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

Yigal Nochomovitz, and it's my pleasure to be hosting the autoimmune cell therapies disorders, panels. So today, we're hosting Steve Harr, the CEO of Sana. We're hosting Rachel Haurwitz, CEO of Caribou. And for Fate, we have Bob Valamehr, did I pronounce that correctly, who is the incoming CEO for Fate. So I believe this is his first panel. So we're very honored. So thank you all for being here. It's a pleasure to host you for this panel.

Thank you for having us.

Yes, thank you.

All right. So I'm just going to jump into a couple of panel level questions. I'd love your thoughts on all of them, and then we'll switch over on the second half to a couple of company-specific questions.

So just -- let's just level set for everyone. What is a high-level overview of your approach to allogeneic autoimmune cell therapy. Briefly, how is your asset designed, what it differentiates it from the space? And what indications are you currently focused on?

You want me to start. Well, thank you for having us. Sam and Yigal, and thank you to the audience for joining us. And I'm sure as you -- I'll say it for all of us. I think we're all going to make some forward-looking statements. So take a look at the Sana, Fate and Caribou 10-Qs and 10-Ks for all your risk factors. So -- this is Steve, by the way, from Sana for those who are listening. And -- so the company was founded really around the idea of trying to go after a couple of the major challenges of cell and gene therapy. And within cell therapy, our goal was to tackle the challenge of allogeneic rejection. And I think the -- if you take a step back, there are 2 important arms of the immune system. There is the adaptive immune system of B and T cells. We've all kind of learned about that with vaccines and other things. It's actually relatively easy to grapple with in the context of allogeneic rejection, you knock out MHC Class I and Class II. Unfortunately or fortunately, viruses and cancers figured that out a while ago. And so we evolved something called the innate immune system. And things like natural killer cells will kill cells that lack MHC Class I and Class II. And that's been the biggest challenge that fields had to grapple with now for 15 or 20 years. And the company's insight and hopefully, advantage is that we believe we've come up with a system that allows us to overcome both allogeneic, I should say, both innate and adaptive immune rejection. And therefore allow allogeneic transplant without immune recognition. And I'm quite confident telling you that we've solved the problem for monkeys. We've solved the problems for mice. We've solved the problem for humanized mice. And we're in the context of understanding how well we've solved that for humans. And so we brought forward a couple of different ways to go after this challenge. One and probably the area that has the most external interest in where we've put a lot of our effort is in gene modifying stem cell-derived therapies. The stem cell derived cells. And in this case, in particular, pancreatic islet cells for type 1 diabetes. And there the challenge you have to overcome with allogeneic and autoimmune rejection. And we'll tell you where we are soon on that. And that's -- I think you'd asked me to get in the allogeneic CAR T space. The other is an allogeneic CAR T cells. And so in that context, our goal has been to really go after and create a scaled supply of T cells, CAR T cells that we can then hopefully dose in both the autoimmune and cancer setting. We have chosen over the last several months to really focus much of that effort in the autoimmune space. I think one of the really great things about the last couple of years is we've begun to really understand the power of cellular therapy in the space. I think that's attracted all of us. And I started in medicine where we gave people a sledgehammer of something like chemotherapy and trying to knock out their immune system as part of treating their autoimmune disorder. And often the drug was as toxic as the disease. And starting in the late 1990s and early [ outs ], we had drugs that targeted singular pathways, TNF and things like that, that allowed us to hold the disease at bay, but they still were pretty chronic diseases where patients have had substantial long-term consequences and sequela as well as being pretty financially toxic. And the real excitement that I think we've all felt over the last couple of years is this opportunity of cell-based therapies to really go after patients with curative intent. I think you'll hear from all of us that we're optimistic that there are multiple ways to crack that egg. I think we have one of them. I can't imagine we have the only way to go about this. And I'm pretty sure there will be room for a number of us to treat different patients in different diseases over time. So the differentiation for us is we knock out a couple of genes MHC Class I and Class II, we overexpressed something called CD47. That seems to allow us to overcome allogeneic rejection. And for us to make allogeneic cells at scale that act and behave and look from an etiologic perspective like autologous cells. And that's our goal. So that's a little bit of an answer.

Rachel?

Sam and Yigal. Huge thank you for the opportunity to be here today. At Caribou, our initial focus has actually been on oncology, and we're today running Phase I studies for 3 off-the-shelf CAR T cell therapies in hematologic malignancies. The first of those programs, CB-010, is an off-the-shelf CAR-T targeting CD19. We've dosed more than 50 patients in that study, continuing that development is we're very encouraged by the efficacy and safety we've seen so far and actually hope to initiate a pivotal trial in oncology by the end of next year with that program. All of that has really created significant enthusiasm for us for its potential in autoimmune diseases. And so earlier this year, the FDA gave us the green light to bring CB-010 into lupus, both lupus nephritis and extrarenal lupus. And we are in the final stages of getting that trial initiated and up and running. CB-010, as I mentioned, is an off-the-shelf CAR-T targets CD19. And really, its key differentiator is that we've also knocked out the gene that codes for PD-1. And really the goal is to take the brakes off the CAR T cell itself to prevent premature CAR T cell exhaustion and allow it to have really meaningful cell-killing activity. I think we really are beginning to understand its impact in the clinical setting in oncology. And some of our oncology data from a translational perspective, and happy to dig into this in a bit, Sam, really, I think, helps derisk CB-010's potential in lupus. So excited to hit the ground running there.

So thank you for the invitation. So Scott here -- I will say whatever I want, I blame Scott for it. But I'm Bob Valamehr. There's a transition and I'll be CEO in -- starting January 1. And again, thank you for the invitation. It's great to be up here with pioneers of the field. How we're different is that we use induced pluripotent stem cells. So this is something that Fate has been working on for the past 15 years. And if you add the years that our founders have been working on it is multiple decades. And these cells are very unique. They represent cells that can become any other cell type found in the body. And our main focus for the past 10 years has been to make NK and T cells. And so that's the focus. We want to eliminate the target cells. And we put a lot of genetic edits into it to have functional persistence, multiantigen targeting, and we can make a ton of these cells. When we make a master cell bank of a uniformly engineered iPSC that's been derived from a single cell. That's the starting point for making uniform consistent batches every single time and with the master cell bank being up in the hundreds of vials and working cell banks that can come off being hundreds of vials from individual vials of the master cell bank. My children's children don't need to worry about making another master cell banks. And that's how much -- that's really -- I don't want to say unlimited because nothing is ever unlimited, but it's just like any other biologic. I worked at Amgen we would make master cell banks for erythropoietin producing CHO cells and other things. And that master cell bank was supposed to serve for decades in a commercial setting. And this is exactly the paradigm. So we can make a master cell bank that is uniform, consistent. Our release criteria is not somewhere between 20% and 80% CAR positive, if that's the engineer. It's 99%. You got to be 100% of that product or something is wrong because the starting material was 100% of that edit. So uniform, consistent is cheap based on scale. We're down to $3,000 per dose. And I'm very confident that once we go to 1,000-liter volume and 10,000 liter volume, our master cell bank can support that, it will be even less than that in terms of cost. So it's truly off-the-shelf, truly cost effective. And with the engineering ability, we can essentially do anything we want. As we heard, there are multiple edits that can be incorporated to make the cells perform better. We have a multitude of ideas that we've incorporated in our solid tumor paradigm. We have 7 edited CAR T cell that really goes to the heart of overcoming the tumor microenvironment is ability to suppress the T cell function. And in autoimmune, we just have entered it with a very unique CAR T cell. And this is, again, off-the-shelf here, 99% expression of CAR that is basically at a molecular level, there is no TCR expression, so you'll never get GvHD. You don't need to do anything mechanical anything additional. So the product purity and the ability to do multiple things to it, it really makes us different than most other traditional platforms.

Okay. Great. some of the competitors, obviously, in the space are the autologous products as well as the bispecifics. How do you see the allo products fitting into that to those 2.

I mean I'll take a stab at the first piece at least. I think as you look at this patient population, an off-the-shelf strategy is incredibly compelling. And for different reasons and in oncology, right? In oncology, speed to therapy is critical. And I think that's where the allogeneic approach has an advantage over auto. I'm not sure that the same is true in say lupus, but there's a very different reason why off-the-shelf is so valuable in the autoimmune setting, and it has to do with what patients have to undergo to actually undergo apheresis. So undergo removal of their T cells to make a CAR T product. If you're a lupus patient, you have to go down or off virtually all of your meds, if not all of your meds for a period of time before you can undergo apheresis. Then maybe you go back on some of them and then go off again before you get your dose. Going off or down twice is untenable for a huge fraction of these patients. And obviously, in the allo setting you never have to worry about undergoing apheresis. You can maintain your meds for a longer period of time and reduce the risk of flare right before receiving a cell therapy. And so we certainly believe that for this patient population an off-the-shelf strategy is mission critical for reaching a broader swath of patients.

Okay. Bob, you want to add anything? Go ahead.

I'll add. So essentially, we want to be a biologic. The value points of a T cell engager is, as mentioned, off the shelf. We want to really go into the community hospital setting. We believe that you should not -- just because you live next to MSK doesn't mean that's the reason you get treated. So our focus is to really go into the community setting. And I think this is why we're pioneering and being bold enough to go into situations where we're either giving our products without conditioning chemotherapy or we're combining it with maintenance therapy because we believe we have unique attributes. And so we want to really understand that. And if those things need to be improved, we'll just further engineer it. But the notion here is we're going to be like a small molecule. We're going to be at a pharmacy available. And when you come in and you need that indication. If the physician believes he should be treated with it is right there, is available, and there is no uniqueness to it. You just thaw the cells and put it into the vein.

Go ahead.

Okay. Just a couple of things. One, this is an incredibly diverse number of diseases that we're talking about, and most of them aren't diseases, they're syndromes, right? And so they're very heterogeneous, and it's almost certain that different patients will want to have -- will require different modalities and different targets over time. And the second is that patients are at different courses of their disease in their lifespan, and they're going to have different goals. And so I think it's behooves us to be patient to see how the field plays out. I do think that there's a math problem that you have to think a little bit about. And so what's so unique about these most recent cellular therapy data is the ability for someone to wake up and treat for curative intent. And to have people in this new world of this drug-free remission is not really a word we usually talked about or a phrase we thought about 2 or 3 years ago. And the unique aspect of T cells are 2 things. Maybe just to take a step back, our B cells, that's the problem that we have around 200 billion B cells in our body. And they're all over the place, right? And in fact, the real problem tends to be tissue resident plasmablast, right? So these aren't things in your blood. So you have -- in our blood, we have maybe 1 billion to 5 billion B cells, right? So it's a single-digit percentage of the overall B-cell repertoire. And so the math problem is you have to have a modality that will grow and kill all of those B cells so that you can hit the control-alt-delete, and reset the B cell. And you have to have a volume of distribution that allows you to get into all of these tissues, right? So if you're giving a -- the beautiful thing about T cells as they divide and they expand log rhythmically, right? So you can give hundreds of millions of them and have the impact of killing maybe hundreds of billions of cells is what we've seen in oncology, right? And that is very difficult for any other modality. And the other is they go everywhere. T cells really go almost everywhere in your body. And most of these other modalities have difficult times getting out of blood, lymph node, spleen and bone marrow. And so that's a real benefit of these therapies over time, and they allow us to go after that. So I'm very optimistic that T cells allow you to get that control-alt-delete of an immune reset and potentially curative intent. We don't know about these other modalities. We'll have to see. I do know immunology has humbled me or humiliated me multiple times. And so let's see what the clinical data really show. But I think you can be very confident that when people are utilizing T cells that there will be multiple approaches that will lead to very, very profound benefits for patients. And hopefully, all of us have one of us.

All right. We're halfway through our time. So I'm actually going to switch over to some company-specific questions, make sure we give you each a little bit more time to speak on your own platform. So Steve, if you don't mind, I'd like to start with you. So you've been highlighting more recently, I believe, your fusogen platform for generating in vivo CAR Ts a little bit more frequently than you had maybe say, in the past recently. So do you think that autoimmune indications are well suited for in vivo CAR T? And if so, when could we see you bring that fusogen platform forward.

Like many things, it depends, right? The thing that is -- I think that the aspect of these CAR-T therapies that the field has been grappling with has been both consistency of results, but more importantly, for what patients is it appropriate to give the lymphodepleting chemotherapy, right? And that is not going to be everybody at all times. One of the most challenging, but also beautiful aspects of an in vivo CAR T cell program is that you cannot give lymphodepleting chemotherapy, right? Because you have to have T cells to transduce and make into a CAR T cell in the body. So to the extent that it works, it's kind of the killer app, right? You're going to have this ability with a singular therapy to potentially generate a therapeutic, you own -- these patients' T cells that will lead to curative intent with really no chemotherapy at all. It's implausible to give it. On the flip side of that is that's a biologic challenge, right? So you have to be able to transduce these cells in a way in a high enough efficiency and get the CAR to be expressed without activation and then get them to grow without the cytokine support of lymphodepletion and have them persist. I think that is much easier in autoimmune diseases than it is in oncology. It's easy to kill all your B cells then to kill all your B cells and all your cancer cells, right? And it's hard to imagine you can kill cancer cells before B cells. So I do think it's possible. I think it's very -- something that is doable. We have done -- we have an in vivo CAR T program, and we did a GLP tox study. We finished in about a year ago. And we saw very high levels of delivery to T cells. We made CAR T cells. We saw no off-targets really, and it was very safe. It was targeting human CD19 so we didn't know the pharmacodynamic effect on a nonhuman primate, but I think we can very safely deliver high levels into T cells. What we're going back and doing is ensuring we can get profound and deep B-cell depletion with a CAR-T that targets the nonhuman primate B cell. And I think we've learned a lot in the last year, you're going to see some stuff soon. I'm quite optimistic. This is something we can turn into a therapeutic. Let's see, maybe relatively soon that happens. That's always longer than you think, though, we'll take a couple of things. One, a GMP manufacturing campaign and things like that. And so it's not going to be 2025 when we're in humans. And the second is my guess is dose escalation will take time here. It's just, well, it's a novel modality, and it will be challenging. But it is kind of the killer app, the in vivo delivery. It should work and I'm optimistic that we're able to give very cell-specific delivery. But we'll have to ensure that we get the activation, expansion and persistence is adequate to get the immune reset without any activation of lymphodepletion. And then that's the challenge and the real upside.

Yes. No, looking forward to seeing that piece of the autoimmune cell therapy landscape really, really progress and your data in particular. But I'm going to switch gears a little bit more. I want to talk about your type 1 diabetes program. It's a sort of different than everything else here with Caribou and Fate, but a big focus for investors, right?

It is an autoimmune disease. So it's on point.

It is. It's just not lupus, yes. And so there's been a lot of focus on this. You've had a bunch of donors come in for your IST that for one reason or another that you've told us the last time we chatted weren't the right fit. So I guess, can you share an update on where the IST stands? And when can we see -- what's your latest thinking of when we'll see that data?

So I think everybody here knows that type 1 diabetes is just an autoimmune disease where the patient's immune system attacks and kills the beta cell, right? And so the patient is no longer able to make insulin at all, right? And so one of the really kind of advances of the last 20 years has been cadaveric islet transplants. And in the context of immunosuppression, it's been curative for patients. Now some patients are up 15-plus years. The challenge is there aren't that many patients for whom lifelong immunosuppression is better than a lifelong insulin. And it's not a very scalable or applicable supply source. So we've seen others now, several groups make stem cell-derived therapies. And in the context of immunosuppression, you'll see, again, really nice glucose control off of all insulin. But again, there's that challenge of really, there aren't that many patients who want to take lifelong immunosuppression. So what we're doing right now is we're doing a study where we're gene-modifying cadaverously derived islets, and we will transplant them with the goal of seeing just the cells survive and function. And it doesn't take that long to understand if it works or not because they should be rejected within a matter of a few days, right? There is a preexisting immune response to these cells. And so it's taken us some time to get this study going. First, we had to get it through regulators, and we had to find the recipients. Then we've had to wait for someone to die who has a really high-quality pancreas and who will donate it and -- or family will donate it and then manufacture the drug. We've had a manufacturing failure we've told the world about. So stay tuned. I do think we'll have data soon. It's inevitable that this will happen. If it works, I think, a cure for type 1 diabetes becomes inevitable. Our goal is a gene-modified stem cell-derived islet program where we can make this at scale. It may not be us that does it. I think we're well on the track and we'd be in the right position to do it, but we'll have some execution really to get done to make that happen. So those data will happen. And if they work, I think, they could be really encouraging to a lot of patients into the cell therapy field more broadly around what we can accomplish going forward.

Got it. And last question for me for you, Steve. Should we expect your SC291 data, which is kind of the topic that we started with at the start of this panel around the same time as the type 1 diabetes data? Or could that -- is that not tied at all?

Yes. Who knows?

Good answer.

I certainly haven't thought that far ahead. I think it will depend a little bit I think many of you know in this field, it's pretty capital-intensive. And we will need more capital over time. We have money into 2026. We have to raise money in the next, I don't know, 6 months or so or do a partnership or something. And to the extent we have data, we may put more out there. So what we have, it may not be the right time to do that. We'll see, perhaps not really thought that far ahead. We want to make sure we have 1 set of data to share with you before we think about 2. But I'm optimistic. We'll get you both sets of data in the not-too-distant future.

Sounds perfect. All right. Yigal, I'm going to turn it over to you.

Okay. Rachel, you started talking about CB-010 at a high level with the construct. But could you go into a little more detail now in terms of what you've learned from the experience with the dose escalation in ANTLER. What you've learned in terms of genotyping with the HLA strategy and how you're pursuing that in the Phase III to increase the probability of success?

Yes, absolutely. So we, as I mentioned, have dosed more than about 50 patients in that Phase I trial at this point. And we shared quite a bit of data on the first 46 patients at ASCO this year. And what we shared was a really interesting retrospective analysis of those 46 patients. And what we did was actually subdivide them by the relative level of HLA allele matching between the patient and their donor. This is, of course, all by chance because it was just randomly which dose was shipped to which site for which patient. And what we saw is that if patients and donors share a minimum of 4 alleles. Now this is 4 out of 12. So it's a pretty modest match. But a minimum of 4 alleles, they had significantly better outcomes and significantly more durable outcomes than the other patients on study. And in fact, that subset of patients demonstrated a PFS at the time of our data cutoff that really meaningfully rivals that of Yescarta in the ZUMA-7 study, which is the trial that led to the approval of Yescarta in the second line setting. And really, that's our objective for CB-010 is to develop an off-the-shelf CAR-T that can meaningfully rival the autologous CAR-Ts. Now I think the biology here is probably not rocket science. If you think about hematopoietic stem cell transplant, solid organ transplant. The importance of HLA matching has been well understood for decades. But we're talking about a very different scale, right? For a given transplant site, you might need 10 out of 12 matches or 12 out of 12 matches to be considered a good donor. We're talking about 4 out of 12 driving these really meaningful outcomes. So we're very encouraged by these data. And we've now created a new 20-patient confirmatory cohort that we're actively enrolling. We are explicitly providing patients a matched product that has at least that 4 match. And we're doing that not only in this cohort, but actually in a second cohort, of patients who have relapsed after prior CD19 targeted therapies and in our autoimmune study with CB-010. So we're all in on this strategy. and look forward to sharing some of those data in the first half of next year. Though the N is small in the initial data that we shared, it was a cohort of about 11 patients. The PK data also really supported this observation namely the greater the number of matches, the better the expansion and proliferation, which certainly matches our hypothesis here. So looking forward to sharing those data next year.

And then with regard to the Phase III design, maybe talk a little bit more about the comparator arm you spoke with the FDA about what the right way to do that is. Can you just elaborate a little bit?

Yes, absolutely. So about -- almost exactly a year ago. We spoke with the FDA, and they agreed that a control arm where patients would receive auto stem cell transplant would be acceptable to them. Now we're continuing to cook our Phase I study. And so we certainly have additional engagements planned with them to finalize the key details of that trial. So certainly, stay tuned as we continue to have those conversations.

Okay. And then could you touch a little bit on the second and third programs in myeloma and AML. What the targets are, what the constructs are, whether you need this HLA matching strategy potentially there as well or you may not dispense with it?

Yes, absolutely. So CB-011 is our multi-myeloma asset, targeting BCMA. We recently disclosed that we've seen encouraging initial efficacy in this Phase I study. And we plan to provide our first fulsome data update in the first half of next year, committing to at least 15 patients at efficacious dose levels. One of the nice things about our trial design is it's quite flexible. And so where we see efficacy, we can backfill additional patients at the same time as we continue to dose escalate. So we're actually enrolling patients in multiple dose levels, multiple cohorts today. We're also exploring multiple lymphodepletion regimens, and we've found success with a modestly deeper lymphodepletion. So very encouraged by the initial efficacy that we're seeing there. In this program and in the AML program, we use our immune cloaking editing strategy, which actually gets rid of all endogenous Class I. So now we have Class II, which is part of the puzzle, but it's as if we have matched the 6 alleles for Class I. So we're very carefully mining our data in CB-011 and CB-012, the AML program to see whether additional Class II matching will be additive for those programs or not. CB-012 as I teased, is our AML program, and it targets CLL1. As far as we know, CB-012 is the only allo CAR T in development against this particular target. We exclusively in-licensed the binder for Memorial Sloan Kettering. And they and we got excited about CLL1 because it's highly expressed on AML tumor cells, but not expressed on healthy hematopoietic stem cells. It's a really important discrimination as far as we're concerned. Because if you're successful with an active CAR T, you only blow up the tumor and not also the hematopoietic compartment. Said another way, if you go after something that shares the antigen with healthy HSCs, you necessitate a transplant, whereas this target gives us the potential for a therapy with a disease-modifying activity on its own. We're continuing to execute on that trial, and we're enrolling patients at dose level 3, no DLTs to date. So stay tuned for updates there.

Okay. And last but not least, let's move over to Fate and talk about the strategy there. You also had a product which was being developed in oncology, which you're now also leveraging in lupus. So can you talk about 819, what the edits are and what the strategy is in lupus?

Sure. As you mentioned, FT819 is a CAR T cell derived from iPSCs. We have 2 very unique edits. One is taking a novel CAR, chimeric antigen receptor and putting it into the track locus. So here, the control of the expression is regulated by endogenous gene as opposed to a super strong synthetic promoter that would drive exhaustion here. The gene expression is more accommodating to T cell activity as opposed to overstimulation. And by doing that, in a biolytic manner disrupting the track locus as you also eliminate TCR expression on the surface. So as we've talked about, not having TCR expression in an allogeneic CAR-T product is very important because if you have TCR, you're going to elicit GvHD, graft versus host disease. So here, most companies, what they have to do is they have to not only knock out because they're doing it in a pool setting, knock out TCR expression, but also eliminate the cells that did not get genetically edited. So here because our starting material is genetically disrupting track expression. We have no other issue when we create T cells. They are pure for CAR expression and completely negative for TCR expression. So this makes it a perfect allogeneic CAR T cell. Now the product, first, we started in aggressive lymphoma. And we saw a very unique profile there. We saw what traditionally seen with primary derived CAR T cells, PK curve that peaks in the second week and goes away and the curve goes up, the maximum PK goes up with dose. And we saw activity in aggressive lymphoma. So we're excited there. But as we see, you really need today CR profile, a complete response profile up in the '90s to have a product that could go all the way. Here, we thought that perhaps the potency of 819 was more relevant for a disease that has lower disease burden because we see very good data in follicular lymphoma. So we believe follicular lymphoma, the number of disease cells in follicular lymphoma is much equivalent to the number of disease cells in, say, SLE. We're calculating there's a magnitude of somewhere between 10 to 8, 10 to 9 disease cells in each study. And because we have such a unique profile being very safe, having no really adverse effects that are problematic, no [ icant ], no GvHD greater than -- or no GvHD period. We basically thought this would be a pretty good transition to go from oncology to autoimmune disease because the effective target ratio is in our favor, and we should get really good results in autoimmune disease such as SLE.

And now you have some data, which you've presented, and you also have in SLE and you also have some data in oncology, which provided some good insights into the B cell activity. So can you maybe just walk through that? And then you're also going to have some additional data at ASH if I'm not mistaken.

Yes. So for oncology program, as Steve mentioned earlier, we see that our T cells can traffic into primary, secondary and tertiary tissue. So in oncology, we saw that when patients have B cells it was quickly eliminated and cells -- diseased cells in different tissues were also eliminated. So we had good trafficking and we had good potency. So we really learned there in oncology that the cells can traffic and have activity. And they had a -- with combination with conditioning chemotherapy that had a PK curve that resulted in activity and response. And so then when we transition to SLE, we saw a very similar case where we saw B-cells be depleted. There was a period of immune reset. And so when we had B cells come back in patient 1, we saw really good profile of what was more of a naive nature. So this -- the first patient had 20 cells per microliter of B cells in a disease setting at baseline. It went down to 0, and it came back to 100 cells per microliter, which is of a normal range, and we were at first concerned that, oh, there's relapse. But the population that really came back to a healthy number was actually of the preferred. So really, what we've learned in oncology, we were starting to see -- we're starting to see, at least in patient 1 in SLE. So a nice transition. We've already -- we've dosed patients 2 and 3, which we will highlight at ASH. And also, we've dosed the first patient in Regimen B, which is a combination with maintenance therapy. So they're basically in combination with maintenance therapy, whether it's methotrexate or MMF. We want to see whether 819 can synergize with maintenance therapy and help the patient drive a deeper response.

Okay. And I gave Rachel a chance to talk about some nonautoimmune topics with AML and myeloma. So it's your turn, you do also have the 825, which can you just very quickly just comment on the HER2 cell therapy, and that one's partnership.

Yes. And I should mention that 819 in autoimmune is in partnership with CIRM. 825 is in partnership with Ono Pharmaceutical. And here, they gave us HER2 binder that was very unique. Most HER2 targeting strategies, whether it's Herceptin and HER2 target the wild-type HER2. Here, we're able to not only target the wild type HER2, but we also can target the p95 and also misfolded HER2. So it really has a unique binding profile. That's associated more with disease. Because we know that solid tumors are very brutal environment to go into and have activity. We've created a 7 edited CAR T product. So these have 7 edits that really try to overcome tumor heterogeneity, the immunosuppressive environment of a solid tumor and the trafficking -- enhancing trafficking capabilities of a CAR T cell to go into the tumor itself. And so this multiplex CAR T-cell, multiplex engineered CAR T cell. It's really exciting for us because not only we can go after HER2 uniquely, we can also go into the tumor, reside in it and in combination with MAP, not only target HER2, but also target EGFR and other targets. So it really goes after tumor heterogeneity and starts to overcome all the challenges we're starting to see with solid tumor.

Okay. And now lightning around just go down the line, let's just do 10 seconds, 15 seconds. What are the key data points to watch for each of your companies over the next 2 to 3 quarters? Bob?

So 819 in autoimmune and 825 in solid tumor.

Three data readouts in the first half of next year, CD19 second-line large B-cell lymphoma, CD19 relapse and multiple myeloma.

So what was the question?

Data readouts for you in the next 2 to 3 quarters? What do we need to watch.

What was that?

Data readouts? Do we have data readouts or catalysts, anything you wanted to talk about in terms of what investors need to watch for over the next 2 to 3 quarters from you.

Yes. So I just have very bad hearing, and it goes out that way, and I apologize, I didn't get everything. But stay tuned. We'll have a bunch of good data for you soon, I hope. We'll have data. I hope it's good. And we're quite optimistic that it will be as type 1 diabetes and an ability to go after patients with a whole bunch of B cell-mediated autoimmune disorders. Thanks.

Thank you very much, everyone. Appreciate it.

Thank you so much. Thank you for being here.

Thank you.

Thank you.