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

Good afternoon, everyone. Thank you so much for joining us. We are really pleased to have Steve Harr, President and CEO of Sana. I'm Salveen Richter, biotechnology analyst at Goldman Sachs. Steve, before we jump into the programs, can you just provide us with a quick overview of your ex vivo hypoimmune and in vivo future platforms and discuss how you're differentiated in this field of cell therapy.

So first of all, thank you for having us. It's great to see you. And thank you for joining us both on the webcast and in this gray and dreary Southern California day. I mean we don't have many of them, so thank you so much for joining us in the room. So I think as you probably know, we'll make a few forward-looking statements, so please do peruse our Q for risk factors and things like that. So just as a way of background, we believe that one of the most important transformations that will happen in medicine over the coming decades with the ability to modulate genes and use cell as medicines. And our goal is to build one of the leading companies of that era. And the -- as Salveen mentioned, we started the company really around 2 platforms, really going after what we think are some of the most important challenges in making the vision a reality of cell and gene therapy. So first, is the hyperimmune platform, what we call. And the goal of that is to hide cells for -- allogeneic cells from immune detection. And so there's been some progress in cell therapy, I think everybody recognizes autologous CAR T cells are a great example of something that's having a meaningful impact, and we should likely see something with HSCs and others going forward. But when challenges have been there, one, there aren't that many cells where you can actually make autologous cells on. They have to exist in suspension. And when you can, it's a complicated and expensive process. It has limited access. So our goal is to make cells from a single source or from aggregation of sources, and make genetic modifications to hide them from immune detection, with transplant, for example, myself into you -- so that you won't reject them. And that will give us the opportunity really to do 2 things: one, build medicines that we think can be at a comparable or better efficacy and safety profile and have scale compared to autologous cells. And they go after cell types that autologous cells can't do. That's probably even bigger. That's what we're doing. So the way we approach this, is if you look at immunology, there are 2 aspects of the immune system we have to deal with. One is the adaptive immune system of B and T cells. It's actually relatively easy to deal with through genetic modifications. And you then have to deal with the innate immune system, natural killer cells, macrophages. And that's actually been really the challenge for the field. And we're really pleased with the progress we've made. And so we do 3 gene modifications. It turns out that's much simpler than we thought it would be. We've shown now in nonhuman primates in over 40 different animals in humanized mice and mice that we can effectively prevent immune recognition immune -- and we have immune innovation. And I would postulate to you that if you were a monkey or a mouse, we'd solve the problem biogenetic rejection. And so the critical question, which I'm sure we got get into, is how do these data translate into humans, right? And we'll know that in a matter of months. And then it will just be how quickly and efficiently can we move that into a number of different therapeutics, which we can go through. So that's the hypoimmune side. We're super optimistic about where that's going to go. If it turns out we have hidden cells from the immune system, it's a pretty straightforward path to making -- you're really having 3 different large categories of medicines, one are, allogeneic CAR T cells for blood cancers. We'll target -- we have 3 different targets. One to be all today at CAR T cells -- second, CAR-T cells for autoimmune disorders. And then a bit harder, but where we've made a lot of progress is stem cell-derived products for type 1 diabetes. So all 3 of those, I think, are areas where we're really pretty -- are very excited. The other platform is the fusogen platform. So the idea of this is potentially look at the field broadly, you can more or less do anything you want to a cell in a petri dish. The challenge has been delivering, right? And there's been some progress in delivery, particularly to the liver, but it's been very difficult to get into other cell types. And so the idea here is to do cell-specific delivery, which is both better from a safety and efficacy perspective, most likely, and also better from a manufacturability. Because the vast minority of cells are going to be your target cells in a scenario, trillions and trillions in our body. And the -- and what we've done is figure out how to package in the gene editing or base editing or prime editing reagents so that you can do gene-specific modification in a cell-specific way. So the first IND around that platform will like -- hopefully be in later this year. I think it will take us into next year to figure out how well is it really working. And if we can do that, we think that, that can offer a lot of promise. I'm sure we'll get into -- that's basically the idea. The way that, that program -- that platform works is, viruses have figured out a long time ago how to get into cells, and we hijack some viral mechanisms to do cell-specific delivery and you have to modify them. And then we've hijacked the entrants of a virus to become a virus-like particle, so it no longer delivers to viral proteins in DNA and RNA, but what we put in there, and then we can scale the manufacturers. That's a little bit how it works.

Great. So Steve, you talked about how we're going to get first proof of concept in a couple of months. And this is really from your lead asset from your hyperimmune platform where we're getting Phase I allogeneic CD19 targeted CAR T data in B-cell cancers. As we think about this data set that you're going to provide us, clearly, the target is validated. So we're really going to be understanding your platform. What are these efficacy measures and safety measures that we'll be able to look at to assess that?

Yes. So a really important question. So what we chose to take a step back. Believing that you can overcome immune rejection of allogeneic cells is a pretty big biology step, right? So into -- what we've decided to do on the gate was you have kind of 4 big risks in drug development, and making platform risk as my platform works, I think it will. Disease biology risk and intercede the important biology, drug development risk. Can I see it in -- actually be able to prove it in people and then commercial risk? Do I really need an unmet need. It was to go and really figure out an area where we could isolate the platform risk. And if we truly did what we thought we did, we'd be able to intercede an important biology. We knew exactly how to prove in humans, and it would be important commercially. So the CD19 CAR T cell -- actually, the beautiful part of this is we can peel back the layers of, I would say, efficacy or proof very quickly. So the most important question, I think, is we have solved allogeneic rejection for nonhuman primates and mice, does that biology translate to people, right? So it turns out that's actually really easy to spin out. So when we make these drugs, let's say, Salveen is the donor. So we take her white blood cells, we isolate T cells, and we gene modify them. And there are 5 gene modification that we do. So -- and then we grow it up and infuse in the patients. We can talk -- we can make a lot of doses from that. But only about half of the cells will have all 5 gene edits, right? So that's a bug, right? But it becomes a feature in understanding our drug because we're going to -- when you give a CAR T cell, you lymphodeplete so you knock out a lot of their immune system, and it comes back over about 2 or 3 weeks, which means that in 1 month if this really works like we think it will. What you'll see is that all of the cells that are not 100% gene edited will be gone. And you'll be left with 100% gene edited cells. And you want to see that maybe over a few months in a few patients and things like that. But if you see that, you know for sure that the data we've seen in animals, which is that we've kind of dealt with immune rejection, translate to humans. That doesn't mean we have a great drug, but it means we have a spectacular platform that's going to be broadly applicable. To really understand the CD19 then is a good drug, which you want to see is take, get patients in the complete response. And the first level of evidence would be that the cells persist for a while. I think the data are very clear. You need, I don't know, 3 to 6 months, these CAR T cells seem to stick around that long and really have that kind of a profound benefit in cancer patients. So the second layer of evidence will be -- first layer of evidence, cell enrichment. And that's a great day. On that day, if we have it, we will change the way we invest internally because we will know the hypoimmune platform is going to translate. The second will be, do you have a high level of complete responses with cellular persistence? I will tell you if you have a great product, there will be people who want to see -- and by the way, we'll know the first soon. The second, we'll begin to figure that out this year. The third is do you have durable complete responses at a high level. That will take us well in the next year to figure out. And different people will probably have different views around what they're willing to accept. But if you have it, what you have is we have a scaled process, right? So this process, I would say the most likely dose, we can make commercially -- doing nothing else to what we do, about 450 doses per manufacturing run today for cancer and about 950 doses per manufacturing run for -- we think will be the autoimmune dose. So it's immediately at a scale that we don't have. It's very derisked. And if it works in CD19, this is -- what we have is we've licensed in -- first of all, we'll probably work in CD19 in autoimmune disorders, right? That will be -- to figure out exactly where and how. But CD19 in cancer, then you get CD19 and autoimmune disorders. And we've licensed in the only validated CD22 CAR, right, where we -- and so it's likely going to work in CD22 because it's not much -- there's no risk, it's just capital execution risk. The biology risk is kind of out. And it's likely going to work in BCMA. We've also licensed in a validated CAR in the autologous setting. So from a single experiment, you can derisk 4 separate drugs, right? And that's a very rare opportunity. It will be daunting for us to execute on that, and it will be expensive to execute on it, to be very clear. We won't be able to do it alone. But that's what we get out of those proof-of-concept studies.

Steve, maybe you can frame for everyone how you would position this drug in the CD19 market if you were to be successful? Like how does it play out versus all the existing therapies, the autologous therapies?

Yes. So first off, I think the data from the autologous CAR T cell space has been really good, both for CD19 and BCMA. And I think you've CD22, what they have is really good. And the companies continue to make progress both in terms of expanding their labels and in being able to manufacture the drug. But what how we will position it will depend upon the data to be frank. But I think what we have is a drug that, first of all, is available today, right? It's not available in 30 days, not in 14 days. You don't have to go find the doctor, doesn't have to find a plasmapheresis site and a swap. I don't -- I think people underestimate the complexity of all of the autologous supply chain, right, from a provider perspective. It's available today. It's off the shelf. And if it looks as good or better, which is what we hope is about autologous as CAR T cell, we are going to want to -- we'll have a couple of things we can do. We can get a broader label. We can go into places like CLL, where it's been very difficult to manufacture the drug. We can go into places like ALL, where it's been difficult to bridge patients through the manufacturing process or time lines. And we go into Non-Hodgkin lymphoma and this is -- it will take time, though. I mean, the companies are generating very good data in the second line, they'll have it in the first line. So these are big markets, and we can -- it's available. I mean, it's just the biggest position will be -- it will be available for our patients, available for patient today right? And it's available everywhere we want to use it. And my hope is that we'll be better and that makes it easier to position. We'd like to be at least the same, but we'll have to see -- the data will tell us.

Let's pivot to your second program from this platform, which is SC451, which is an iPSC-derived islet cell program or asset in type 1 diabetes. And you're going to file an IND this year, but there is an investigator-sponsored trial utilizing your platform to modify islet cells here. So how this program versus your program differ and whether we could look at this program and get some sense of proof of concept for yours?

So by the way, just to be -- I mean, it's the second area. The allogeneic CAR T cell platform will have a lot of INDs, coming in the very near term. So just -- it will be before some of the SC451. Just move on from the allogeneic CAR-T. We have those 4 different near-term opportunities. The next exciting area to take this platform is type 1 diabetes. And type 1 is most simply immune rejection of the patient's pancreatic beta cells, right? So they no longer make insulin. 100 years ago, it was a death sentence. In 1923, there was the invention of insulin and now patients do reasonably well. But they don't do great, right? It's still a real challenge. And surprisingly, if you look at the data, glucose control has gotten worse over the last decade, not better, despite all of the real-time monitoring interventions that exist. And what we know -- so first off, for about the last 15 years, it's been possible to isolate cadaveric islets and transplant them into patients with immunosuppression, and patients can be euglycemic with no insulin but with significant immunosuppression, right, for 5, 10-plus years, right, probably something on genral medicine, other places. Not scalable, difficult to replicate. But hundreds of patients get this treatment per year. there aren't many people for whom lifelong immunosuppression has been in lifelong insulin. We now know from others in the field that you can make a consistent product from stem cells, right? And that, that could have a similar short-term benefit. What we don't know is, can anybody figure out how to overcome immunosuppression, right? So the autoimmune alogeneic rejection of cells. And if you can, you can now put all the pieces together, right? You can make a long-term product at scale where a single injection will leave -- hopefully a patient euglycemic off insulin for years. So our goal of the product, SC451 is to take pluripotent stem cells, gene modify them, grow them up into islets, transplant them simply into the arm and have a patient who can just go about life normally for a long -- a person no longer patient, right, for years and years and years. So what we're doing is that's a hard thing. That will hopefully have an IND next year. That's our goal. Our goal was this year originally, it's just it's a lot of work to genetic stem cells and control the product. And so that's going well now. So then the second -- so what we decided to do is to try to learn the immunology in the short term. So we're taking cadaveric islets -- investigators are. The person dies to get the pancreas to isolate the islets, just like they do for the normal transplant. But they then gene modify them. They put these gene edits in and we will transplant that into a person. And the goal is to see that these cells survive without any immunosuppression. And if they do, you will now have checked all 3 boxes of making a scalable product, right? So a cure for type 1 diabetes moves from being something that's kind of possible to being something that's inevitable. It's completely inevitable and it will happen this decade. We may not get it, right? I mean that's -- there's a lot to do. But it will happen because all of the boxes have been checked. So we're actually quite optimistic about that. It's a CTA, not an IND. It takes place -- the study will take place in Europe. Our goal is to be able to transplant and get the data this year and things are on track for that.

What do you think we could understand from this readout? Like what you'll be looking for in this data set? And what could we understand from that?

So you could get -- it's kind of peeling back the evidence of the onion of evidence, I guess. So from our perspective, the most important question is do the cell survive, right? And you can see that in an MRI on the cells alive. And if you look at a -- if you just transplant allogeneic cells, or if you were to transplant in an autologous with a patient with an -- an islet in a patient with type 1 diabetes that will be dead in about a week, right? And so that the patient the cells. So we see cells surviving at 2 weeks, like our immunology say they're popping the champagne. I asked them just to have a beer and maybe champagne in a month. But -- and so that's really looking for self-survival. The easiest way to see that is the lowest bar is MRI, right? The next better way to -- and I would prefer to see this as -- when a cell makes insulin, it actually makes pro insulin, right? It's a create C-peptide in insulin. The type 1 diabetic has no C-peptide. Those of us who don't have type 1 diabetes will have it. So if you see stable C-peptide, you know you've overcome immune rejection in the cells lift, that would be the second best -- that would be next best. The next best is actually seeing patients not need insulin, right? And we're unlikely in the first dose and in the first patients at a dose so that will happen, but it's not impossible. It doesn't really matter immunologically. And from a learning perspective around what we have, but I think it's more interesting. Maybe it's better for the patient, obviously, if we get them off insulin. So that would be how we think about it, look for innovation by radiography, then look for the biomarker C-peptide, then look for patients not needing insulin. Those are kind of the 3 levels of efficacy.

There's -- Vertex has 2 platforms. Their [indiscernible] platform and their via-site platform with data sets that are starting to read out. How does that -- I guess, how does your program differentiate itself from those 2 platform programs? And then what are the learnings you can take from their data that derisks your program?

Yes. First off, I think they're wonderful companies, and I don't -- I really spend almost no time worrying about them. We do try to learn from them. I always just take the step there are, if we somehow figure this out, then it works. And then we figure out how to scale it, and we can make 100,000 patients of drug per year. And we dose 100,000 people for a decade, and no one ever needs to be retreated again during that time. We will have only treated 20% of the people who need these drugs. So like competition is just -- we need it, right? It will be good for the field. We are approaching problems in different ways, and I think that we can probably learn from each other. First off, just the fact that they're in humans. I think has been really great to see that a stem cell derived. Islet can lead to such great benefits for people. And we do a few things differently. Number one, our -- the goal of our gene modification is completely evade the immune system and not need an immunosuppression or any devices. And so hopefully, that works for people, and they're doing it -- they're approaching it a different way. The second is we put the cells in the body in different places, they're going in through the portal vein into the liver, we're going to go into muscle. More people have been dosed in the liver. I'm sure that's why they're doing it, and that makes sense. There are challenges in the liver: one, injecting to the portal vein is an interventional radiology procedure, it has complications, and it's not easily scalable; two is when these cells are going to deliver, they immediately undergo something called IBMIR, immediate blood mediated immune response. And you lose a very large percentage of the cells. And so that's one of the gene edits that CRISPR is making us try to get around that. We just go in the arm. It doesn't happen there. So it's more scalable going into the arm, it is more -- we think cells will survive better, and the physiology seems to work just as well in patients who have received it with cadaveric highlights. So there's a lot that's different. The edits we make, the cells we have -- I'm sure the process is different, I'm sure the cells are a bit different. We don't really know what each other are doing. So -- but we root for them. We really learn from their successes than their failures.

So you talked about with allogeneic CAR Ts, the CD22 program here, how would you study this target? I guess, where would you position it during -- in your trials? And then...

Where we were.

Where would you position it for first in-person trial.

Very straightforward. I got to know the CD22 target about 10 years ago. And it just seems like the lymphoma, leukemia should be a most -- the hepatitis C, single treatment more or less curable for the vast majority of people. And maybe we're going to get there in myeloma. It's getting more -- and it's pretty clear that at least in some patients, either there's an immune response to CD19 CAR or you have a loss of the CD19 antigen, right? It's a escape mechanism in patients fail. And so we actually -- in a different company, we licensed this drug, this target, and it was kicked out through a series of mergers that was back at the NCI. So this has been -- this CAR has been validated around 100 people, and CD19 failures in lymphoma and leukemia. So we license it for allogeneic in in vivo use. And the positioning will be in people who failed a CD19 therapy, a CAR T cell or maybe a bispecific, right, because those will be important as well. And it's pretty straightforward. That market is just going to keep growing as both CAR T cells and bispecifics grow. And about 2/3 of patients don't have a durable long-term complete response. And they have an average survival of around 5 months at this point if you fail the CD19 CAR, so that's a really big unmet need. So we'll have an off-the-shelf therapy, hopefully, that is very effective, I want to see, for them in that field. So we're just right in that field. So just patients who failed other therapies. Last line, it's very straightforward. There are a lot of patients. There's nothing for them really today. And that actually couldn't be progged to become our first drug that's ever approved, is because it's such a straightforward place to develop the drug. If you were to draw the perfect time line for both CD19 and CD22 today, they end up right on top of each other.

At an R&D Day recently. What were the key points you wanted to get across to the investor community about the company as a whole? And you also talked about the autoimmune disease effort, utilizing CD19, and you mentioned how translatable that is, but we really haven't seen too much on CAR Ts in autoimmune. So help us understand why you see that kind of direct line to translate them.

So I'm going to address that and then I'll come back to the R&D. Because one of the key points we really did want to get across at the R&D Day was the opportunity for the field and then particularly for Sana, in going after autoimmune disorders with the CD19 CAR T cells. So this is something that has been kind of we wanted to do for a long time, right? And it's pretty straightforward. If you just take a big step back, the industry spent about 30 years, trying to figure out where to go and where B-cell depletion drive, leads to a clinical benefit. And there are probably about -- if you look across oncology, and autoimmune disorders, there are over 80 diseases where there's been some evidence of efficacy. And it turns out the CAR T cell is the greatest B-cell depleter man has ever made, right, or human has ever made. And so that's kind of where we want to take it. But the challenge with autologous CAR T cells has been twofold. Number one, every time you move into the new disease, you have to redo your process, right? And you have to prove your regulators that you have a stable and controllable drug. So that makes it hard to move across diseases. And two is just a scale problem. So we have, just from for Phase I studies, plenty of drug on the shelf to figure this out. So actually, the best -- so really -- what really jump-started this was data last August, September that was published in nature from a German group in lupus nephritis. And if you haven't read the paper, it's like every time I read it, I find another thing that's just so spectacular for these patients. And you now have -- these patients are completely off therapy. The longest patients out over 2 years. Every single physiologic parameter is normalized, every single biomarker like double-stranded DNA or ANA antibodies that you use to look at lupus have totally normalized. I don't think you're going to see that this works 100% of people as you go to broaden it. But the data are just really [indiscernible]. So it's worth reading. And that same group has shown that not only does this work in lupus nephritis, but it works now in scleroderma, a giant unmet need and in myositis, and it will likely work at a whole host of other areas. So what we can do because we have the drug product is actually go right into -- maybe there's a small dose finding portion and then run a basket study across multiple indications. And you can then -- as we get signal, really just go into broadening and towards registration type studies depending on what it is. So that's an opportunity we can process near term. And we were very fortunate. It's not why we went and hired them. But we brought in Doug Williams as our Head of R&D. And if you know, Doug. Doug, who is part of the -- he was the Head of Research at Immunex, they developed this [indiscernible] very broadly. He then -- he was Head of R&D at Sedona, then he actually ran R&D at Biogen as they went through both Rituxan and ocrelizumab and other -- someone who's steeped in immunology understands the drug development paradigm here. I think you can actually move very quickly. And because of the manufacturing -- instead of you asking us how we're going to position this versus autologous, we'll be able to beat them to the market potentially in some areas, at least. So we're really excited about that opportunity. It's going to be harder than what the Germans have shown it today because I can imagine that it's such a perfect drug, and we'll have to figure out the right patient populations and things. But it's one where -- there's no reason to believe that CAR T cells can't be as big in autoimmune disease and as transformative for autoimmune disease patients as they have been in hematologic malignancies. In fact, they'll probably be bigger. So that was one thing we wanted to get across. We don't have a lot of time left, so you figure out where you want to go next.

I just want to touch base on the fusogen program and platform and where we will see that first proof of concept. I know you've incorporated a second-generation manufacturing approach to increase potency, but help us understand how you derisk that platform.

So fusogen are super exciting. And again, you can do gene-specific modifications with cell-specific delivery, right? And the first program in humans, we'll be looking to develop -- sorry, to deliver the DNA to make a CAR into in a CD8-positive T cell. We'll hopefully have an IND later this year. Last year, we took a step back because we had a -- you mentioned improved manufacturing process. We improved yield, purity and potency, right? It's about 55x improvement. Just to put that into context, we put more medicine now into 7 CCs and we used to put into a leader, right? So it's decade. Like it's a true game changer from what we have as a drug product. And it turned out that as we did that, we got 2 or 3x improvements as well and how well it worked in T cells, right? And so it became to something we needed to do. So first program will be in T cells. We'll figure it out as we move through next year, how well that works. But what I really like is do things like HSCs, where you can deliver the gene editing or base editing or whatever machinery directly to hematopoietic stem cells. We've shown we can do both. It will take a few years to turn that into a drug that it's ready for human testing. It's a process, right, to really do this. But that's one where -- with proof of concept, you can see a single shot, no lymphodepletion, an opportunity to really go after a whole host of areas. First, where we already know that fixing the genome works, and then we'll go after other areas, that's coming at us quickly.

Great. Well, with that, thank you so much, Steve, really appreciate it.

Thank you. Really appreciate it. Appreciate the time, the audience as well.