Welcome back to the next session of the 2024 Leerink Partners Global Biopharma Conference. I am Mani Foroohar, Senior Analyst of Genetic Medicines, and I am very fortunate to be hosting Rachel Haurwitz, excuse me, Dr. Rachel Haurwitz, President and CEO of Caribou Biosciences. Rachel, how is Miami treating you?
So far, so good. Can't complain about the sunshine. Thanks for having us, Mani.
I'm glad to have you guys, and I guess I'm hosting in more than one capacity. So welcome to my adopted hometown. Before we dive into individual nuances around timelines, etc., let's chat a little bit about where Caribou is in terms of pipeline prioritization, strategy, etc. Obviously, there's been an open debate around cell therapy world in terms of both indication selection and, more broadly, oncology, non-oncology use of the same technology. Give us a little bit of where your philosophy is and how you think about monetizing the opportunity in your core tech.
Yeah, fantastic question. Big question. At the heart of Caribou is what we call the chRDNA technology. This is our next-generation CRISPR technology platform that is dramatically more specific than first-generation CRISPR-Cas9. It allows us to carry out quite sophisticated multiplex genome editing while maintaining genomic integrity. Now, you can imagine a wide variety of potential therapeutic applications. Our initial focus has been on off-the-shelf cell therapies for oncology.
So today, we have three clinical-stage allogeneic CAR-Ts targeting three different hematologic malignancies. That said, and I know we'll get the chance to dig into a lot of those programs, we recognize that is but the tip of the iceberg in terms of the broader potential of both our cell therapy capabilities and our genome editing technology writ large. So certainly, our research team today continues to really dig in on potential new applications, ranging anywhere from the potential for cell therapy in autoimmune all the way through to the potential for in vivo genome editing for rare genetic diseases.
So let's talk a little bit about the autoimmune side of the story, in part because obviously, you and I have had a number of conversations, yourself talked about it, different conferences around your approach to your core oncology target. There's been a lot of investor interest in cell therapy approaches to immunology. Some of that's existed in large-cap pharma, some of that's in other private companies, etc., a little bit driven by academic data, interpret as you will.
Where does Caribou fall in terms of your development efforts in that area, and is that something, does it fall in the universe of ongoing BD effort? Is it something that most appropriately lives in a JV, an approach that your predecessor on this panel, Natalia, talked about? How do you think about monetizing that value and capturing it in the right way that doesn't burden your own OpEx, your own infrastructure, limited headcount, etc.?
Well, I think you've just hit the nail on the head in terms of how do you not be a kid in a candy store and get massively distracted by the incredible breadth of potential of what our cell therapy and genome editing technologies can do. So today, the lion's share of our resources are dedicated to the continued clinical development of our three phase 1 assets in oncology. And as we look at some of these other things that our research team is exploring, including autoimmune, we have to think really hard about what is the right way to invest in some of these opportunities.
Is it directly off our own balance sheet? Is partnership a better way, the only way? I think for some of these opportunities, let's take in vivo genome editing, there are a lot of pieces to the puzzle. Delivery is one of them. That's probably a great example where partnership might be the only way to move into that space. I think some of these other opportunities, there are multiple paths forward. I will say the early autoimmune data, largely in academia, to your point, really intriguing, right? It's a small number of patients, but very intriguing in terms of what a fundamental reset of the immune system can do.
And I think CB-010, which is our CD19 allogeneic CAR-T, has some interesting attributes as you think about that. Certainly, in our dose escalation work in oncology, we've demonstrated that we can drive the same kinds of durable responses you can see with an auto CAR-T, and I think that's very promising.
It does introduce a little bit of a strategic question of, obviously, you mentioned CB-010 has some unique attributes. It raises interesting questions of, would you partner off a construct that functionally is CB-010, or would you want to have a completely separate construct for manufacturing or other reasons, which perhaps might be against the same target? How do you think about those decisions and what product you would then partner?
I think there are really interesting pros and cons to all flavors of this. I think we're also at a point in time where the field writ large understands more, maybe not a lot, but more, about how CAR-T cell therapies can work in oncology and what success looks like. It's super early in autoimmune, and even the phrase autoimmune is a huge umbrella term, right? There are lots of different diseases that are functionally driven by incredibly different mechanisms. You might want different approaches for some of these different diseases just based on the field's understanding today.
That makes sense to me. So stepping away from that particular debate, I wanted to chat a little bit about some of the—you guys released earnings last night. I want to talk a little bit about some of the nuances and disclosures you made last night. We'll start with the earlier stage kind of extended pipeline about the CAR-NK side. Talk to us a little bit about how you think about the role, where that technology falls in the portfolio, and sort of the impact of last night's disclosures in terms of timing to clinic, development strategy, impact on OpEx, etc.
Yeah, great question. So to level set for those who did not read our earnings release last night, Caribou has two cell therapy platforms: healthy donor-derived allogeneic CAR-Ts, and we have three phase 1 assets from that platform. The other is iPSC-derived CAR-NKs. We had a program called CB-020 that was targeted to ROR1 for certain kinds of solid tumors, and what we shared yesterday is we've decided to hit pause on that particular asset. It is a comment on that asset, not the CAR-NK platform writ large, and in fact, we remain quite enthusiastic about the potential of editing iPSCs and differentiating them into armored CAR-NKs.
We just did not have the confidence that CB-020, what we call this asset, had all the right bells and whistles to bring the right features to the clinic for patients.
So let's roll forward. Is the presumption that we will see a replacement CB-020, next-generation version of CB-020, whatever the—a future CAR-NK asset, is that something we should expect on a defined time horizon? Is this sort of a little bit back—are we back to defining our development candidate, back to target selection? Where are we in that technology?
Yeah, I'd say our research team remains as busy as ever in terms of continuing to really progress the capabilities of the CAR-NK platform. In terms of future cell therapies in oncology or otherwise, I think we'll always think really hard about the two platforms that we have and which may be the most appropriate chassis for the disease or target in front of us. I didn't answer the other piece of your question, so I'll come back to it for just a moment. You asked how these impact runway.
Actually, largely independent of this change. This change had a very modest impact. We did share yesterday that through our sort of regular way budgeting and prioritization, we've extended our runway from Q4 2025 to Q1 2026.
That's helpful. There was another—I was lost in my seven-part question—but around where are we for CAR-NK? Are we back to target selection or not? Should we expect a future CAR-NK to also be directed against the same target, or is it a reasonable supposition to assume it would be a different target?
Different target, I think, is the right expectation.
Okay, that makes sense. So we're going to pivot over to ANTLER and its descendants, if that's okay. So I'll also disclose last night in earnings alongside financial updates, etc., discussion of CAR-NK was a little bit of an update and some commentary around pivotal trial plans and timeline of disclosure for the CB-010 program. I still mentally think of it as the ANTLER program, even though that's just the name of the study. Maybe a shortcoming of my own. But why don't we talk about what the update was, and then we can sort of dive into some nuances, what it means from a modeling perspective, etc.?
Yeah, absolutely. Again, to level set for those who are a little less familiar than you are, Mani, CB-010 is our lead program. It's an allo CAR-T targeting CD19 for non-Hodgkin lymphoma. Last year, we concluded dose escalation. We enrolled 16 patients in dose escalation. We shared these data in the middle of last year, where we demonstrated that CB-010 can drive the kinds of durable responses achievable with auto CAR-Ts. Based on those encouraging data, we actually went to the FDA and proposed moving into the second-line setting, which they greenlit.
Since then, we've been working in dose expansion exclusively in second-line large B-cell lymphoma patients. Yesterday, we shared that we've hit our target. We wanted to enroll 30 patients in dose expansion. We have now done so, and we are on track for sharing initial dose expansion data in Q2. So what's in that? What's coming next? In the Q2 update, we will have at least 24 of those 30 dose expansion patients. That'll give us a chance to really dig in for the first time on efficacy, safety, and translational data, including PK and PD for these patients.
Now, for those paying attention, there were actually 4 second-line large B-cell lymphoma patients in dose escalation as well. So all told, 28 patients who are second-line large B-cell lymphoma patients. Now, we're not going pencils down. We are continuing to enroll patients today in ANTLER to continue really building out the dataset. And you're absolutely right. We have guided that in Q2, we'll give a lot more details in terms of the timing of the start of the pivotal phase 3 trial. We had initially anticipated it would start by the end of this year.
Where we sit today, we believe it's now a 2025 initiation. So what are some of the moving parts here? A couple of pieces that I'll highlight. One is, as we've now really studied a lot of the data coming out of ANTLER, we have observed that there's actually an enhanced efficacy for the subset of patients who received product manufactured from a donor who had a minimum level of HLA matching. We are really intrigued by this enhanced efficacy, and we would like to actually proactively study this.
And so that's really our focus for our continued enrollment in ANTLER right now, is to look proactively at this and enroll another, let's call it, 20 patients to do that work. We also recognize that there's a geographic difference between what we're doing in phase I and what we plan to do in phase III. So pretty much all of the patients we've enrolled so far in phase 1 have been U.S based patients. For the phase 3 trial that we envision, we fully expect that we will largely enroll that ex U.S.
So there's some really important differences between these patient populations. Here in the U.S., obviously, at the major medical centers, patients have a lot of options. Physicians have a lot of options in the second-line setting. And so we know that many of the patients have come on ANTLER specifically because their physicians think they cannot wait for auto CAR-T, whereas many of these ex-US geographies, I'll point to Australia as maybe the poster child, we have no commercial CAR-T in the second-line setting, creating a really nice window of opportunity for us to recruit patients who look a lot more like the patients who were recruited into the pivotal trials, ZUMA-7 and TRANSFORM, for Yescarta and Breyanzi.
So let's talk a little bit about that dynamic where, in some ways, you've almost got two different patient cohorts between the U.S. and OUS, especially in places where you have functionally no available CAR-T. Use the Australia example, but there's lots of other places. Let's talk about how to think about evidence of incremental efficacy for those U.S. patients who, I'm putting words in your mouth a little, but you characterize as rapid progressors within the time horizon of vein to vein.
What is the right comparator and bar for efficacy for those patients versus the right comparator bar for patients who look more like a population that would be enrolled in TRANSFORM? What does Zuma?
Yeah, look, I think the fundamental bar, as we think about the future of CB-010, irrespective of patient population, is to develop an allo CAR-T that can meaningfully rival the auto CAR-Ts. Now, how we actually quantify that obviously depends on the patient population. To your point, the work that we did in dose escalation, largely third-line and later patients. And so we used the initial pivotal trials for Yescarta, Breyanzi, and Kymriah as kind of the comparator set there. Now that we've moved into the second-line setting, the gold standard studies are ZUMA-7 and TRANSFORM.
And so I think understanding these patient differences are an important part of digesting the totality of the data. It doesn't change the fundamental goalpost for us, though, right? The goalpost is to develop an off-the-shelf CAR-T that can meaningfully rival the auto CAR-Ts.
So while that might be that sort of goalpost, North Star, how you want to think about it is unchanged. Interpretation of the datasets will be more complex. So should we expect that data from those patient populations or from patients similar to each of those populations in the ongoing ANTLER study to be disclosed separately? Because investors will need to evaluate, well, what is this going to look like in each of these two settings? Because commercial realities are different. As you mentioned, competitive realities are different. How does that affect how you expect to present data and what we'll see when from ANTLER?
Yeah, those are really good questions. I'd actually focus on the other piece of what I just updated on in terms of how we'll think about what are the subsets of patients and what do they tell us about the go-forward strategy. Let's focus on the HLA matching piece. What we have observed is that for a subset of patients, they happen to have a certain threshold of HLA matching relative to their donor. And that has led to a really encouraging enhancement of results for those patients. And so I think you'll see us really try to double-click on that in terms of understanding the role that HLA matching plays in the opportunity to enhance efficacy for these patients.
I do want to come back to that. Talk to me a little bit about how to think about the scale of that subset and how large of a proportion of eligible patients it is.
Yeah, that's a great question. It's probably about a third of the patients on study who fall in that category, which gives us enough to say we think there's a there there and why we're now going to study this proactively moving forward.
Is that true across geographies, or is there sort of—or is that something that varies?
True across geography.
So is it reasonable for us to expect some, this is the kind of place where I would expect preclinical data, cell-level data, product-level data would be something that would be very helpful to investors? Is evaluating essentially the durability of the immune response, for example, what would be a time horizon where we could expect to see those data?
Q2. Q2 is where we will really dig in on this.
Great. Recognizing that your ability to point to a date is somewhat limited, what is a venue that would be reasonable in your mind?
So we've been specific about quarter, but not venue. Mani, as you know, historically, we've used both medical meetings and corporate events to share data updates both around the table.
You're not going to tell me Chicago versus Spain, are you? So let's take a step back. What does the enrollment pace in the U.S. tell you about the depth of unmet need in this population? Because there has been a debate of whether or not enrollment was even plausible in the U.S., given your comparator arm. So walk me through what that means in terms of the universe of eligible underserved patients in the U.S.
Yeah, let's start with the commercial setting, and then I'll back it into our enrollment rates. Gilead has been quite public that only about 10% of U.S. second-line patients are getting auto CAR-T today. 90% are not, right? And so I think that demonstrates a really meaningful opportunity for an allo CAR-T to start bridging that gap. And by the way, the fact that physicians are choosing the ANTLER trial specifically because patients can't wait for auto CAR-T, I think, is proof positive that an allo CAR-T strategy really can grow the pie in terms of the number of patients who can access a CAR-T cell therapy.
In terms of enrollment, our team continues to work very hard in terms of opening additional sites. Sitting here today, we have 29 sites open. 24 of those are U.S., 2 Australia, 3 Israel. That's obviously a lot more than we had about a year ago when we started dose expansion. We started dose expansion at the end of March 2023, so not even a year ago. We've now concluded enrolling the 30 patients we were targeting. I'm pleased with the enrollment that we had with fewer sites then and excited to have even more sites partnering with us now.
That makes sense to me. Let's take a step elsewhere in the pipeline and talk about multiple myeloma. Your partner, not partnered.
Friend, our friend.
Yeah, fine. And what is the ongoing role of Pfizer in that program, other than providing some fairly senior-level expertise? And when we're going to get what from that program?
Yes. So in June of last year, Pfizer made a $25 million equity investment in Caribou, laser-focused on CB-011. So in exchange for making that investment at a pretty healthy premium to the then stock price, they get a 30-day right of first negotiation to CB-011. And they also get a seat on our scientific advisory board and quarterly meetings with the team to get updates on the program. And they have no control over the program. So from our perspective, it's a really lovely relationship.
We get to take advantage from their expertise and their advice while we maintain full control of the asset and the clinical development for this program. We have committed to the first data update from this program by the end of this year. So to give people a sense for where we are, we've successfully concluded dose levels 1 and 2, no DLTs. We're actively enrolling at dose level three. By the end of this year, we expect that update to include initial safety, initial efficacy, as well as some initial translational data as well. Our objective is to have a meaningful enough initial dataset that allows us to really inform the key next steps in terms of the future clinical development to CB-011.
So you've talked about how the development in myeloma is very different than CD19 because there isn't like a 6-month, there isn't this like 6-month CR rate sort of metric. It's more of a traditional oncology endpoint dynamic. So when you talk about disclosing metrics that are what you would use and decide in the future of the program, what might those look like?
I'll start with what's the bar to beat. You know, we spend obviously a tremendous amount of time with myeloma physicians, really trying to understand from them, what are they looking for? What does an allo CAR-T need to accomplish to be relevant for their patients? And I think we have to ask the humble question, does it need to look like Carvykti to be what they're looking for? And uniformly, the answer we hear from them is, bispecifics are the bar to beat.
As they look at the broader patient population, and of course, keep in mind, myeloma is a much larger patient population than lymphoma, as they think about that broad patient population, bispecifics are the only thing that they have that is broadly available and readily available to those patients. And that's fundamentally how they're going to make that judgment call.
Now, here's what they tell us they're hoping for. They're hoping for an allo CAR-T that's in the same ballpark from a response rate perspective as the bispecifics. But they also expect it will have a better safety profile compared to the infection rates that are challenging with the bispecifics today.
So I think one implicit component around that bispecific, non-CAR-T, etc., dynamic is the availability of bispecific versus cell therapy. Obviously, the J&J and others have talked about expanding infrastructure, making investments, supporting growth, however you want to, whatever phrasing you want to use in various conference calls, to allow for broader market access and expansion. When you think about that as something that could influence and change the structure of myeloma treatment in the US and the EU5, I suppose, how does that fit into when you think about what the bar is? Like, at what point is there any risk that bispecifics are no longer the bar? Like, how broadly accessible does CAR-T have to be?
That's a really important question, right? We can't just focus on what does supply and competition look like today. We have to think out into the future. Certainly, even the physicians we speak with, even as they imagine significant infrastructure scale, they still see a pretty meaningful gap between supply and demand in terms of the overall patient population. So at least based on their feedback, I do think there is a gap that will need to be bridged. I also think we cannot lose sight of the value of the readily available one-and-done strategy for patients that an allo CAR-T offers that neither auto CAR-T nor bispecifics offers to them.
So let's talk a little bit about exactly that dynamic around availability and around the debate. And obviously, our colleague Daina Graybosch has done a lot of work on this. How far are we? And how much of an infrastructure build would there need to be for something like a CB-010 to be very broadly accessible outside of major population centers? Is that a five-year process, ten-year process? What does that look like?
There are probably a lot of other CEOs in the auto CAR-T space who might answer this question more specifically or better than I do.
There's no bias whatsoever, I'm sure.
But at least the way I think about it, supply is a piece of the puzzle, but it's only a piece, right? You're still asking patients to undergo this very lengthy multi-step process of getting in a queue to undergo apheresis, depending on their disease, likely need some sort of bridging therapy to keep them healthy enough, let alone alive long enough to ultimately undergo lympho depletion and receive their therapy. And today, that's largely centered on the patients who are being treated at the MD Andersons of the world.
Taking lymphoma as an example, a huge fraction of second-line patients are in the community. They're not at the MD Andersons of the world. I think serving those patients can best be done by an allogeneic strategy that does not require half of the steps that I just outlined there, more than half of those steps, and that can be potentially delivered to a community setting more readily than an autologous CAR-T cell therapy could.
That makes sense to me. And I think when we think about access to the community setting, let's talk a little bit about the safety profile and infection risk that you mentioned around the bispecifics. Obviously, a real challenge. To what extent is that blunted uptake in your mind, or shifted patients towards other therapies? And what would you need to show with an allo to clearly take advantage of that opportunity?
My suspicion, based on public comments from other companies, is that at least especially in lymphoma, the fundamental challenge is not inpatient versus outpatient. It's where are the patients? And the majority of those second-line patients are in the community setting, whereas it's the MD Andersons of the world that are the centers of excellence for delivering CAR T, which, by the way, many of them are using outpatient strategies today, right? Many commercial CAR Ts are given not in an inpatient unit, but outpatient with very thoughtful and careful safety protocols, usually keeping the patient within a certain geographic proximity for a certain number of weeks, for example.
And there are lots of ways to address and handle side effects if and when they occur. I think it's less the outpatient versus inpatient and more just fundamentally, where are they? Obviously, I drank the Kool-Aid a long time ago. Mani, I really believe an off-the-shelf cell therapy is the only kind that can make a broad impact in the community setting.
Right. We're coming down to the end of the half hour now. Obviously, we're going to see a lot of further updates from you guys somewhere into the queue. Looking forward to seeing that. As always, a pleasure to have you.
Thank you, Mani. Really appreciate it.