Good afternoon, everybody. Thanks for joining us for the next session here today. Very pleased to have Sana with us for the session. Quickly before we get started, I need to read a disclosure statement. Please note that all important disclosures, including personal holdings disclosures and Morgan Stanley disclosures, appear on the Morgan Stanley public website at morganstanley.com/researchdisclosures. From Sana, we have Steve Harr, the CEO, joining with us. Steve, I'm going to turn it over to you to make some opening comments, and then we'll go right into Q&A. Great. Thank you, Matthew. Since you made your disclosures, I'll do mine quickly, which is we'll probably be making forward-looking statements. We and our lawyers spent a bunch of time on the risk factors in the Q, take a look at those before you make any decisions. It can be very helpful in understanding our risks. Sana was a company founded on the belief that one of, if not the most important transformation that will occur in medicine over coming decades, is the ability to modify genes and use cells as medicines, is what we call engineered cells. Our goal is to build one of the leading companies of that era. To go straight to the point, one of the most important decisions that the company made when it was getting going was that a lot of people want to pigeonhole us into being a gene therapy company or a cell therapy company. We really looked at those as the same thing, right? We are an engineered cell company. We just do it sometimes inside the body and sometimes out. What that's allowed us to do is build capabilities at a scale that we wouldn't have been able to otherwise. It's allowed us to attract better people, because the best people want to work where they can have the biggest impact. It's left us with a portfolio where we actually have very different risks. One of the great things is the capabilities are relatively similar, but our risk profiles are quite different when we go in vivo or ex vivo. The way we approached the in vivo gene modifications was a really simple idea, and that is that in order to modify the genome, you have to deliver a payload, and that payload then has to do something, right? It turns out that you can do most things you want to the genome in a petri dish, and the real challenge has been in vivo delivery. We focused on the outset on delivery, with the goal of being able to deliver any payload, DNA, RNA, nuclease, protein, whatever, to any cell in a specific and repeatable way. Every time we do one of those four things, we create a whole new category of medicines. We started with technology that allows us to do cell-specific delivery. For example, just to a CD8 T cell. We can really deliver any type of payload, right? DNA, RNA, protein. We've spent a lot of time and effort on our gene modification, gene editing capabilities as well. That's one platform. On the ex vivo side, you want to be able to manufacture cells at scale that will engraft, function, and persist, right? That's how you make medicines. The field's made really good progress in doing all of that with autologous cells, but they're really hard to do at scale. With allogeneic cells, it's very hard to get them to persist or hide from the immune system. We made the choice to really start the company around some technologies that allowed us to hide cells from the immune system, and we've made really great progress. We can now show you in non-human primates, we can transplant allogeneic cells with no immunosuppression. They live out months and months, right? Those are the kind of the platforms around which we founded the company. We're moving forward with a pretty broad pipeline. There are probably about 12 drugs in latter stage pre-clinical development for candidates, as you call them. Not all of them will make it. Our goal is to build a pipeline that allows us two to four INDs per year. We'll start as early as next year. What comes out the gate will be really focused on T cells, right? There's an allogeneic T-cell platform utilizing this cell cloaking or hypoimmune platform. Then there is an in vivo CAR T-cell generation, which is really from this ability to deliver payloads directly to cells in vivo. We'll expand beyond that. I can get into that as we go. That's just a little bit of the background. We certainly aren't a T-cell company. T-cells are an important part of what we do, and it's going to be the first parts that go into humans and likely tell us we have more work to do or we're making real progress with these platforms. We are then, behind that, areas where we leverage them going after a host of different diseases. Great. Good. Thank you for that intro, Steve. Maybe we could take each of the two platforms sort of in hand and just walk through them. Maybe we could start with fusogens and just give people some background, because I'm sure not everybody's familiar with them, even though there are a lot of fusogens out there. What's a fusogen? How are you using it? Then, why did you pick in vivo CARs as the first way to go about demonstrating their utility? First, one of my lessons has been, if you're faced with a complex biologic problem, see if Mother Nature's already solved it, and if she has, leverage that system. Viruses are able to deliver genetic payloads in vivo in very specific ways to different cell types. For example, HIV only goes into your CD4 T cells, right? COVID only goes into cells that have the ACE2 receptor on it. We have really leveraged this system in mammals. For example, human sperm only goes to human egg. It doesn't deliver genetic payload to any of the cells that it goes by on the way. We really took that system, and we're leveraging it in a cell-specific way. What we did is we took a viral fusogen, we neutered it so that it no longer recognized anything. We put on it a binding moiety, so it will find the cell that we're interested in. I guess called a CD8 for CD8 T-cell. We do a lot of protein engineering to build back up the potency of that ability to kind of sphere into and deliver genetic payload into the cell target. That's basically how we build them. We put them. You have to put that fusogen onto some lipid bilayer. It could be a cell, which we do. It could also be on a virus-like particle. What we started with was taking a modified lentivirus. Remember, lenti is a modified HIV. We took the fusogen on lentivirus is something called VSVG, that targets the LDL receptor, which means it gets into basically every cell in the body. Right? We get rid of that, and we put on our own fusogen so that we can get cell-specific delivery. We can then put in different packaging, so we don't have to use all the random integration of lenti, or we can choose to use it. It's our choice. That's what we started with. The reason we went after T-cells first was, one, it just happened to be one of the first places we got this to work. We can kind of get it to work in almost any cell type you ask us to, but where we got it to work with really high efficiency. Right? The second is that we're using a payload, essentially a carrier, that has a pretty limited volume of distribution, right? It doesn't go everywhere in your body. That is beautiful from a safety perspective if we're going after things that exist in the blood, bone marrow, spleen, and liver. It's challenging if you want to go in the brain, right? Because you have to direct it. It's really nice from a safety perspective for us to go after T-cells or HSCs, just because they can get there pretty easily. It doesn't go a lot of other places. It was kind of practical and biologic that led us there. It's kind of where we got it working first. Maybe the third is it's a really simple place for us to prove this works well enough or not, right? We know the model. Let's make an example. We went after Alzheimer's disease, and we didn't have the right effect. We wouldn't know if we had the wrong biologic hypothesis or if our platform didn't work well enough. By going after creating a CD19 CAR T-cell in vivo, if it doesn't work really well, where the biology has already been proven, if we get it right, it will be a very meaningful medicine because it's just a marked improvement over the way things are done today. If we have a challenge, we'll know it's because our platform isn't working and we need to modify it, not because we had the wrong biologic hypothesis. CD19 CARs clearly kill cancer cells, right? That was kind of why we chose that. Okay, great. Good. Maybe since you're obviously not in patients yet, talk about the de-risking data that you do have, especially the de-risking data in animal models to give you confidence that you have solved some of the problems that you talked about. There are a couple things that we've done. In vitro, we want to show as much as we can that we are specific and that we're highly efficient in getting into the right cell. Right? That's done. The second is to look at really in vivo, first of all, can we transduce a reasonable number of cells in models, which is true, and then what is the biologic impact, right? Two ways to test that. One is there is kind of the definitive mouse model that's been utilized for B-cell malignancies across the board for CAR T-cells. It's called the Nalm-6 mouse. There what we do is we compare a single intravenous injection of our medicine into the mouse with kind of like a CAR T-cell that's been made outside the body and delivered. Our data is in the S1 or our presentation. You can see we get comparable efficacy, right? The second, and I would say a more challenging one, is to put this into a normal immune monkey and deliver a CD20 CAR. It just turns out you have to do CD20 because CD19 doesn't cross-react between humans and monkeys. See if we can deplete B-cells, which is the target of CD19 or CD20, right? That gives us a real insight into, in a normal immune animal, are we able to deliver with just a single intravenous infusion, enough medicine to make enough CAR T-cells to have a clear biologic effect. There, in the majority of monkeys, the first time we did this at a single dose, we saw B-cell, really meaningful B-cell depletion. That was, to me, that's kind of the killer experiment. That leaves us running really towards three things to get our human testing going. 1 is scaling GMP manufacturing. Always easier said than done. Right? The second is these animal models aren't really that. They're not the same as going into a human cancer patient and doing our best to understand what should be our starting dose. The dose is really not clear. The third is the normal, is its pharm/tox work. Where does this go? Does it go anywhere besides T-cells? When it gets to whatever cells it goes to, how does it integrate? To be really clear, CD8 is on T-cells and some NK cells, so we will get into some NK cells as well. You can say that's good, and you can say that's not so good, depending upon your belief of NK cells. I happen to think it's good. That's a little bit around kind of where we are. Okay. Okay, good. Our goal is to be in human testing, right? As things go well, we'll be there, hopefully next year, we'll get that, off we go. Yeah. Maybe now is a good time to touch on manufacturing, what you're doing there in terms of being ready and any sort of unique challenges that you face in manufacturing here different from some of the other cell therapies. Yes, it's very different. Think of this more like AAV or lentivirus than a cell therapy. It's a gene therapy in terms of manufacturing. If you watch this AAV panel a couple weeks ago at the FDA, I think one of the things you noted is, if you watch this, the FDA is getting to understand that we measure dose, and there are capsids that have all the genetic material, and there are some that don't. Right? This is a system that people have been working on for 20 years. Right? You can imagine a system that we are getting going, we have to really understand what percentage of our cells are really well packaged and what happens in those that aren't. We have to scale that process in a proprietary way. Again, like any gene therapy, you're going directly into the body, so you want it as pure as possible. Right. This is one where manufacturing is a challenge, and I feel good about where we are. We don't have every I dotted and T crossed, but I feel good about where we are, to go in and run our first in-human studies. I think if you said, where are you in terms of commercialization, we have work to do to be at the scale, and I would say quality and predictability of what we want to have for when we would want to globally commercialize this. I don't think that's unique. Antibodies have that, antibodies, you know exactly the roads you need to take to get there. We're more like the antibodies 15, 20 years ago, where we're going to have to figure out some of that road as we progress going forward. We have work to do for some of that. Okay, perfect. Before we move on to hypoimmune and ex vivo concepts, maybe just talk about beyond T cells, how you're thinking about using fusogens? How we're thinking about doing what? How you're thinking about using fusogens, but beyond T cells? Yeah. There's a lot to do in T cells, to be clear. There are a couple things that are true. One, using the carrier we use today, as I mentioned, our volume and distribution is somewhat limited, and it's a great thing. The cell types that are most obvious for us to go after would be things in the liver, either hepatocytes or something like liver sinusoidal endothelial cells, something like that. Different cells in the liver, and then hematopoietic stem cells. Right? We can get at both of them with reasonable success. We now need to take that reasonable success and turn them into medicines. Right? If you look at those, both for prioritization and complexity reasons, as well as just for bandwidth and biology reasons, anyway you say it, those are earliest are a couple of years out, right? We're really focused next year on bringing forward the T cells. What I would say as we exit T cells, I think it becomes more and more important for us to think about what payload we put inside the cell and how we deliver it. Within T cells, we can rely upon millions of patients who have HIV and thousands of patients who have had CAR T cells to say that when you go to T cells, you can integrate DNA safely and not lead to T cell malignancies. As you exit T cells and go into hematopoietic stem cells or hepatocytes or other cell types, we don't have that same comfort. Right? Doesn't mean they're not safe, we don't have all that data. It makes more sense to us than those to maybe go and deliver the gene editing payloads, whether that's simple things, call them CRISPR, TALENs or whatever, or more complex things like base or prime editing or some other novel things that deliver bigger or different things than just knocking things out. Stay tuned. Those are the things that we're doing there. We're making real progress, both in getting into the right cell and delivering new interesting payloads. Okay. Great. Good. Why don't we turn to hypoimmune then, and just like we did with fusogen, and maybe explain the concept behind the hypoimmune platform and what your first target is there. Yeah. Really, as the field of stem cells got going in the aughts, 1 of the first things that the real leaders realized was that unless you could overcome the problem of allogeneic rejection, the field would be pretty limited in its impact and for human therapy based. You put my cells in you're going to reject them in this form. A couple of different places got the same advice, and that was, this isn't that complicated. Really what you need to understand is the paradox of pregnancy. The paradox of pregnancy is that we're all half mom and half dad, and the only reason we're on this call together is our mothers didn't reject us. Really, none or very few of us would be good organ transplant donors to our mother. Really what's different about that maternal fetal border was the question. Really, the teams came up with a really clear roadmap, I think. The system that was built really seems to be working. What we've done now, the challenge, and this is where the field has struggled, is you have to grapple with two arms of the immune system. There's the adaptive immune system of B and T cells, and there's the innate immune system with things like macrophages and natural killer cells. The way that you deal with the adaptive immune system, generally, is you get rid of MHC class I and class II. That's been known for a long time. The challenge is that that's what viruses and cancers do to hide these cells from the immune system as well. We've evolved a natural killer cell to go after cells that miss this, and really figuring out how to turn off both of those arms of the immune system at the same time has been something that people have struggled with. As far as I can tell, we're the first group that's really gotten this to work. Again, we have to get it to work in humans. Where we are is we've shown that we can inject allogeneic gene-modified cells into monkeys, and with no immunosuppression, they will live for months and months and months. We can do that de novo. We can also first inject non-gene-modified cells, which the monkey will create an immune response to and reject, and put our cells in. Even when there is a preexisting immune response, our cells are hidden from the immune system. It gives us a lot of optimism that we can deliver allogeneic cells, whether that's T cells or stem cell-derived islet cells or anything else, that we can do that, and we can redose if we need to. Right? We can do that even if a patient has a preexisting immune response, like in type 1 diabetes and multiple sclerosis, and still hide these cells from the immune system. Where we are, generally, is We've done really, I think about as much monkey work as we need to get into human studies, and we're really making GMP reagents, and GMP therapeutics to move into human testing. Hopefully, we'll get the IND in next year for the ALLO-T program. That's where the odds are stacked in our favor. Right? You're putting an allogeneic T cell into a cancer patient who's immunosuppressed from cancer. They get a little conditioning chemotherapy because that's what you need to do to get any T cell, autologous T cell, or allogeneic to engraft. We're going to go with CD19, knock out their B cells. End of the day, we probably need B cells to last for a few quarters, right? That's the easy one. The hard one, which we'll do hopefully, the IND in 2023, things go well, is type 1 diabetes is probably the first place. Can you really go after really hard problems where you've got a preexisting immune response to the cell, there's going to be no immunosuppression, and you want this to last for years and years and years to be valuable. Those are a couple of programs that I'm really optimistic about. That's sort of where we are. No, that's perfect. Can you talk about cell lines, cells, all the work, because obviously you've got the technology to avoid the immune system, but then another piece is getting the right cell lines, getting the right cells and production there. Where are you in that regard as well? Really, we're applying this in two different camps, right? One is a donor-derived allogeneic T cell. It might be you or me just donating T cells. There we have to do three things. Right? We have to show that we can get very high efficiency in our gene modification. Right? We're doing five modifications, so it has to be very efficient. The second is that we can create very high-quality T cells out of that, right, at scale. The third is that we can control the donor variability. If for one group of patients it's me and others for you, that they get the same product, right? I feel we're really good on the first two. The third one, not that we've had a problem yet, but we're still in the process of proving to ourselves that we can do that. Right? That's where we are on that. You have the stem cell-derived hypoimmune cells, that's a more complicated endeavor. First of all, we have to start with a GMP, it's called iPS cell bank, with freedom to operate. Right? There's a lot in there. Once you have all of that, there are only a subset of those that meet our immunologic criteria. You have to do two more things. You have to ensure that they really do go to the cell type you want. iPS cells sometimes love going to liver cells and hate going to beta cells, as an example, or heart cells, as they say, right? They have predilections where they like to go. The second is that you need genomic stability through that. The way that the field has generally looked at this in the past has been karyotyping, which is a little bit like trying to figure out if a patient has cancer by palpating their abdomen. There are more sensitive ways to do this and trust that we are doing them. That just takes time. We have to do the gene edits and then ensure that, again, we have a very stable genomic stability, and that that gene-edited iPS cell will go to the cell type we like. It just takes time. It's very complicated. Making sure that we have GMP reagents all the way through, that they're very high quality, and we don't know a lot about the biology yet, and that we have freedom to operate. We're working on all of those, and again, I feel really good about where we are, but it just takes time. That's why with the iPS-derived hypoimmune cells, it will be a few years before we're in human testing. It's just because all of those things take time. Okay. Good. Maybe a broader question just about initial data. Let's just for sake of argument, assume with both fusogen and hypoimmune, you can file an IND sometime next year and maybe in 2023 or maybe it takes a little bit longer than that, you have some initial data. How do you think about that initial data in terms of, let's call it de-risking the overall platform? What does that do in terms of demonstrating safety or efficacy or just delivery across those platforms to then allow you to maybe move more aggressively across a range of different biologies? I would say, that gets at where is their correlated risk, right? In some regards, right? Correlated risk, there's correlated upside too. With the allogeneic T cell program, the question is, have we really nailed the ability to hide cells from the immune system? Ultimately, that is the question. There are two separate questions in there. One is, when you do that to an allogeneic CAR T-cell, do you see a meaningful clinical benefit in terms of durability of response? That will define how valuable what we're doing is in the oncology field, right? We're behind, if it turns out that they live longer, but it doesn't matter, that's a lot of work for not a lot of effort, right? A lot of upside, I should say. The flip side of that is, if they live a long time, what you've done is you de-risk the whole platform. Right? Now when you think about the beta cell program, your probability of success goes way up. All the other glial progenitor cells, all of these things that we're trying to hide from the immune system, they really do change in terms of how you think about it. Really, There are going to be two questions that come out of that. One, does it matter clinically? Does it happen? If it happens that they live longer, the whole platform becomes more valuable. Then, does it matter clinically, which will tell you how valuable the allogeneic program is. Sounds right? On the fusogen platform, what you're really asking is, this works. We can deliver payload in vivo in a relatively cell-specific way. It's really, can you do it well enough, right? I think, to the extent that you can, and that you can manufacture at that scale, right? You get more confident that these other cell types we're going after, we'll figure it out. We'll get there. Right? Again, I think that in and of themselves, it will be very valuable if you can show a single injection, no conditioning chemotherapy, nothing else. One treatment, and a patient might have a curative cancer therapy, with really, hopefully, very limited toxicity and ideally nothing more than what you get from a CAR T-cell, maybe even less than what you get from a CAR T-cell and CRS and neurotoxicity. Might be worse. You have a really important drug for really generalizing accessibility to patients. I do think that they're very independent platforms and they're very independent in terms of their risk, but within each of them, they will read through to multiple programs. Okay. If you fail, you will say, "Boy, you guys, If it doesn't work for T-cells," you'd say, "How's that ever going to work for islet cells?" Right? Yep. We have some real thinking to do. Yep. Okay, good. Steve, maybe in the last two minutes, we could just touch on some of the, I guess, what should I say? More far-out stuff that you're doing, maybe some of the cardiac regeneration work that you're doing or otherwise. I'll let you pick sort of what you think is most interesting. Well- Maybe just give people a view into some of the other stuff you're working on. It's always fun to talk about some of what we've just made a little bit of progress and that maybe people haven't paid that much attention to. The cardiac's a good one. The idea here is that still, if you go through all this, the number one killer in the world is heart disease, and heart disease is worse than all cancers put together. In the U.S., they go neck and neck every year. There really have been very limited novel medicines, right? The problem is for most patients who have congestive heart failure, it's a disease where the cells are just gone. You've had a heart attack, they're not there yet, and so they need to put them back. The challenges are threefold. You need to put in cells that engraft and function. You need to hide these cells from the immune system, and you need these cells to integrate with the electrical conduction system of the heart. In particular, what we've seen is that first few weeks, there's a real risk of an arrhythmia. We've shown you data that says, "Hey, we can put cells back into a monkey, and that monkey will get largely recovered to almost normal heart function." It's in a small N, so be careful with it, but they clearly recover, at least meaningfully. We've shown through data from our hypoimmune platform that at least in certain settings, you can hide cells from the immune system. What we recently showed was that through a series that Charles Murry did through his lab, is that we can make a series of gene edits, some you turn on, some you turn off, and you make these cells, they're no longer arrhythmogenic. They integrate with absolutely no issue around arrhythmia. What we haven't shown is that they function as well. We have to show that over time. We've shown they engraft, and we've shown they function, but we need long-term function because you are knocking out some pretty important risks. Most of the time, I think about these ion channels and hearts being there for a reason, and so we still have that to do. I look at that as a really good example of a hard problem that we tackled head-on and where, through really a combination of rational design and a little bit of luck, made some substantial progress over the last few months that could lead to a really great drug, which again, if all went well. That will take a bit more time for us to enter into human testing. To put all those edits in, it would probably be, again, a few years out, but something that I would say is very exciting. Okay. Great. Well, wonderful, Steve. Thanks for being here. Thanks for spending some time with us. Always a pleasure. Great to see you, and thank you everybody for your time today. Appreciate it.