Sana Biotechnology, Inc. (SANA) Earnings Call Transcript & Summary

There it is. Okay. Thanks for joining everyone. This is the fireside chat with Sana Biotechnology. Happy to have with me CEO, Steve Harr. Steve, thank you for joining us. Really appreciate it. Before we get started, just let me read this disclosure statement. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures.

With that, let's go ahead and get into it. Steve, I think at this point, people are likely familiar with the Sana platform, but just to level set for everyone, do you want to talk for a minute about what you think some of the key milestones have been for the business recently and just what the hypoimmune approach generally is conceptually and then we can go into specific programs from there.

Sure. First of all, thank you for having us, and thank you, Morgan Stanley. Thank you everybody here for joining us. And I'll do my quick disclosure, which I'm sure you guys know we're going to do some forward-looking statements. We spent a good bit of time on our risk factors, so take a look at those in our most recently filed 10-Q. So just take a step back, the company was founded really with the goal of making an important company in the era of cell and gene therapy. And we wanted to go after what we thought were some of the more important challenges. And I'd say there are 2 major platforms inside the company. The one that we're not going to spend much time on today is probably related to cell delivery and you probably recognize, you can do -- or gene delivery, I should say. You can do pretty much anything you want to a genome in a petri dish. The hard part has been doing it inside of a body. And so we really set out with the goal of being able to deliver any payload RNA, DNA protein to any cell in a specific and repeatable way. And every time you do one of those 4 things, you think you open up a whole new category of medicine. And we've got this fusogen platform, which really gives us payload diversity and cell-specific delivery. I'm not sure if you'll help us with some of the other aspects of that. And stay tuned. We can -- we slowed that down a little bit. We couldn't do so many INDs last year, but we've continued to invest in that on the research side, and we're excited about where it's going or optimistic about where it can go. So you brought up the hypoimmune platform. And again, if you take in just a broad step back, the goal in cell therapy that we have is to be able to deliver cells that engraft function and persist, right, and to be able to do that at scale. And the real challenge to date has been related to, first of all, just persistence and most importantly, overcoming the barriers of allogeneic rejection. And that's gotten -- people have approached that problem since the advent of transplant medicine in 1 of 2 ways. One is profound immunosuppression so that the immune system has no ability to find the allogeneic cell, and the other to use autologous cells. And both of those have pretty profound implications on your ability to scale. And so we've kind of embarked on a rather audacious goal of being able to overcome allogeneic rejection. They call it the hypoimmune platform. And so, step back, very simplistically, there are 2 elements of the immune system that we have to grapple with. There's the adaptive immune system of B and T cells. It's actually relatively straightforward to deal with. You knock out in some way, disrupt expression of MHC Class I and Class II. And you won't see T cells recognize these cells and you don't get B cell, you get antibody production. The challenge to that is that viruses in tumors figured that out a long time ago. And so we've evolved something called natural killer cells that recognize these cells. And so that's been what's really set the field back for the last several decades. And our belief is that we found the key to turning off that innate immune system, that's our killer cells and macrophages. And that is over-expression of CD47. And so we make really 3 gene model applications in this hypoimmune platform. It's relatively straightforward. Disruptor, knock out MHC Class I and Class II expression, so you get no -- there's no MHC on the cell surface, not even minor HLAs, and then the other is over-expression of CD47. And we've tested this extensively pre-clinically. I would even, if I wanted to be bold, say, we've solved the problem of allogeneic transplant rejection for mice, for humanized mice, for nonhuman primates. And the real key is understanding how it translates into humans. And so we've shown you a little bit of data in people. It was early this year. It was 4 patients. It was early in the process. What I would say is it's doing everything we hoped it would do. I think there's still a little bit of work to be done to ensure that what we've seen pre-clinically fully translates into humans, but there's a lot of reasons to be optimistic. And if we get this right, we think we can have a pretty big impact. And we've got 4 different drugs in human testing across 3 major areas in 7 diseases, right? So it's type 1 diabetes, B cell-mediated autoimmune disorders like lupus and multiple sclerosis and vasculitis, and then blood cancers. And happy to go into each of those, but those are kind of the big clinical categories that we have, and then we have some things going on earlier stage.

Great. Let's try to unpack all of that in the next 30 minutes. But first, Steve, I have a question for you on the platform, generally speaking. One question we've received is, if the secret to the hypoimmune platform is essentially these 3 edits, what stops another company from being able to make these same 3 edits, and what keeps your approach proprietary to Sana?

What keeps it proprietary to Sana? Well, let's just start with, I can't imagine there's only one way to do this, right? And I think to the extent that there is more than one way people will figure it out over time, right? We -- the most important thing we can do to kind of exploit the advantage of this system for patients is to move quickly and urgently in developing medicines. We do have a very robust intellectual property estate. This has been a challenge people have gone after for years. This is clearly something that's pretty very proprietary to the company where we've built a nice patent portfolio. But I mean I think one of the things you have to recognize over time is that if you build something important, people will try to join the club. So I think the most important thing for us is to make sure it's important, and then we'll work really hard on defending our position over time.

Okay. With that, maybe we can start with type 1 diabetes first? I know that you have potentially a data readout coming out from an investigator-sponsored study there. So let's talk about that program first before we go to autoimmune and oncology. So for type 1 diabetes, you presented nonhuman primate data earlier this year. Could you just unpack that for people just to kind of explain what you thought were the key de-risking events from that data set? And then what are the open questions now when you try to translate that signal into humans? Like what are the leaps of faith that you need to make to bet on a signal in humans?

So, type 1 diabetes is a very simple disease, right? I mean it's very complex biologically. But at the end of the day, what's happened is the immune system has specifically recognized one cell in our body and kill them all, the pancreatic beta cell. So pancreatic beta cell is what makes insulin in a glucose-dependent factor and really keeps all of us in homeostasis. Up until 100 years ago, it was a death sentence for patients. And at that time, there is the advent of porcine-derived insulin. And actually, the first biologics were human insulin. And so the innovation in this space has been important over time. So what we're trying to do is very simply make a replace to cell that's missing, right? And so that is the pancreatic beta cell. And you know it worked, and you know it works because for the last 20-plus years, people have been transplanting cadaveric derived islets in patients with type 1 diabetes. And in many of them, what you find is that they have normal blood glucoses without needing any insulin. It's a pretty profound effect. Two issues though. The first issue is that that's not a very replicable or scalable manufacturing source, right? And the second is that there are only a limited number of patients for whom lifelong immunosuppression is better than lifelong insulin, right? And so others in the field now have subsequently shown that you can take stem cells and make them into a pancreatic beta cell and transplant that. And in the context of immunosuppression, see really, again, profound clinical effects. That is probably almost certainly a more replicable and scalable manufacturing source, but you still have the challenge of the immunosuppression. So really what we're trying to do in the short term is understand, can you get rid of the immunosuppression and overcome allogeneic and autoimmune recognition and killing of those cells, right? Our long-term goal is very simple. It's a gene-modified stem cell-derived beta cell, where with a single treatment, patients are euglycemic, normal blood glucoses, with no insulin and no immunosuppression. So it's kind of like a functional cure. Give them back their normal life and give them back their normal life expectancy and give them back their normal organ function expectancy. So that's the goal. So this -- what we're doing in the short -- what we showed last year or what we put in -- published earlier this year was a nonhuman-primate model of type 1 diabetes, where we induce type 1 diabetes chemically, right? So as a STZ will knock out the pancreatic beta cell and these monkeys had florid diabetes, or this monkey had florid diabetes, difficult to control, but doable with insulin, received after several months a stem cell derived -- sorry, a islet -- gene-modified islet transplant into the muscle of the animal. And the animal was euglycemic, normal blood glucoses off insulin, went from living in a cage in very difficult life to going back into his colony. We followed that animal for 6 months. To ensure that it was our cells that were having the effect, we then gave it a kill -- we have a kill switch, and we killed those cells. They went away and the animal went right back into florid diabetes. So a wonderful proof of concept that this works in nonhuman primates. So we're trying to -- we're basically replicating that experiment in humans with this investigator-sponsored trials with really 3 important differences. So these are leaps of faith probably. One is it's a human instead of a nonhuman primate, right? And so the leap of faith here is that the immunology is pretty similar. The second is that the mechanism for diabetes is different, right? So in the human, it's autoimmune, meaning the immune systems attack the beta cell. In the monkey, we chemically induced it, right? And so you have to believe that we -- that we've overcome the autoimmune component as well as the allogeneic component. We can get into why we believe that. And the third is it's just a lower dose, right? So don't expect to see most likely euglycemia off insulin, right? I mean I think it's -- our -- what we're looking for success is to show that these cells survive with no immunosuppression and function, right? And so detectable C-peptide is the biomarker we're looking for. And for those of you who forgot your basic biology, the beta cell makes proinsulin, and it then secretes it as C-peptide. Cut -- cleaves it as it accretes it and you get C-peptide and insulin. So when you see C-peptide, what you know is that the patient is making their own insulin, right? So that's what we're really looking for is evidence of that, and that would be very clear to us that we overcome and we have overcome the allogeneic and auto immune mechanisms of rejection. So I hope that's helpful. It was a long-winded answer.

No, that's helpful. So cell survival, safety, C-peptide levels, those are kind of your thresholds for, I guess, a viable data readout from the initial IST data set is what you're saying?

I think to the extent that you've shown that you don't have a -- you have cell survival, a cure for type 1 diabetes becomes inevitable, right? I mean, it may not be us that gets there. I sure hope it is, right? But at that point, given the things that we brought up earlier around what you've seen with cadaveric islets, what you've seen with stem cell-derived beta cells, a functional cure, meaning you euglycemia off insulin is -- it's going to happen, right? And then our goal is -- our job is to make sure that we're a part of that.

Got it. Is there any way that there could be a false positive in either the cell survival component of the readout or the C-peptide expression levels of this readout? Is there any reason why if you were to see both of those things that you would still have doubt that these cells are doing what they're supposed to do biologically speaking?

Maybe just a -- so is there any way that this doesn't translate into what we're trying to do and you say, boy, they really haven't overcome allogeneic autoimmune rejection. We've never been able to come up with one. A couple of the readings were false negatives, but you can't come up with one for a false positive. It's really unprecedented to transplant -- It's basically an organ transplant, right? You're transplanting a cell with no immunosuppression and to see cell survival. It's unprecedented. So it would be transformative, I think, for the risk of the company. We like to think it'd be transformative for patients over time as well and what we can deliver for them.

And I know you've mentioned previously that you don't need to see very many patients worth of data to get proof of concept from the IST. So for this initial data set, I know you're not guiding to specifics, but do you think it's a handful of patients at most that we would see data for? One patient?

I think if you saw one, it would be all you need to see if it works. If it doesn't work, we understand why it doesn't work. Some patients don't -- when you get -- some patients who get a primary islet transplant, they just don't agraft, right? That's not -- that just happens and we need to redo it again. It may be that there's immune rejection of our cells, right? I mean that would be, I think, really important. And again, that would be where an end of one would be very instructive, right? We really want to understand that. It's probably a true negative, right? And that's something we need to better understand and discuss with you and others. If it's a true positive, we'll also want to discuss it.

Okay. So I know you mentioned that seeing a patient to get off insulin is not reasonable to expect for this initial data cut. But if you have a patient with cell survival, C-peptide levels, would you then follow them? Would you keep following them in IST to see if they eventually get off insulin?

We'll follow them. And the longer it lasts, the better we all feel about how this will play out over time. So we'll definitely follow them for a long time. And just having a detectable C-peptide is a clinical benefit for patients. They -- historically, that's led to very clearly lower risk of both lower hemoglobin A1Cs and lower risk of severe hypoglycemia. So it isn't no clinical benefit for the patient if they have detectable C-peptide. They're making insulin. You get a basal level of glucose-sensitive insulin secretion, right? That's a really nice thing to have. But it would be unreasonable or unlikely to think that in the first study of a first-in-human experience at the doses we're going to start with a patient -- this person would have a normal blood glucose without insulin. It could happen. Again, my general view is if that's your investment thesis, don't invest.

Fair enough. Great. Fair enough. So let's say that this initial data readout hits and meet your expectations on safety, C-peptide expression. What does that mean for SC451, your internal T1D program?

Yes. So SC451 is a gene modified pluripotent stem cell that we make into a pancreatic islet, right, mostly beta cells. And so the reason -- I'd like to define that. So an islet has alpha, beta, and delta cells in it. They're endocrine cells and you put all of them in there. It's the beta cells that we really care about here. So the -- so to make that product work, we've kind of thought about it, there are 4 major scientific questions to really get through. One, and this is a part that probably proved a bit more challenged than we expected early, but -- and that is to make a gene-modified pluripotent stem cell master cell bank, where you're comfortable with the genome integrity or stability over time. I think we got it. I wouldn't guarantee that to you yet, but we think we got it. That took us a bit of time. The second is to make drug product at a purity potency and yield that is adequate for Phase I testing, right, just to get going. Again, we think we got it. I wouldn't lose a ton of sleep over that if I were in people's shoes. The third is can you overcome allogeneic and autoimmune rejection. That's what we're going to learn, right? So I think that's something that will be majorly de-risking for the program. And then the fourth is can you make drug at a purity potency and yield to be really commercially important, right? And we have a ways to go to be there, to be very clear. And so that's kind of how we thought about it. We haven't really gated any of the investments in those first 2 questions around master cell bank and Phase 1 scale on success of this study. I think we'll feel more -- we're underinvesting right now, I think, realistically and truly the science of scale. This isn't just, we'll put more manufacturing suites together. This is really, it's a scientific question. We need to put more resources towards that. I think that will be easier to do with positive data. And so that's the part of the SC451 program that I think would accelerate with some good data.

Got it. From a regulatory standpoint, thinking a couple of years ahead, assuming SC451 has progressed, it's going through initial studies. What is the best way to think about how much data the FDA might want to see for this kind of a product in this disease space to feel comfortable with the grant, an accelerated approval?

Yes, I have to say we don't have a good answer for that yet. It's probably the best answer to the question. If you're looking at proxies, I mean, one of the things that we have in the field is there's a company that's ahead of us that's doing gene -- I'm sorry, that's doing non-gene modified but pluripotent stem cell derived islets. And I think that that's at least a guidepost to think about. Our program could be very different. We might have different safety things to worry about. We might have different efficacy. But it's a better guide than anything else that's out there right now because there haven't really been many things out there to date.

Fair. Okay. Fair enough. Any final, I guess, thoughts or guidance you provide on type 1 diabetes and the upcoming readout before we switch gears to autoimmune?

Any other thoughts or guidance?

There's been a good amount of focus on how to best...

I would say, we're really optimistic that we can do something important here. And if we get this right biologically, it's going to be, we think, a very important drug, but it really will be dependent -- it's important it will be dependent upon our ability to scale it, right? If we really nail this, it's hard to imagine that our supply isn't the right limiter kind of for success for a prolonged period of time. So we need to really start investing more heavily in getting that part of it, right? But that's one where we really think can be something very important for the company and for patients.

Sorry, final question on this program, and then we'll move on to others. Snapshot of where the IST currently stands. You're enrolling patients. Have you dosed a patient yet? Have you disclosed that?

What?

Have you dosed a patient yet in the IST? Have you disclosed that?

We just told you that we had patients that are on the transplant list, and that was as of early August, and that we weren't going to provide any update until we had data most likely. I mean we don't need people hanging out outside the hospital while we're doing this, trying to figure this thing out.

Fair enough. Okay. That's helpful. And probably a good segue to start talking about your autoimmune efforts then. But before we get into your trial, maybe you could talk a bit about your view of just autoimmune CAR T in general. We've had a couple of data sets come through from other companies now, also from the academic setting. What is your view of what we're seeing from an efficacy and safety standpoint? What's your view of the third-party data sets out there in autoimmune CAR T?

So I think it's been really kind of encouraging, surprising in many regards and almost thrilling to see the clinical benefit that, first of all, George [indiscernible] has found, right, in his single-center experience these B-cell-mediated autoimmune diseases. And it seems like what you get is a control-alt-delete of the B cell repertoire, right? And so when you think about the competitive dynamics, you have to kind of think about that biology. So what's really important in that biology? First is just what cells are you trying to get rid of, right? And so you need to get rid of B cells, memory B cells and tissue resin plasmablast. And the challenge with that is to do all of that, you actually have to have a volume of distribution of your drug that gets into tissue and allows you to get into the tissue humoral centers, right? And so there are a host of things that you could do that just can't do that, right? The second is you have to get rid of it seems like pretty much every single cell. I mean you don't really have to get rid of the pathologic cells, but it's just luck if you don't get rid of every cell. So when you think about that, that means you have the estimate of how many cells are there. There are 3x 10 to the 11 B cells, right? That's kind of the estimate, right? So 300 billion. So it's very difficult to do that and to dose anything that's that potent unless these cells -- unless you have something that [indiscernible] grows, right? So those parts -- I think that's the first thing I think about within the competitive dynamics. The second is that these are complicated patients and giving these drugs is very complicated. So I think this is an area where, in particular, the allogeneic platform can offer a substantial benefit even over -- even more so than in oncology, over autologous CAR T cells, right? Patients have to come off of -- if they want to do an autologous cell. Any type of immunosuppressing, which they're all on because they all have a severe disease or they're not going to take this stuff. Before they can get plasmapheresed, then they put back on what the drug product is made. They have to be taken back off once the drugs are made. So the allogeneic really simplifies that, right? Take them off and you does them. Very simple. It's a lot lower risk for the patient. And the risk the patients as they go through all of these changes is otherwise they'll flare, right? You've seen some examples of that in the field, right? So the second is just scale, right? We can do this at a scale that is very, very different. We'll make hundreds and hundreds of doses per manufacturing. You use a number of 500. That's just a nice round number. Put into context, 100 manufacturing runs of an autologous product is you get the best 100 patients that you can dose. If we do 100 manufacturing runs, that's 50,000, right? And so when you kind of go through the scale of the autoimmune diseases, it really gives us an opportunity both simpler and also just more readily available for patients. We would love to be better as well. I just kind of going to hold off hope on that and see where we end up. When I think about where we are and the data we've disclosed, so that we gave 4 patients with cancer data. You know we deplete B cells, right? We showed you that, right? So you have a question really left around, okay, does that translate into the autoimmune setting? I would argue it's way easier to deplete B cells than deplete B cells and get rid of tumor cells, which is what you have to do right in oncology. And so then the second question then is, how -- what's your clinical benefit long term really look like, right? And I think it's very reasonable to say, well, maybe we won't be as good as some of the other things out there, and we'll have to find our niche, all right, and it's very reasonable to say we have the category killer, right? And the clinical data will prove out which one of those we end up with. But we do have a scale and hopefully a potency and durability that will give us the chance to do that. I would guess that if you look at autoimmune versus oncology, you probably don't need the cells to stick around quite as long, right? Because all you do is trying to get rid of the last cell, and that doesn't matter. And again, it just takes less time most likely to get rid of B cells than to get rid of B cells in every tumor cell, right? Just common sense tells you that.

Sure. I guess on that point, what is the minimum amount of durability you think an allogeneic CAR T needs to show for a disease like lupus?

What is it for?

Minimum, like amount of durability that an allogeneic CAR T therapy would need to show for a disease like lupus to be considered kind of commercially viable in your perspective?

I would like to think that there isn't a big difference between with the allogeneic CAR T cell and the autologous CAR T cell delivers. There is -- it will probably have a commercial place if it's not quite as durable. But the rate limiter in many -- some of the rate limiter for patients right now in terms of market size is the lympho-depleted chemotherapy. So while it's easier to retreat with an allogeneic CAR T cell, it's really simple. It just sits on the shelf and off you go, you would have to live with deplete the patient again, and that's something we'd prefer not to do. So I think we get away with some variability around that. And I'm sure there will be variability across different studies that just isn't even real, right, whether we're better or a little bit better or a little bit worse. I mean I think that is just patient population and luck. But I think you probably want to be pretty similar. I don't think you want to -- we don't really start out trying to be worse.

Right. Fair enough. So coming to your program then, for the data set expected this year, what can people expect in terms of number of patients, amount of follow-up, the number of indications that could be represented?

I can't really answer that except for one thing. And that is that the first patient was dosed around May, right? You know that from clinical trials. So it's not really realistic. I think there are going to be a whole bunch of patients with 1 year follow-up, right? And we're limited in dosing early on because we're going through dose finding, right? And so the most you can do is 1 per month until you find a dose. And I haven't yet been a part of a study where -- in CAR T cells where we're good enough to enroll every 4 weeks, right? Usually, there's a little bit of very -- there's a little bit of time lag in there, just screening, scheduling, lympho depletion, then you dose the CAR T cell and then you wait another 28 days before you start that cycle again. So you're not going to get tons of patients. But our goal is to say, okay, first off, does what we see on B-cell depletion translate into what you see in the autoimmune space. I think that's very low risk. Then you say, okay, does that translate into clinical benefit for a patient? I mean, I think that's an important question. And once you have that, it's actually pretty straightforward, then. You're more or less de-risked until you get a ton of data, right? And the ton of data would be your question is, how do you compare to others in the field, right, in a very competitive space. And that will take a lot more than we can generate this year.

Got it. Okay. Do you think it's reasonable to expect data for indications outside of lupus this year?

From what?

Reasonable to expect data for indications outside of lupus for this year?

Well, I think it's reasonable to expect data. I think that's about as much as I'd be willing to get into, and even that has some variability of timing that's beyond our control. So I think last year, you probably remember, we thought we'd have data at ASH, and we didn't get an ASH abstract. So we put it out in the first week of January. So there are elements of this that we don't control. We'll do our best. I kind of look at this, we're building -- we want investors to come along with us on a journey and building something very valuable over time. And we have to make sure that we have transparency around what we think are the important kind of deriskers along the way. And part of that is always going up being early clinical data to see does it do what you hope it would do. And then it takes a lot of information to say, truly define the drug characteristics after that, right? But we should be in a place, I think, this year, we're going to start to tell you, does is a smell like we think it should.

Okay. Fair enough. We have 3 minutes left. I want to make sure we touch on oncology before we close out. So ARDENT, you reported some initial data from ARDENT, like you mentioned earlier this year. And I think you've guided to more -- like a further data update, I think later this year as well.

The first what?

A further data update for ARDENT later this year as well, right? I guess what are you looking to see from the next update to kind of get conviction that this is a differentiated CD19 therapy? Is it durability? Is it depth of response? Like what do you hope shows that this is differentiated?

Well, to be very clear, our goal has been to do 2 things, right, which is to -- what you've seen in the allogeneic CAR T cell field to date is they get pretty good early responses, right? But they're not very durable generally because the cells don't last post the time of the immune system recovery. Either that or the immune system is really heavily knocked out, and that's come with its own toxicities, right? So our goal would be able to show with kind of normal lympho depletion that we can -- really, our goal is to see durability that is better than what's been seen out in the allogeneic space and really more comparable to autologous, right? And it's available on off-the-shelf and scaled, right? And so it becomes to smell more like a drug and our ability to deliver it for patients. We're not trying to be better than autologous. We would love that. If that happened, we're trying to look the same. So we'll see where that ends up. It's a competitive space. And so we need -- in some of these other places, like if we're not quite as good in the autoimmune space, it's like I think we'll find a place for the drug. It works, it will find a place where it's really important. I think in this space with the establishment, you got to have a higher bar for what you expect from us. So that's kind of -- at least we have a higher bar for what we expect to justify ongoing investments in the space. So that will be what you should be looking for is to say, hey, it's it starting to smell like something that can really compete and be similar to autologous cells or not, right? I mean, I think that's the important element of what we're trying to establish.

Okay. Understood. 30 seconds left. Maybe we can just touch on Sana's cash balance, your current cash balance, the associated runway. And then importantly, how much, I guess, pipeline development is contemplated in that run rate. So like where does your current cash get you for T1D, oncology, autoimmune? How far can you get those programs?

No. Yes, we're going to need more money, I would just say, right? And we're going to need more money in the not-too-distant future, right? When we raised money earlier this year, we kind of -- we told investors we were going to raise more or less as little as we needed to. We want to kick the can down the road to get more data, right? And so we want to kick the can far enough down the road we didn't trip over it, right? But really, we thought the next time we raised money should be with some clinical data that helped to -- you and us to define kind of what is -- what the company's future looks like and what areas and what level of investment we're going to need. So I mean, our -- we have a lot of levers we could pull to extend the runway. But our goal is that we get good clinical data, and we kind of push forward in all these areas, right? It may not be possible, right? We may have to make some choices. I'd like to think they would make those choices based on clinical data, not finances, but cell therapy is expensive, and we live in the real world, we'll figure that out.

Great. And with that, we're actually at time, so we'll go ahead and close out. Thank you, everyone, for joining. Thank you, Steve, for being here. Really appreciate your time.

Thank you, Vikram, and thanks, everybody, in the audience. Appreciate it.