger patients or older patients, before we move into our pivotal study. I don't think it will be that long to do it. I mean, we've made a lot of progress on it, but we still have a bit of work to do, and you have to assume that once you've locked your process, it's still another nine months or so at least before you can start enrolling a patient. We have work to do to get there. Yeah. You have time. As you're generating clinical data, you continue to lock in the commercial process for the pivotal. We're working on it already. Yeah. Okay. You're working on it. Yeah. Good. This is different than some other fields that you're used to, where the science is understood and it's a capital investment that needs to take place. Right? If you kind of think of it as you have the scale is number of batches per run times number of runs. We're still in the number of batches per run problem, which is a scientific problem. Right? That's good from a capital efficiency perspective. It's a bit more of a scientific journey we're still on and some unknowns that we have to grapple with. Yeah. The foundation of the science is amazing, right? You just need to figure out a couple technical part. In terms of the current expectation for the pivotal, do we expect it also will be the all-comer? If that's the case, and then how many patients you actually need to enroll compared to type 2? I think type 1 is different kind of situation. Yeah. I think it's going to be something on the order of a total of. You have a precedent for another field. It'll be like 50 patients. The precedent is about 37, total of 50 between phase I and phase II. I can't imagine we're logarithmically different than that. I can make an argument we should be smaller. I can make an argument we should expect to be bigger. It won't be meaningfully different. Again, it's relatively straightforward. You know if this works or not, right? If we have a patient who's insulin-free, that is implausible. It's entirely implausible without the intervention. There doesn't need to be some control arm around other mechanisms for control, right? Patients will die without this. They will die within a matter of months. Yeah. You're curing those people, giving them the insulin or giving them the beta cell. In some regards, it's like the clinical simplicity, and I think actually the patient passion, like an orphan disease, right? That you see or even an ultra-orphan disease, but with the market size and the unmet need that is of a very large market. I think you mentioned cell therapy. I cover a lot of the cell therapy. The long-term risk is always the case, but how you think about the initial approval, right? You may get to six or a year follow-up to get to the efficacy, and then you can get approval label, and then you keep following them as a post-marketing, or is that the case? How long you need to follow them before you can actually get the initial approval? Well, let me just say, we've had no discussions with dialogue with regulators around what a pivotal study might look like. Anything that I'm saying is speculation. Right? I don't think we really know. I know we don't know. If it were me making the decision, I would ask us to do a relatively minimal. You're going to know if it works or not. Right? The safety is likely something that is rare and long-term, right? There isn't a lot of value in a larger registration study. I'd have a relatively small registration study and very onerous post-marketing follow-up requirements and something like that, where we have to really kind of keep an eye out. We have to anyway. It's a gene-modified therapy, right? A 15-year requirement by law for following every patient. We will have that in our bag of things we need to do regardless. Yeah. Okay, great. I think we should touch on the in vivo CAR T platform. I think a lot of interest, a lot of the buzzword or chatter out there. I think your in vivo CAR T is different, right? Your combination of many delivery, gene editing, and then the mRNA, the construct. Tell us why you choose the platform you're using now, and then how you think you will be different from others. Yeah. When we started this, we made two really fundamental assumptions, and they may prove to be right, or they may not be. The first one was that the signal or the gene that you insert needs to integrate into the target cell's DNA. The reason being is that in the CAR T space, there's another alternative, which is just put mRNA in, right? You might make 100 million CAR T cells, something like that, and if you just look at B cells, that's all you want to get rid of is B cells. We have a few hundred billion of them. Right? What you see in the CAR T field is this logarithmic expansion of cells. Right? Expansion being growth. They divide, and with mRNA, when you divide, the signal can't go with both progeny. Right? The mRNA just goes with one. When it's in the DNA, the progeny each have the same signal, so you can get that logarithmic expansion. We believed you had to integrate. That's part one. I think that's generally proven out so far in the clinic with the really robust data you've seen from other VLP, virus-like particle approaches. The second is that cell specificity matters, and so you only want to go to the cells that you're targeting, the T cells. I think many others would say that's not as important, as you just need to get enough cells into the T cells, right, and the others will take care of itself. If it turns out both those things are true, we have a best-in-class platform. I think I can be fairly confident in saying that. If it turns out that neither is true, we made things really complicated and other things are going to win. Right? That's the bet we made. Right? We're different, and I think we've shown that in our ability really to target just the T cell, and that's something we think it will help us from a manufacturing. T cells are rare in your body, so if you go to other cells, you're going to have to make a lot more drug just to get in because you're going to go to the liver and other cells. It's going to help because when you go to off-target cells, particularly antigen-presenting cells, you run into immunology problems. Right? The third is I don't think you really want to be inserting DNA and expressing proteins in off-target cells, just because there's only bad things that can happen. It may not happen, but nothing good's going to happen. That's why we thought this, and it's taken us some time to really make this work. We almost went into the clinic a few years ago and really kind of ran into some challenges that thought would make it less likely our drug would work. I think we were right in taking our time, and I think now we have a very high probability this works. It's not guaranteed, but it should work. We'll have data very soon to bear that out. Makes sense. Yeah, it's interesting. It seems you are more targeted, and then I would argue it's even more powerful because you integrated it to the DNA versus others with more RNA kind of a peripheral. How should we think about the initial clinical setting, and then what is the data we're going to see in the coming year? Yeah. Maybe one other thing that's different. What you see if you look in the literature as these virus-like particles are starting to get going, is you've seen a new side effect emerge, which is a post-infusion toxicity that occurs within hours and lasts for a few days, and it's like a second case of CRS, or cytokine release syndrome. Patients have fever, cardiovascular collapse. It's been manageable. No one's died or anything from it yet, but it's been problematic. That happens most likely partly because it's a virus. It's a virus-like particle. A lot is because the portal of entry that all the other technologies use is CD3. When you bind a CD3, you activate the T cells, and you get this very rapid activation of T cells that is helpful in helping these T cells grow but is difficult in terms of a side effect profile. We use a different portal of entry into the targeted T cell. I think that gives us just a, again, hopefully something that's a bit more of a safety advantage. We'll have to see. These drugs look pretty good so far, so it's a high competitive bar for us. Our path here, we have a CD19 that we're going to start in lymphoma, non-Hodgkin lymphoma, large B-cell lymphoma. If that works, which I hope it does, we will continue down the path of expanding oncology, and we'll rapidly, hopefully, move in the autoimmune disease setting as well. We also have a BCMA, which we'll begin human testing a year from now or something that is also right in there. We're not going to pile failure upon failure, so if the first one doesn't work, we probably don't go forward with the second. If the first one works, we really unlock a lot of opportunity of potentially to move into CD19 and on multiple tumors, lymphoma, leukemia, to move into multiple autoimmune diseases, and to really push rapidly with our BCMA-targeted CAR as well. Excellent. Awesome. Yeah. All right. Thank you, Steve, for being with us, and thank you everyone for listening and watching. Thank you, Roger, and thanks everybody else. Have a great afternoon or morning. Take care.