Hi, welcome back, everyone, to the next session of Citi's Virtual Oncology Leadership Summit. It's my great pleasure to have with me the president, CEO, as well as founder of Caribou Biosciences, Rachel Haurwitz. Rachel, welcome. Thank you so much for taking a few minutes out of a super, super busy schedule to chat with us. Just as a reminder, everyone, if you do have questions for Rachel, you can just email me, and I will check my email and be able to forward them over to her in real time. So with that, again, Rachel, thank you so much for the time. I think a lot of people are obviously very familiar with the company, but it would still help to give us a quick overview of the founding technology, the chRDNA platform.
Why is chRDNA different? How does it differ perhaps from some of the, you know, the earlier approaches to CRISPR technology? And then from there, we can go into all the, all the clinical data and the clinical trials. Thanks.
Fantastic. Well, Yigal, thank you so much for the invitation to participate today. I really appreciate it. You're right, at the heart of our company is what we call the chRDNA technology. This is a next-generation CRISPR platform that we've invented here at Caribou, and it's far more specific than first-generation CRISPR-Cas9. What we did is actually invent new guides, so rather than using the all-RNA molecules that Mother Nature gave us, a few of my colleagues invented new guides that are hybrid materials. They're part DNA, part RNA, and it turns out the inclusion of DNA dramatically improves the specificity of genome editing compared to first-gen CRISPR-Cas9. We can actually use these guides with multiple different CRISPR enzymes. So, for example, our lead program is manufactured using the chRDNA guides with Cas9, whereas our next two clinical stage programs are manufactured using the chRDNA guides plus Cas12a.
In each case, what this lets us do is carry out quite sophisticated multiplex genome editing while maintaining genomic integrity. And so today our focus is really on using this platform to advance a wholly owned pipeline of allogeneic CAR-T cell therapies. There are three clinical-stage programs I, I know we'll have the chance to talk about today. And across each of them, we use the chRDNA technology to armor, to enhance the potential anti-tumor activity of these cells in one or multiple ways, and, and I look forward to getting into some of those details today.
Okay, perfect. We don't want to spend too much time on the technical aspects, but they are important, so thanks for the brief overview. Well, let's start, of course, with the lead program, which has been in the clinic for some time, CB-010. This is the CD19 CAR T, PD-1 knockout, and you're developing it in large B-cell lymphoma. So first of all, just tell us about the PD-1 knockout and why that is so important, especially with respect to the durability thesis, which, as we all know, is, you know, one of the areas where you need to show strength in the allo cell development space.
Yeah, it's a really important question. So maybe taking a few steps back, we're obviously deep believers that the cell therapy field has to move to an off-the-shelf setting to really deliver the power of cell therapy to a broader patient population. But then you fundamentally have to address head-on the challenge that an allo cell therapy is foreign to the patient's immune system and will likely be rapidly rejected. And so that really sets up our whole thought process at Caribou about how we can use our genome editing in a multitude of ways to address this challenge. You're right that with CB-010, we address it by knocking out PD-1. So I want to be clear: I don't think the PD-1 knockout alters the immune-mediated rejection, but what we intend it to do is to prevent premature CAR-T cell exhaustion.
So that means our vision is that while CB-010 is present prior to immune-mediated rejection, it does not become exhausted through the PD-1, PD-L1 axis, and that it can maintain a higher anti-tumor activity state for a longer period of time. Preclinically, we're actually able to ask a lot of these really interesting and sophisticated questions, and so before we ever brought the program into the clinic, we actually ran an animal model where we evaluated CB-010 head-to-head against an otherwise identical CAR-T, but that was still able to express PD-1, so it didn't have that PD-1 knockout. And so that really gave us a lot of confidence that the PD-1 knockout strategy could be quite meaningful in this disease setting.
Okay, and now you've done quite a bit of work, obviously, in the dose escalation so far. I believe it's, I think, 16 the NHL patients that you've treated, 69% CR rate, 44% six-month CR rate. So beyond that, just kind of give us the overall, you know, color on the dose escalation and how you believe the data that you're generating is gonna fit into the broader CD19 CAR-T landscape.
Yeah, so our fundamental goal for CB-010 is to develop an off-the-shelf CAR-T-cell therapy that can meaningfully rival the auto CAR-Ts. Now, how you actually measure that quantitatively or qualitatively depends heavily on which patient population you're treating or talking about. So our dose escalation work was largely in third line or later patients. That's obviously a very traditional place to begin a first-in-human phase I study. And across those 16 patients, you're right, we're able to see a very compelling overall response rate of 94%, the 69% complete response rate that you mentioned, and I think most importantly, seeing that 44% of those patients achieved a CR at six months or greater. So to put that in context, in the third line setting, about a third of patients who receive an auto CAR-T achieve a CR at six months.
And so these data really, really taught us that CB-010 can drive the kinds of durable responses achievable by an auto CAR-T. Now, based on those encouraging data, we actually went back to the FDA and proposed an amendment to our protocol to leapfrog directly into the second-line setting for our dose expansion work, and the FDA greenlit that. All of our dose expansion work has been done exclusively in second-line large B-cell lymphoma patients. Perhaps stating the obvious, this decision was driven both by an understanding of the evolving unmet medical need as well as the commercial opportunity and fundamentally, our desire to be able to serve the largest patient population possible. We're excited to be in the second-line setting today.
Okay, and so have you discussed which dose you're going to be taking forward in this, in the second line, in a pivotal trial? And then obviously, you had some discussions with the FDA very recently, just I think towards the end of last year, which gave you clarity on the control arm. So can we talk about that? Because that was an important point in the development to go against chemo immunotherapy versus the auto CAR-Ts.
Yeah, thank you. So where, where are we headed next? What we've committed to is that in Q2 of this year, we will disclose initial dose expansion data, and we can certainly dig in on some of those expectations in a moment. That keeps us on track for our expectation to initiate a pivotal trial in the second-line setting by the end of this year. We will disclose in Q2 what the RP2D is. However, Yigal, I will share in the dose expansion work, we actually brought all three doses from dose escalation into dose expansion. So that was 40 million, 80 million, and 120 million CAR T cells.
And really, the goal was to be responsive to Project Optimus, which is the FDA's expectation that sponsors do robust dose-finding work, and to ensure that we had done a sufficient amount of that in our target population in the second-line setting as well.
I know we can all go back and look at the swimmer's plots, but for everyone just listening to us talking, the 44% six-month CR, was there some correlation with dose there? Or just... Can you talk about that a little bit, just so people get a better perspective on how you're moving with the dose respect, with respect to efficacy?
Yeah, quite remarkably, our starting dose, which was 40 million CAR-T cells, was an efficacious dose. Now, if you compare 40 million to many of the other CD19 assets, including the approved auto CAR-Ts, it's actually a pretty low dose. And Yigal, I believe that the efficacy we've seen at that dose is really driven by the impact of the PD-1 knockout. We'd certainly hypothesized, based on our preclinical data, that the PD-1 knockout could result in enhanced therapeutic index. Perhaps I didn't expect it to be quite as enhanced as what we actually saw in dose escalation. And Yigal, I realize I didn't pick up the last piece of your previous question, so I'll come back to it quickly.
Okay.
You're absolutely right. We had a very productive conversation with the FDA at the end of last year, and that discussion really centered on the control arm for the phase III trial we'd like to start later this year. And so we proposed, and they agreed with using chemo immunotherapy followed by stem cell transplant as the control. This looks a lot like what was done in both TRANSFORM and ZUMA-7, which were the trials that led to the second-line approvals of Breyanzi and Yescarta, respectively. Now, we owe a visit to the FDA again this year to really align on all of the key elements of the protocol.
Yeah, I was just about to ask. So you mentioned you're going to have the dose escalation data in the second quarter, and then, as you say, another meeting with the FDA. At that point, is it fair to assume... I know you can't tell us which dose it's going to be, but is it fair to assume you will pick a dose for the pivotal versus sort of multiple doses versus-
Correct.
It'll be a recommended Phase II dose. That's correct.
Okay. Okay. So we talked about that. And then with respect to the patient population, anything more you can say in terms of where you might hone in there, in terms of the second line, and in terms of demographics, or is it just going to be sort of like an all-comers situation?
Yeah, great question. So I, I can talk both about the work we're doing now and, and the phase III as, as we're thinking about it. So the ongoing dose expansion work in our phase I ANTLER trial, we're targeting to enroll about 30 patients in dose expansion. We started that at the end of Q1 of 2023, to give people a sense for how long this study has been going on. So in terms of our Q2 update, we've committed to that being a meaningful fraction of the target 30, which obviously will include a range of follow-up, as I just highlighted when the first patient came in, and, and obviously, some of the patients have been dosed much more recently. So our objective will be to, obviously disclose the data, the RP2D.
It'll be an opportunity to talk about not only overall response rates, but also duration of response, and also a chance to dig in on some of the key translational data, including PK and PD. And obviously, all of these data really wrap around, informing that determination of the recommended phase II dose and the go-forward strategy. In terms of the phase III itself, I think there's a tremendous amount that we can continue to learn from both TRANSFORM and ZUMA-7, and I'll just highlight that those two trials have a key difference between the two of them. TRANSFORM permitted what's called a crossover, whereas ZUMA-7 did not.
So that means TRANSFORM patients could cross over to the CAR-T arm if they had an insufficient response to the chemotherapy, whereas in ZUMA-7, that wasn't possible, and it resulted in needing far fewer patients to enroll TRANSFORM than ZUMA-7. So these are some of the details that we're paying careful attention to as we think about our future engagement with the FDA.
... Okay, although I guess that cuts both ways in terms of the size of the study versus potential overall survival outcomes and so forth. But that's something we obviously will have to work through.
Yeah, interestingly, as I understand it from some of the physicians who are part of ZUMA-7, many of those patients went on to have CAR T anyways, but it was outside of the context of the trial itself.
I see. Okay, got it. That makes sense. And then with regard to the dose, just again, I mean, it sounds like you're seeing effects across all three doses. So I mean, it's not... To me, it seems like there's not necessarily a clear choice as far as which dose. I mean, maybe the FDA just will say, you know, Project Optimus, raise the Project Optimus flag, and just tell you to go with the lowest dose. I don't know. I mean, it's just, what are you thinking?
If you look across the cell therapy field, by and large, it looks very different from the other therapeutic modalities, where in other cases, you'll often see an almost linear dose responsiveness. In the cell therapy setting, it's often more of a threshold, where if you're below the threshold, you have little to no efficacy. Above the threshold, you have efficacy without huge differences between increasing doses thereafter. And so I think certainly the CB-010 data that we shared last year are consistent with that, and in fact, it appears all of the doses are above that particular threshold. And so we'll be using really the totality of efficacy, safety, PK and PD data to inform our determination of the RP2D.
And then, as you look beyond, you know, pivotal trials into the hopefully commercial setting, the second line, you know, there'll be a lot of choices for physicians, allo for you, others will think about auto. How are you? What does your channel checks tell you about what physicians are thinking about what their approach is gonna be in the second line? And what kind of, you know, what kind of pricing power you can implement there?
Yeah. So as we think about CB-010, right? Our objective is to develop an allo CAR-T that can meaningfully rival the auto CAR-Ts. So that gives us actually two pieces of the pie as we think about the commercial opportunity. One is to compete directly with the auto CAR-Ts for the market share that they have gained and are gaining today, and the other is to actually increase the overall opportunity by serving patients who will never be served by autologous CAR-Ts. So to put that in numbers, at least from a patient population perspective, there are about 10,000 second-line LBCL patients here in the U.S. every year, and right now, only about 10% of them are receiving auto CAR-Ts.
Now, DRG thinks that by, you know, 2030, 2031, that's maybe a third of patients getting auto CAR Ts. So we think CB-010 could have the capacity to compete for, for part of that market share, in addition to serving patients who are not able to receive auto CAR T. Yigal, I think some of the work that we're doing today is already proving out that second thesis. We know that some of the patients who are being enrolled in ANTLER, their physicians have recommended ANTLER specifically because they think they cannot wait for auto CAR T manufacturing.
Interesting. Given the, you know, the off-the-shelf nature of allo, and easier accessibility, you know, with that, you would like to bring with it, you know, easier, perhaps easier tolerability. Or how do you think about the, you know, the classical side effects we know of with these types of therapies, the CRS, the ICANS? You know, you wanna hopefully avoid, you know, avoid the inpatient experience, if possible. You know, just tell us a little bit about how you're thinking about that, given you want an allo product that is sort of, quote unquote, "user-friendly.
Yeah. Yeah, absolutely. You know, starting with our data from last year, you know, for that 16-patient cohort, we were really pleased to see that CB-010 has been generally well-tolerated. Obviously, there are three AEs of special interest in this context: CRS, ICANS, and infections. Across those 16 patients, there were no cases of Grade 3 or higher CRS, only 2 cases of Grade 3 or higher ICANS, and only one Grade 3 infection. So I think a very favorable comparison against the auto CAR-Ts in that context. You are absolutely right that the holy grail for cell therapies is to get them into the community setting, and I think it's much more feasible to understand how an allo CAR-T could truly be delivered at scale in the community compared to an auto CAR-T.
There's obviously a lot more complexity in terms of the blocking and tackling for an auto versus an allo, so we think the community setting is ultimately going to be the holy grail for this opportunity. Step one will be the opportunity to evaluate CB-010 in the outpatient setting, and that's something I hope we'll be able to do in the future.
Okay. Let's move on then, unless you wanted to make any more comments on, on CB-010. I think we covered, covered it in good detail. So the next one, CB-011, going in sequence, so this is the BCMA CAR-T. Here you used the immune cloaking, the beta-2 microglobulin knockout with the B2M-HLA-E insertion. So I think, first of all, just remind everyone what that means and why that provides the immune cloaking, and what the choice—why the choice for that, for this particular product for BCMA?
... Yeah, thank you. So I, I guess going back to the problem statement, recognizing that allo CAR-Ts will be rapidly rejected compared to auto CAR-Ts. This program, we felt it was most appropriate to try to immune cloak the cells. And I want to be clear, this is about slowing down immune-mediated rejection. We're not trying to create the perfect stealth cell that lasts forever. We actually don't think that's necessary or, or even particularly helpful, but this is about buying additional time for additional anti-tumor activity. So the strategy has two steps, as you've outlined. The first is to knock out a gene called beta-2 microglobulin. So beta-2m is necessary for the presentation of all class one on the surface of the cell. So it's a very simple way of just wiping out all class one in one fell swoop.
So that solves one side of the equation, right? Now, if you're devoid of class one, the CD8 cytotoxic T cells in the patient's immune system won't be able to readily recognize and clear the therapy. But if you stop there, if that's all you do, you now run the risk of the patient's innate immune system, through their natural killer cells, rapidly rejecting the therapy. Because NK cells are specifically tuned to recognize what's called missing self, or the complete absence of class one presentation. So that's why we take it a step further, and we site-specifically insert a transgene that fuses beta-2M with HLA-E.
So the goal then is to have a CAR-T product that lacks all endogenous class I, only expresses this HLA-E transgene, and therefore should be able to keep the initial CD8 cytotoxic T cell and natural killer cell onslaught at bay to buy additional time for additional anti-tumor activity.
So I understand the point is that the CD38-positive, they can't see the beta-2m HLA-E. Is that right?
Right. Exactly. Exactly.
Right, and then you also solve the problem with the NK cell. So I think we understand that. Now, you're running a study, you're in dose escalation. I think you said you're gonna have initial data by the end of this year. So I know you're fairly limited in what you can say, but what can you say in terms of the enrollment? Well, moreover, you know, what are the goals? What do you want to see? What is the TPP that you want to see? What will get you excited in terms of your first cut of data that's gonna make you want to invest more dollars here?
Yeah. First off, I'll confirm that's correct. We have committed to an initial update by the end of this year. The bar to beat really is the bispecifics. So we spend a lot of time with KOLs in this space and ask them about the rapidly evolving nature of treatment opportunities for myeloma patients. And what we're told is really that the only approach that is broadly available to this patient population is the bispecifics, and so that's how these physicians will compare and contrast an allo CAR T as another off-the-shelf, broadly available strategy. So what does that mean? That's about a 60%-70% overall response rate.
Importantly, they have expectations that an allo CAR-T would have a more favorable safety profile than the bispecifics, and I think they're particularly thinking about the high rates of complex infections that these patients experience. Now, I'll also highlight, many of them point out to us the real advantage of the one-and-done nature of an allo CAR-T over the bispecifics, for which patients are really tethered to the infusion clinic for a very long period of time, and therefore are experiencing these side effects for a very long period of time. So to bring it back to the work that we're doing right now, we have disclosed that we have successfully concluded dose levels one and two, no DLTs.
We're actively enrolling at dose level three, and our goal for this update by the end of this year is to have a sufficient initial data set to really speak to the safety, the efficacy, the PK of CB-011 to drive the fundamental next steps for the clinical development of this program.
I'm going off memory here, but I think it was 50 to start, $50 million to start with. Is that right?
Great memory. Yes.
Okay, so is it gonna be the... My, really my question is, is it - do you expect the same sort of threshold effect as you were pointing out for the lead program, or is it, could it be different?
Given that it's been generally true for many cell therapies, I think that's a reasonable starting hypothesis. You're right, our first dose was 50 million CAR T cells. The second dose level was 150 million CAR T cells, and today we're enrolling at dose level three, which is 450 million CAR T cells.
Okay, got it. All right. And so before the end of the year, it sounds like we're not gonna hear a lot, maybe, maybe just a little bit on enrollment and so forth.
I think that's a fair expectation.
Okay. Okay. And what are you hearing from, you know, in terms of the investor enthusiasm for the study and the enrollment? You know, you mentioned the bispecifics obviously have a, you know, fairly high bar. You know, what are physicians telling you, just anecdotally, about how this could fit in? Does it have to be better? I mean, you mentioned the repeated having to go to the infusion clinic a lot for the bispecifics. Are you saying you need to beat the 60%-70%, or not necessarily? What can you say there, just kind of generally?
Yeah, I'll say, KOL enthusiasm is off the charts for this program. There's a lot of excitement about the potential for an off-the-shelf CAR-T. I think one of the ways I can demonstrate that is just the really rapid number of sites who've wanted to be part of this initial phase I trial, which we're very excited about, a number of very high-quality locations throughout the United States. What we hear from physicians is, they'd like to see an allo CAR-T in the same ballpark. So they'd like to see an allo CAR-T with at least a 60%-70% overall response rate. Again, with that sort of built-in expectation that the safety profile, especially with respect to infections, would be more favorable for an allo CAR-T compared to the bispecifics.
... Now, I should have asked earlier, I mean, these are relapsed/refractory. What can you say? Like, are they third, fourth? Are these the salvage patients that have seen all the anti-myeloma mechanisms, you know, the Velcade and so forth, all the old Celgene drugs? Have they seen some of the bispecifics or no?
Potentially. So in terms of our inclusion/exclusion criteria, patients have to have seen at least three prior lines of therapy, and we've been specific about those, so that's a proteasome inhibitor, an IMiD, as well as a CD38 antibody. They can have seen other BCMA-targeted therapies. However, they cannot have had a prior cell therapy exposure. Yigal, I'll highlight maybe another piece of evidence of interest in this program, is the investment that Pfizer made in Caribou in the middle of last year. So they made a $25 million equity investment at a pretty healthy premium to where the stock was trading at the time. And that was with a specific interest in CB-011. So they actually get a 30-day right of first negotiation on the program. They get a seat on our scientific advisory board.
We provide them quarterly updates, and those funds that they invested in Caribou are specifically earmarked for the clinical development of CB-011.
Okay. And does that $25 million, does that basically cover the cost of your- this first trial?
I don't know that we've ever been super specific about the exact costs-
Okay.
But suffice to say, the $25 million are meant for CB-011.
Okay, and you mentioned the 30-day right of first refusal, so that is... Like, what's the trigger? I mean, when you report your first clinical data, they get 30 days, or some other trigger?
It's a 30-day right of first negotiation rather than refusal, so I'll highlight that. It's a pretty light right.
Okay
... they have 36 months from when this started, so that was, I don't know, more than half a year ago at this point, and it's triggered by someone else coming to the table. If someone else-
Oh, okay
... expressed an interest in CB-011, they'd get to jump in and have their 30 days.
Got it, got it. And then just to be clear so everyone understands, the—for the 50, 150, 450, so the 50 and the 150, are those closed now? Are you still backfilling, or is it just the 450 that's open?
That's a great question. So 450 is the open dose from a safety perspective. Our protocol has tremendous flexibility built in, that we could theoretically backfill at cleared doses. Right now we're focused on enrolling at dose level three.
Okay, and is that... The protocol goes, you're not gonna go to 1,350, right? That's too much. That's-
1,350 is a big number. I doubt you'd see us go there. Maybe to put this in context, Abecma, which is one of the two approved auto CAR-Ts in this setting, has a label that actually has a range. Now, I suspect that's due to the complexity of autologous CAR-T manufacturing. That range is 300-450 CAR-T cells, so we're somewhere in the same park as their label dose.
You're in the right zip code, as, as people like to say. Okay. All right. So that's CB-011. We'll stay tuned for that. And then the most recent one is CB-012, so this is the CLL-1 for relapsed/refractory AML. And then here, interestingly, you joined forces on strategy to do PD-1 knockout and immune cloaking, which is... I was always wondering when you would do that, so that's good to see. So tell us about why this is interesting. AML is a very different situation from the lymphomas and the myelomas, which are kind of, you know, slower progressing. So tell us how what the strategy is here and why AML?
Yeah, great, great questions. You know, look, fundamentally, as we thought about how best to execute on our vision of an off-the-shelf strategy for CAR-T cell therapies, we felt it was really important to begin evaluating our vision that armoring, whether it's PD-1, immune cloaking, or otherwise, is really important. To do so in the context of what we know success looks like, there are two of those. That's CD19, and that's BCMA, and so we've initially put our focus on those two targets, with the goal of, of advancing therapies that we think can really do something meaningful for, for those broader patient populations, and having the backdrop of the auto CAR-Ts to, to know what success looks like for those patients.
And then quickly, we wanted to be able to take all these capabilities and start deploying them against diseases which have no meaningful cell therapies today, and I think AML is the poster child for that. It's an incredibly challenging disease, and I would personally hypothesize, Yigal, this is a disease where I think it's allo or bust. Said another way, I think your average AML patient has such chewed-up T-cells that it is unlikely that an auto CAR-T will broadly be a useful strategy for that patient population. And so I think it puts a lot of emphasis on our capabilities as a genome editing company to actually bring together multiple of these key strategies, including the PD-1 knockout and the immune cloaking.
To actually make CB-012, it is 5 genome edits, 3 separate knockouts, two separate site-specific insertions, and the chRDNA platform allows us to do this at quite high efficiency while maintaining genomic integrity. So I, I think it really sets us apart in terms of the sophistication of the kinds of therapies we can bring into the clinic. Now, clearly, target is a really key piece of the equation here, and we know that target identification has been one of the big challenges in this space, and so we got really excited about some of the work being done on CLL-1 at Memorial Sloan Kettering. CLL-1 is highly expressed on AML tumor cells and even on leukemic stem cells, but importantly, it is not expressed on healthy hematopoietic stem cells, and that was a really important discrimination to us.
And so we were actually able to exclusively in-license fully human binders that MSK had developed against this target, and we're using one of those in the CB-012 CAR today.
Okay, so I think you've answered the question in terms of the, the advantages potentially of allo. But that does raise the other question, which is the, you know, the preconditioning, the lymphodepletion, and how, how are you gonna help the AML patients, you know, manage through the, the lymphodepletion to get to your allo therapy? So what, what is the thinking there? 'Cause I know with the, you know, with the other programs, the, the, it is a pretty intense lymphodepletion regimen. Is it, is it different for this, or are you gonna stick with the same thing?
Actually, each of our programs is different.
Mm-hmm.
So ANTLER evaluating CB-010 has a deeper lymphodepletion. It's the same two drugs, it's cyclophosphamide and fludarabine, but it's a little more of each than you would typically see before in auto CAR-T. And that was really driven by very strong advice from our scientific advisory board, as they were acknowledging that CB-010 has the PD-1 knockout, but does not have immune cloaking. And so felt that a deeper lymphodepletion was important and appropriate for that patient population to allow for sufficiently robust initial engraftment. Now, in CaMMouflage, in the trial where we're evaluating CB-011, because we have the immune cloaking strategy, we've actually reverted to the more traditional chemotherapy approach, more traditional lymphodepletion approach, and it's the exact same regimen you would get before an auto CAR-T. Now, AMpLify for CB-012 is somewhere in the middle.
It's a little more lymphodepletion than CaMMouflage, but not as much as ANTLER. Again, being very responsive to the specific nature of the disease and the therapy itself. I will highlight, Yigal, these patients in AMpLify have to be eligible for a bone marrow transplant, and so that means they have to be sufficiently robust. And so the approach for lymphodepletion is less than what they would experience in the context of a bone marrow transplant.
Okay, that's helpful information. Now, you know, we do a lot of work on the small molecule side in AML, less so on cell therapy, for the reasons you cited. So you know, typically what we've heard is that the sort of de facto bar is this 20% CR rate, and a six-month duration of response. Is that an applicable benchmark in the cell therapy world, or what is the thinking there? Has the FDA weighed in on that at all? Do...
I'd say the KOLs have, right?
Okay.
A KOL conversation usually starts with something like, "We'd really love a cell therapy that has some activity for some of our patients, please.
Yeah
... which is obviously a very different conversation than where the lymphoma or myeloma docs start. I think it's a reflection of just how wide open the space is in AML and how severe the unmet medical need is. When you really start to push them for more quantitative metrics, you know, on average, we'll hear that they're hoping to see an allo CAR T with something like a 25%-30% overall response rate. In terms of CR rate, probably something like 15%-20%, and they've certainly shared with us their frustrations that with the limited auto CAR T data available for these patients, the median duration is less than six months, and that that's a problem for those patients.
Okay, so the study is going to... And you mentioned AMpLify. I think, so everyone is aware, that is the name of the phase I trial. That study is starting, I believe, in the next few months. Is that correct?
Yeah, we've committed to dosing our first patient in the first half of this year.
Okay, and what more can you say? It's just going to be the typical late-line AML patient, well, that have failed all multiple mechanisms, you know, like the obvious standards of care, the venetoclax, azacitidine, 7 + 3 chemotherapy, things of that nature?
Yeah, it's a great question, and I think the answer speaks to the complexity of this patient population and wanting to make sure that we're working with patients who are sufficiently robust to understand the activity of CB-012, rather than simply the complexity of very, very late advanced disease.
Yeah.
As we've designed the trial as it stands today, patients are relapsed or refractory following at least one line of therapy, but no more than three prior lines of therapy, and that's how we're trying to achieve that appropriate sweet spot for patients. They can have had prior stem cell transplant, allo or auto. That is permitted. However, they cannot have had a prior CAR T-cell therapy.
Okay, so that's interesting, 'cause you already said one to, at most, three prior lines, which obviously raises important questions around future combinability, potentially. I know this is just—you're just getting going, but I'm sure you're having initial conversations with the thought leaders. I mean, is the thought that potentially you could combine CB-012 with some of the more classical regimens, like the ones I just mentioned? Or that, I mean, you obviously have to see the data, but is that sort of in the opportunity set or not? Or you hope that this would be a monotherapy, and then, you know, you would move on?
The big vision for CB-012 is that it could be a monotherapy, right? And one of our reasons for choosing CLL-1 specifically is because it's not expressed on healthy hematopoietic stem cells. And so that means with sufficient activity, CB-012 could have disease-modifying activity on its own and not even necessarily have to be yet another form of bridge to transplant. Now, we recognize that most patients coming on a phase I AML trial, the big goal in any physician's mind is going to be to get their patient to transplant. And so we know in the near term, the goal is to see whether CB-012 can drive meaningful responses to get patients to transplant. And someday, Yigal, I hope we get to understand whether CB-012 can have disease-modifying activity on its own.
Okay, I'm just scanning my email. I'm getting some questions in from the investors. One person is asking, just in terms of going back to ANTLER for a second, sorry to deviate. But just, you're gonna show data at 40, 80, and 120, if I'm not mistaken, at this Q2 dose escalation update... dose expansion update. Just to confirm, so we will have some patients from each of those three buckets? And do you have a rough sense as to how many per dose?
Yeah, great question. Yes, we have dosed patients across all three of those dose levels in the context of dose expansion. And we were targeting a minimum number for each of those cohorts to ensure that we had sufficient information. We haven't been super specific about that number, so stay tuned.
Okay, and then someone else is asking about, I think you said second quarter for the dose escalation, that it sounds like you're not gonna get more specific than that at this point in terms of, when in the quarter or conference.
Correct. Correct.
Okay.
Thank you.
And then another question, if you could put the CB-010 data in context relative to the Allogene asset on an efficacy/safety basis, if to the extent you could.
Yeah, it's a little hard to do. Allogene has evaluated a number of different lymphodepletion protocols, a number of different doses, even different dosing strategies, single dose, multi-dose, and the majority of their work has been done in third line or later patients, though now I believe they're moving away from that particular strategy. So if you look at the totality of their data, at least the data that they have shared publicly, CB-010 has a favorable efficacy and duration of response profile. Certainly a key, I think, safety differentiator between their strategy and our strategy is their use of an anti-CD52 antibody as part of the lymphodepletion protocol, which we certainly hear from physicians can be challenging from an infection perspective. We do not use an anti-CD52 antibody.
Okay. And then just in, as we finish off here, just a few more topics. So moving away from hematologic malignancies, you do also have a new program, a CAR-NK program for solid tumors. This is the ROR1 iPSC. What is the status of that, and what is the goal with targeting ROR1 for solid tumors?
Yeah, great, great questions. This is a preclinical program. We've not provided timelines for next steps on it. In terms of the target, ROR1 is a very interesting target because it is expressed on multiple different kinds of solid tumors, so it actually creates quite a bit of flexibility in terms of potential future clinical development. It is also a target that's relevant to certain heme malignancies, and so that means other people have put anti-ROR1 cell therapies into the clinic. And so there are some safety learnings that we can glean from those as well, that have motivated the interest in this target.
Okay. Since you're, you know, kind of going down the path of developing the iPSC-derived CAR-NK, would you consider the iPSC-derived CAR-T at some point?
I think it sounds really interesting. You know, there are obviously a lot of potential benefits we see on the NK side for the iPSC-derived strategy. One is simply that you get to do all of your editing at the iPSC stage, and then you pick out that single-cell clone that has all the attributes you desire, and then expand and differentiate that population so that your end material is fully edited with all of the sophisticated attributes that you desire. It's obviously different from the primary blood-derived allo CAR-Ts we're making today, where we edit a meaningful fraction, but not 100% of cells. So I love the whiteboard exercise of thinking about iPSC-derived T cells. However, T cells are incredibly complex, right? If you take blood from any one of us, T cells are not one thing.
There are many different phenotypes of T cells, and it turns out that rich mixture is really important in terms of the overall functionality and potency of a potential CAR T cell therapy product. I haven't yet seen an iPSC to T protocol that recapitulates that rich mixture, and if anyone on the call is aware of one and wants to share that information, please look us up.
Okay, fair enough. Okay, so then another big topic is manufacturing. Tell us how you've invested in that. Well, who's doing the manufacturing now for the clinical trials? Is it sort of on an in-house scale? Are you using external vendors? But more broadly, what are you planning to do for, you know, for launch? How that's going to work, and what initial investments have you made or, you know, pilot projects to understand that?
Yeah, great, great questions. I'll say one of our very most recent investments is actually bringing on board our first CTO, Tim Kelly. I'm delighted to have him on our leadership team as of last month. Someone who has decades of experience in the tech ops and manufacturing space, and really excited to have his expertise. Internally, we have built strong process development and analytical development capabilities. So what that means for any one program is that our internal process development team owns everything from that initial handoff from the research team through to scale-up and optimization, and then they tech transfer the process to CMO partners, who make the materials for us.
We've not disclosed who the CMO partners are, who we work with, but we have specifically chosen organizations who have not only clinical experience, but also commercial experience and capabilities to support us over the long term. We obviously have not dedicated resources at this point in time to any internal build-out. However, we may choose to do so in the future for strategic reasons.
Okay, and then, you know, and I guess I never really asked this before, but, you know, when you think about off-the-shelf and, and you have a commercial manufacturing facility, I mean, what... There is still a lag. I mean, it can't be instantaneous. So what is-- what does that actually mean? By the point where the physician orders, how quickly would you foresee actually getting the cells? You know, if they're made in Boston, and you need them in New Mexico, for example, what are we talking about? Is it literally, like, a day, like, by FedEx, or...? Well, just help us understand that, 'cause that is kind of like the Holy Grail.
Yeah, this part of the supply chain is very straightforward.
Okay.
So as far as the patient is concerned, their product is already sitting in a freezer somewhere, and the moment they need it, and they begin undergoing lymphodepletion, it is drop-shipped to their clinical site. You're right, it's an overnight, it's a one-day shipping.
Yeah.
We actually deliver it in a bespoke dewar that can hold the material at the appropriate temperature for some period of time. So the site doesn't even have to have sophisticated storage facilities. The dose-
Mm
... comes with its own storage.
Okay, got it, got it. Now, of course, I would be remiss if I didn't ask you one more question about another angle, which you know what I'm gonna ask, is, are you thinking about autoimmune? Everyone seems to be jumping on the bandwagon, a lot of your peers in the cell therapy space. You know, you must be at least at some, at the discovery level, looking at it. I get the question all the time. What can you say?
Yigal, I'm so surprised by the question. No one has ever asked us this before. No, in all seriousness, obviously, the lion's share of our resources have been dedicated to the clinical advancement of our three oncology programs that are in phase I studies today. And Yigal, you've known us for a long time. I think you know Caribou is a company that stays pretty laser-focused on doing and resourcing a small number of things that we think we can excel at and progress to important milestones and go/no-go decisions. All of that said, we recognize that those three programs in oncology represent but the tip of the iceberg in terms of the potential of not only our cell therapy capabilities, but our chRDNA technology platform writ large.
So our research team absolutely continues to think through and evaluate a number of exciting opportunities, anything from the potential in autoimmune, all the way through to the potential for in vivo genome editing for rare genetic diseases. I, I think for any potential new application, we will always carefully ask ourselves the question of the best way to resource one of these opportunities, whether it's directly off of our balance sheet or potentially in collaboration, depending on the scope, the nature, the opportunity, and, and other technologies or expertise that may be warranted.
Okay, makes sense. And then just in the last minute or two, didn't touch on the financial strength of the company. I think you have a runway for a few years. Can you just tell us what the runway looks like, how much cash you have? And then the... You mentioned we've already talked about some of the catalysts, like the IND filing for the newest program, the expansion data for ANTLER, but just to recap the catalysts for this year.
Absolutely. So we ended Q3 of 2023 with $397 million in the bank. That gives us a runway into the fourth quarter of 2025, so strong balance sheet. And you're right, a number of key milestones for the team for this year. For CB-010, two milestones that we've guided to, the dose expansion data and the RP2D disclosure in Q2 of this year, and then the intention to initiate a Phase III trial by the end of this year. For CB-011, we've guided to initial dose escalation data by the end of this year. In CB-012, we've got the trial open. We've guided to dosing our first patient in the first half of this year, so a lot of exciting moving pieces for the team.
Awesome. Well, we certainly look forward to all the updates and seeing, seeing how the pipeline evolves. Thank you so much again for the time. I know, you got a lot going on, so I appreciate it.
Thank you, Yigal. Really appreciate it.
All right, take care. Bye.