Okay, welcome everyone to our next Fireside Chat. I'm pleased to have Poseida Therapeutics with us today. Specifically, we have Kristin Yarema, CEO and President of Poseida Therapeutics. Kristin, thank you. It's great to be here, Ben. Thank you. I think our typical roadmap is we'll get into Q&A, but maybe you could start with sort of just a high-level overview of Poseida and just highlight near-term events that we should be focused on. Absolutely. Yes. So it's a very busy year for us, and particularly a very busy second half. So we have guided that we will be giving one or more data updates for all three of our clinical stage programs. So we have... This is Cell Therapy Day, so we have three clinical stage allogeneic CAR T programs, two in hematological malignancies, one in solid tumor, and we'll be giving updates on all of those. We will also have a Cell Therapy R&D Day this fall as well. So lots of opportunities for investors to get familiar with Poseida and to see exciting new data coming out. Okay, great. I think one thing that Poseida has always kind of been behind, and that's the sort of non-viral approach to engineering cell and genetic medicine. So I guess, question number one, just at a high level, why take that approach, and what advantage does it, does it provide? Yeah, absolutely. Well, we, we think there are a lot of advantages, and you're right. We have been on the non-viral allogeneic bandwagon for a very, very long time. Poseida has never been afraid to be different, and I think we're really starting to see, you know, the fruits of that play out. What's, I think, fundamental to understand about Poseida is that we have a very different, a unique and proprietary set of tools, and that those underpin everything we do. So we use them ex vivo to produce the allogeneic CAR T. We also use them in vivo in our genetic medicine side, but I think we'll focus on cell therapy today. And among the advantages of those unique tools is that they allow us to produce a CAR T that is really, really rich in T stem cell memory cells. So this is not like other allo CAR T, it's not like other auto CAR T. Really uniquely differentiated in having these stem cell memory cells. And to get that phenotype, you really need to use a non-viral technology. Why is that important? It's important because T stem cell memory cells, or the stemness in your product, has really been shown through more and more literature over time to correlate with depth and durability of response. So if you want the right allogeneic CAR T that's gonna really give you the efficacy profile that you're looking for, you need the T stem cell memory cell. And it also has additional benefits on the safety side, because you can think of T stem cell memory cells as prodrugs. They take a little bit of time to differentiate in the body. What that means is, if you infuse a patient with our CAR T product, you're not giving them a huge bolus of effector cells. Those are the things that kick off side effects, like cytokine release syndrome or neurotoxicity. So we always anticipated that we would see a better safety profile compared to other CAR T technologies by using the T stem cell memory cell, and indeed, we feel that that's really starting to play out. So using this set of tools, we think we're really emerging as a leader in the allogeneic CAR T space, maybe the overall CAR T space, and it's through the strength of these clinical stage programs, our early multiple myeloma data, and we also have our own GMP manufacturing, so we're doing that as well. I think we're getting really recognition and validation of that through a couple key partnerships with large pharma companies that we have, one with Roche and one with Astellas. So it was, you know, it was, it was a different bet. Maybe it was a something of a risky bet in the early days not to do viral technology and not to do autologous CAR T, but we've been very pleased with the platform that it's been setting up for us. Okay, excellent. I think I want to just touch on the last point you just made. You also have been very focused on allogeneic cell therapy. You know, this is one where I think the one question that we often get from investors is, how do you avoid allo rejection? So how do you think about that? Yeah, absolutely. So you know, that is a key question. I think was one of the original challenges maybe facing the whole allogeneic space, right? Will you see rejection of these cells? Will you... You know, will there be alloreactivity? Will they be safe? Will you see graft-versus-host disease? So we're not seeing any of that in our data. How we get there is we really use our proprietary tools to knock out the T-cell receptor, so we edit out the T-cell receptor, and we also do a partial edit on beta-2-microglobulin. So we edit those features out, and that gives us a product that we feel is seeing quite good persistence of the cells in the patients. We've actually observed that. We reported in our ASH data back in 2023, we followed a patient and took samples. This is a multiple myeloma patient that malignancy is in the bone marrow. So we took a bone marrow sample, well over six weeks after the patient was infused with our CAR T, and we found that those cells, the CAR T-cells, were about 15% of all of the mononuclear cells in the bone marrow at that time. So, you know, we're seeing really nice persistence using this, this very, I would call it, a very light gene-editing approach to our cells. Great. Okay. Well, let's dig into the clinical data, starting with the heme malignancy programs, which, of course, you have a collaboration with Roche. Maybe actually, we'll start with this collaboration. What precipitated that collaboration? And maybe just kind of frame for us who does what. Absolutely. So, you know, when Poseida was looking to get into this space, we absolutely knew that the heme malignancy space was gonna be very competitive. It's always been competitive, right? There was a time, and you will remember this, that three of the top 10 drugs, not just in oncology, but in all of pharmaceuticals, were multiple myeloma products, right? So it's a very, very competitive space, and expensive for development. So we thought it would be key to have the right strategic partner to really help us develop and also, you know, commercialize these products. And Roche, you know, is just a fantastic partner because of their scale and capabilities. So, you know, I think undisputed leader in lymphoma, and one of our programs is a lymphoma program. Myeloma is a place where they have great, you know, aspirations to be, and so a lot of commitment to multiple myeloma and interest in the right program. And, you know, and then you have the intangibles. So a very high science company. I think we at Poseida think of ourselves as a high science company, too- Mm-hmm ... so there's a great cultural fit, and the right kind of passion for patients. You know, Roche, very recently, once again, topped a very famous poll of patients and patient advocacy groups as the standout number one company in oncology. So, you know, we are really, really pleased to have that collaboration set up, and it is extensive. So we have two. Roche has licensed two programs, the myeloma program and the lymphoma program. They have options on two preclinical programs that are very rapidly approaching the clinic, and there's also a research component to this collaboration, where they can nominate up to six more research targets. Poseida, you know, runs things from the beginning. So we execute the phase I studies. We eventually transition the licensed programs to Roche to complete full development and commercialization. So, and Poseida does the clinical manufacturing in our own GMP facility. We'll eventually transfer that as well. So it's a really, really nice model because it has enabled us to move the programs very rapidly, to fund the company on the strength of the milestones and payments that we're receiving from Roche. You know, for a cell therapy company, our burn rate is quite low because Roche is, you know, carrying a lot of these development costs, and we're earning these milestones. Just this year alone, we've already collected $45 million in milestones and payments through this collaboration. So I think it's a real win-win for both companies. Okay, that's great. Okay, let's start with the BCMA program, which you call P-BCMA-ALLO1. You've accrued phase I open label data with this, tested a number of different doses, different regimens. I guess maybe start off, summarize those data, then let's maybe get into more detailed questions. Yeah, absolutely. So, you're right. We started. Really, our first presentation of data was last December at ASH 2023, and that was very early data. We had been exploring different doses, different lymphodepletion regimens. We've been looking at that more systematically than anybody else. So that data set, I think the gist of it is that we found really two different dosing regimens that gave great results. One was using 500 mg of cyclophosphamide and fludarabine as a conditioning regimen, the other, 1,000. Both used the same cell dose, so we had an invariant cell dose, and we varied the lymphodepletion. Both of them worked really well. Overall, from those pooled arms, we had a response rate, objective response rate of 82%, so nine of 11 total patients. In the higher of the two conditioning regimen arms, we had 83% response rate, 40% stringent CRs. Everybody was VGPR or better, and we also had patients who had already seen autologous BCMA CAR T, and all of those patients responded. So this was quite remarkable, I think, and something that we'll be, you know, following closely as we move forward. So this was a really heavily pretreated patient population, which I think makes those response rates quite impressive. The other thing to remember is that those response rates that I just articulated for you, Ben, those are on an intent-to-treat basis. So when we think about autologous CAR T data, you're not seeing intent-to-treat objective response rates. You're seeing an as-treated patient population, after all the patients who progressed and couldn't receive therapy dropped out, or the patients whose cells didn't manufacture, sadly, and couldn't benefit from the therapy. So when we're talking about these response rates for CAR T, it really needs to be an apples to apples comparison. And if you, if you do that arithmetic, you know, these rates are, compare very favorably, I would say, to the autologous CAR T, but in a far sicker patient population, who also, a large number of whom have already seen a BCMA-directed agent. And, you know, the label data from those competing products, those patients are all BCMA naive. So Mm-hmm. You know, this was incredibly encouraging. So, of course, the program has been moving forward. And, as I said, we'll give one or more data updates, later this year. Okay. I mean, yeah, it's fascinating that you're seeing efficacy in BCMA-exposed patients. One of the learnings that I took away from CD19 in lymphoma is that that's challenging, to look at a CD19 CAR T in a post-CD19 CAR T setting. Why you... I guess, do you have any hypothesis as to why you're able to see efficacy in this population? Yeah. Yeah, we do. So, you know, I think the first thing to think about is, you know, myeloma and lymphoma are different diseases. So, you know, antigen loss in lymphoma CD19 is, you know, is well-known, up to 40%. And we have a lymphoma program that actually dual targets CD19 and CD20 to address exactly that problem. But myeloma's a little bit different. BCMA is a difficult antigen to altogether lose. You might have downregulation, you might have some mutation, but it's very difficult to lose it. And so I think some of the reasons that we feel we're seeing efficacy are, number one, we are starting with healthy donor starting material. So it's very different from an auto CAR T, where you have these kind of, you know, beat-up T cells that get exhausted and very difficult to maybe reactivate them. We're using fresh T cells that just inherently have more punch, so I think that's the first factor. Secondly, we're using a different binder. So our binder targets a different epitope than, you know, other products that are on the market, whether they're CAR Ts or bispecific antibodies. And I think the third thing is probably just the, you know, the power of the T stem cell memory cell itself. It's just a fundamentally different CAR T in that regard. You know, we're really encouraged by that data. We also presented a poster at AACR this year, where we looked specifically at the subset of patients who had received some kind of prior BCMA-directed agent, whether a CAR T or a bispecific T-cell engager of some type. It was a smaller data set at the time, but we saw 60% of those patients responding well, and that was patients who'd seen BCMA CAR T. One of those patients had seen both a BCMA autologous CAR T and a T-cell engager directed against BCMA. We're now seeing efficacy in patients who failed not just one, but even multiple prior BCMA agents. Again, this is a story that we're watching very closely. We think this is, you know, quite unique data that we're generating, and we're pretty excited about it. You know, going back to something that you said earlier, that stem cell memory T cells, you know, one could kind of think of that as almost like a pro drug, where you're getting, you're getting a T cell phenotype. Rather than just dumping in a bunch of effector cells, you're dumping in... You're, you're giving a, you know, kind of a more nuanced product. Do you see that playing out in the safety profile, in this, in this cohort of patients? Oh, yes, absolutely. So, you know, when we show the efficacy data to, you know, to thought leaders, like we recently had an ad board, for example, at EHA, you know, they're very excited about the efficacy data, you know, and rightly so. You know, we haven't seen this kind of efficacy data from an allo CAR T. You know, we want to see more. But then when we show them the safety data, that's when they get incredibly excited because, you know, we always suspected that there would be differentiation in safety. And sometimes differentiation on safety is hard to show, right? How do you prove something is safer than something else? But it's very difficult to do. But the data that we're seeing, you know, is just pretty clear. So we have seen, first of all, no GVHD, no dose-limiting toxicity, so those are just kind of table stakes. But above and beyond that, our CRS rate is very low. In that ASH data set, it was only 21%, and all of that was Grade 2 or less. Mm. Neurotoxicity, 6%, Grade 2 or less. Infections are, you know, now suddenly in the limelight. There was just a, you know, a study published in the last day or two looking at, at CAR Ts in the real-world setting and showing that about half of patient deaths not due to progression were in fact due to infections. Well, even at the highest levels of lymphodepletion that we've given, infections were quite rare. So it, it's sort of every part of the safety profile is just looking incredibly compelling. And we also aren't seeing the Parkinson's-like symptoms- Mm-hmm. You know, that we've seen in some of the auto CAR Ts. So I think what's really exciting about the safety profile is it opens up the possibility of really, you know, exploiting one of what should be the great advantages of allo CAR T, which is being able to meet patients and physicians where they are. If you have a safe enough CAR T, then you may feel comfortable giving it in the outpatient setting, and we've actually been studying that. Our lead investigator has dosed most of his patients actually completely in the outpatient setting. Oh, wow! Yeah. And so, so we can do that. It's off the shelf, so you might be able to ship it or store it, you know, locally at a wide range of centers. And if you're not concerned about managing, you know, toxicities and side effects, you know, you can really see this not just being a tertiary care, highly restricted, you know, specialty autologous CAR T sort of model, but really having the allogeneic CAR T, you know, move into the community setting. And for myeloma, lymphoma as well, you know, the vast majority of patients, that's actually where they are and where they, you know, begin their treatment. So we see the safety profile, in addition to the ability to manufacture at large scale, and have that off-the-shelf, you know, no wait time feature, as something that will really help unlock the promise of, of CAR T for many, many more patients, and that's really what we want to do, and I know what our partner, Roche, wants to do as well. Okay. Awesome. Well, let's talk about your other program, the CD19/CD20 dual targeting program. How are you positioning this in the clinic? And, excuse me, I guess, why are these the right targets? Yeah, absolutely. So, you know, you alluded to this before. You know, a lot of the patients who are losing response to a CD19-directed CAR T, it's because of the antigen loss. So up to 40% of cases may be just because of this antigen escape, antigen loss of the CD19. So you know, that sets up, "All right, you know, let's use a different target." Well, there really are no two better-studied targets in lymphoma than CD19 and CD20. So because of our toolkit and our technology, we can create a T stem cell memory CAR T cell that carries both full-length CARs. So it's not a tandem, it's not a, you know... it carries two complete CARs, which we think has advantage in being able to, you know, bind antigens and so forth. It's a product that we think could be used for patients at risk of or having exhibited that CD19 antigen escape. But you could also use it, you know, in naive patients as a kind of insurance policy, right? So the T stem cell memory cell, one of its features is it should be self-renewing, it should, you know, it should stick around for a while. So you'll still have that T stem cell memory CAR T in your system. If the antigen starts to escape or you get some, you know, clonal evolution, you know, that CAR T could be ready then to engage cells with the CD20. So we think it's a great combination in lymphoma. It's also a pretty obvious combination in autoimmune disease as well, and, you know, that's, that's something else that we've seen others pursue with that combination of targets in the auto space. You know, this is the first allogeneic, dual CAR T against those targets that's in the clinic. Okay, that's very cool. And then, you did a lot of work, as you talked about, with preconditioning and understanding the kind of the preconditioning, ideal protocol in myeloma. Are there learnings that you can transfer from that, from those investigations, to, to lymphoma? Oh, yeah. Or is that just a different... Yeah. Yeah, 100%. So, and not just the preconditioning, you know, either. So, we have. We have absolutely translated the learnings on preconditioning to our other programs, and so we've, you know, we've started this program at what we suspect will be a more adequate lymphodepletion. So just to answer that. But our system is very similar, so everything about it is the same. We try to take a systems engineering approach to allo CAR T. So our manufacturing process for all the programs is essentially the same. Even the genetic engineering, the transposon cartridge, is exactly the same. We just swap out the CAR Ts. So that makes it very, very easy. We understand our systems very well at the cellular level, at the manufacturing level, and so when we find improvements and we find learnings, you know, in anything that we're doing, it's quite easy to translate them to the new program because there's such a similarity to everything that we do. Excellent. Okay. And so remind us, when did you initiate this study, and when can we expect to see data? Yeah. So we began startup of the study, very late last year. So the study is in the clinic, it is enrolling, and we have guided for a data update, at the end of this year on this program as well. Okay. And is this initial update... will this inform on efficacy, or is that... Does that come later? Well, we'll see. It's a phase I safety study, and it's your typical, you know, three-by-three dose escalation study, so it, you know, does take some time to get going. So, you know, we certainly will be providing safety data. If there is efficacy data that's, you know, mature enough to share, then, you know, we'd be excited to do that, too. But we'll just have to let the data show itself. Okay, fantastic. Last couple of minutes, let's turn to your solid tumor program. So you have a MUC1-C, as it's called, P-MUC1C-ALLO1 MUC1-C CAR T. Maybe start us off. Walk us through the data you've sort of generated in solid tumor so far. Yes, absolutely. So this study, we gave a data update. You know, the last one was a while ago. It was at ESMO in 2022, ESMO IO 2022. It was a very early data cut, but what we saw on the patients that we reported, which is about six patients, we saw a partial response in a breast cancer patient, and we also saw some stable disease in several patients. And this study, I should say, is a basket study in a basket of epithelial cell tumors. That's where MUC1-C, which is part of the dimer of the MUC1-C protein, is expressed. So since then, we've been continuing to enroll across a wide range of epithelial cell tumors, and we have also been refining the lymphodepletion, just like we have in the myeloma program. You know, we've learned some pretty interesting things along the way as well. So at AACR, we said we found that the patients in the solid tumor setting are somewhat more difficult to lymphodeplete than in the heme malignancy setting. They need more, and that's probably because in their patient journeys, the kind of agents that they've seen are generally less lymphodepleting than what you get if you have a heme tumor. So you need a little more then to lymphodeplete them, to create enough space for the CAR T to expand and to do its job. We're pretty interested in what we've been seeing lately in terms of, you know, patient cohorts, looking at different cell doses and with, you know, more adequate lymphodepletion, and we will be excited to give a data update on that program also at the end of this year. Okay. So again, it's all three clinical programs that are going to have, you know, pretty meaty data updates in the second half. Okay, fantastic. And then also, to kind of go back to something you mentioned earlier with respect to autoimmune, I guess why, why, why pursue autoimmune when there's been so many companies that are also pursuing autoimmune? And so, number one, and then two, what is sort of your timeline for that? Yeah, absolutely. So I think the reason why to pursue autoimmune is our platform works in oncology, and, you know, it's, and I'm very excited about autoimmune disease. I've spent a great deal of my career actually in autoimmune disease as well as oncology, so I think there's tremendous potential for cell therapy there. But you want to start with a technology that's proven, and, you know, we've proven it in the challenging setting of oncology. Our platform, I think, is particularly well-suited for autoimmune disease because... Let's just go back to allo and non-viral, kind of where we started, right? So there are very large patient populations in these autoimmune diseases, so a rheumatology practice can have nearly 1,000 patients, some of them. To meet that kind of patient volume need, we really believe you're going to need an allogeneic cell therapy approach. So, you know, we're producing up to 100 doses per batch. We think there's a lot more room to go using this allogeneic approach. So if you want to treat large patient populations, you really need allo. But in autoimmune disease, safety is also a tremendous concern. So you need to have a CAR T that's safe. You want to have something that can be given outpatient, and you want to have something that's non-viral to get you around concerns and questions about secondary malignancy. And I'll tell you from my time running autoimmune businesses, the number one med info request that we always got was, "Give us the data for your product and secondary malignancies. Hmm. So it's top of mind for rheumatologists always. So we just play into those features. Plus, we have an embedded safety switch in all of our programs, so you know, we think this is the winning set of features. We also have assets against targets CD19, CD20, BCMA, CD19 that are well-suited. We've just been trying to figure out the right approach and the right strategy for doing that. But you know, we have said we're actively working on it, and you know, stay tuned for announcements to that effect. Great. Okay, I think we are right at time. But Kristen, thank you so much. This has been a great discussion. Thank you, Ben. Thanks so much for having me, and all the best for this Virtual Cell Therapy Day. Great.