Hello, everyone. I hope you're all having a very productive Jefferies 2024 Healthcare Conference. Today, I have the pleasure of introducing Steven Kelly, CEO of Carisma Therapeutics. Steven, how are you doing?
Doing well, thank you. Thanks for inviting us, and a pleasure to be here.
Always. Maybe we could actually just kick it off with an overview of your company and the platform, and why are macrophages important, and how do you think CAR macrophages monocytes may help improve cancer?
Sure. Yeah, so Carisma is focused on engineered myeloid cells, monocytes, and macrophages. We think that they have a differentiated mechanism of action relative to other adaptive cell therapy approaches, T cells and K cells. And specifically, when you talk about their role in oncology, there's three things that set them apart. Number one is their ability to traffic to tumor. Lymphocytes are generally physically excluded from getting into a tumor microenvironment. Macrophages, conversely, are preferentially recruited to the TME. That's one. The second difference from the CAR-T, CAR-NK approaches is that macrophages are involved in tissue homeostasis, and they can remodel their environment. In our case, what we do is we polarize cells towards an M1 phenotype. And so the remodeling occurs that takes an immunosuppressive environment into a pro-inflammatory environment. And the last thing is macrophages are professional antigen presenters.
And so as they're eating through tumor, they can process and present various neoantigens out through MHC to the T cells they recruited in. So in these three ways, the macrophages are quite distinct. There's a direct killing mechanism through phagocytosis. There's a remodeling effect. And there's a second killing mechanism through driving an adaptive immune response. And so we're excited about it. And we've been in the clinic. We have some early data on our first iteration. And then we have, of course, our next iteration is in the clinic now as well.
Yeah, and so maybe segueing directly into that, you recently had phase 1 data for your CT-0508 in combination with pembrolizumab. Obviously, we've also had monotherapy data on that. Would you be willing to share some of that and some of the key takeaways?
Yeah, absolutely. So our first product was a HER2-targeted CAR macrophage, CT-0508. And it was a first-in-man study for an engineered macrophage. And we learned a lot. I think that the takeaways from monotherapy were, one, we can make the cells. Two, they traffic, like we expected, they traffic to tumor. Three, they can remodel the TME. And we did see a pro-inflammatory cytokine production. We saw recruitment of additional M1 macrophages. We saw recruitment of CD8 and CD4 positive T cells. So we are remodeling the TME. We saw evidence of epitope spread or antigen presentation and epitope spread through increased T cell clonality. We also saw anti-tumor effect through the achievement of stable disease. Now, the things that we didn't see, one, we didn't see objective response yet. And we think that it's primarily due to two things.
One, that we're likely underdosing, and we have relatively low persistence with a macrophage. Secondly, we saw T cell exhaustion in this late-stage patient population. And so we didn't get a chance to mobilize that adaptive immune response. And so our strategy is twofold. One, move towards a CAR monocyte approach, which enables us to create more cells. And secondly, to add pembrolizumab. And as you said, we did some work, some early work on pembrolizumab plus CT-0508. And while it's very early and anecdotal, we have a case study that shows it is adding the value that we hoped it would. We saw the greatest tumor lesion shrinkage of any patient we've treated to date. We were able to see stable disease in the presence of T cell exhaustion markers. And we saw increased T cell clonality. So it is additive.
We're still, of course, looking for the 0525 approach, which is our CAR monocyte, and to drive clinical benefit.
Great. And then there was also from your phase 1 combination of CT-0508 and pembrolizumab, corticosteroids may have played a role with interfering in some of those results. What have you seen with corticosteroid use and CAR and macrophages in kind of your preclinical models, and how are you mitigating use moving forward?
Sure, yeah. It's an unfortunate set of circumstances that happened in our first dose cohort of 0508 plus pembrolizumab. 2 of the patients were experiencing grade 1 CRS. Their families took them to the emergency room. The physician prescribed steroids to address it. It was a shortness of breath. And unfortunately, corticosteroids, we've seen pre-clinically, can kill the macrophages. So it's unfortunate they're not available for clinical activity. They were available for safety. And it turned out that they never had anything worse than grade 1 CRS anyway. But what we've done is to ensure that that never happens again. All PIs are instructed to avoid that. So you can use fluids. You can use Tylenol. You can use Toci if it's really bad. So we do have an approach. We also have instructed the patients and their families to have a card with them.
So they do not ever receive steroids. So it should not be a problem. It was unfortunate that that happened with those two patients, but it did. And so we're dealing with it.
Yeah, it was unfortunate, but I don't want to call it necessarily a blessing in disguise, but maybe a kink that was important to work out kind of early on.
Absolutely. So you can't tell a patient or a physician what to do in that circumstance. They did what they felt was right for the patient. It just turns out that it's not what the PI would have done. And now by educating them prospectively, it shouldn't happen again.
Even before in our discussions, it seemed relatively clear that CT-0525 is what you were going to move forward. Maybe reiterate and maybe go in more depth. Why is a CAR-monocyte approach advantageous compared to the CAR-macrophage approach, both in terms of how can it actually behave in humans and also perhaps manufacturing advantages?
Yeah, absolutely. Yeah, so 0525 is a CAR monocyte. If you think about sort of the differentiation of the myeloid cells, you start with a monocyte, and those cells differentiate into macrophages or dendritic cells. So what we saw with the macrophages is there was a finite number of cells that we can manufacture. It was on average about 1.7 billion cells. Since this is not an expandable cell, they're terminally differentiated. What you get is what you get out of the manufacturing process. We felt, again, that we were underdosing. We were able to see cells in the tumor, but not a lot. And we saw that their persistence was a little lower than ideal. And so we decided to put work and effort into creating CAR monocytes. Very easy to make more cells. So we were able to, we can make on average 10 billion CAR monocytes.
So a 5-fold improvement. What we also know is that monocytes are the cell that typically traffic through peripheral blood. Macrophages are tissue resident cells. And so we were putting a tissue resident cell in circulation, and it's sticky, and it's getting sheared, and things like that. So we weren't getting tumor localization to the extent that we wanted. Monocytes are small spherical cells. They're designed to traffic through peripheral blood. They're designed to extravasate into tissue and then differentiate into macrophages. So what we found pre-clinically is that there's a 40-fold improvement in the ability to get into tumor. So not only are there a higher number of cells, but there's better tumor localization. The last thing that was a little unexpected is that the half-life of monocytes, even post-differentiation into macrophages, is around 45 days. As comparison, CAR macrophages, the half-life is around 5-7 days.
So we were able to get about a tenfold improvement in persistence. So remember that the two issues we were facing were cell number and persistence. And the CAR monocyte approach addresses each one of those. There were other biologic advantages as well. They're better cytokine producers. They're better killers. And so it's a better cell overall. Now, when you ask about manufacturing, there's also some advantages in manufacturing. The manufacturing process for a CAR macrophage is around eight days. The manufacturing process for a CAR monocyte is one day. So it's around 12-14 hours. And we're able to make more cells. We're able to, we can use the same CAR. We can polarize the cells towards an M1 phenotype. And we're able to start the differentiation process before cryopreservation. So the cells will go back to the site. Vein-to-vein time decreases is about two weeks now.
I think overall, it's very much an incremental advantage over what we saw with macrophages. We're excited to be in the clinic with that program now.
Yeah, and any hypotheses of why there's such a longer half-life for the monocytes?
Yeah, so there's a couple of ideas that we've had. One is macrophages in circulation, they're not designed to do that. So there's shear forces starting to destroy them. Another idea or speculation is that they're out of the body for a shorter period of time. Our CAR macrophages were out of the body in a differentiation process in a culture for five days. And so I think there's some data suggesting that the shorter time with the CAR-T, for example, outside the body is a superior cell as well. So it's not surprising. They just seem to be a healthier, more robust cell. And that's what we saw pre-clinically. We'll see if it happens in the clinic as well.
Well, great. So you're moving forward with the CAR monocyte program. How is enrollment proceeding for the phase 1?
We did announce first patient treated. We don't talk about enrollment post that first patient until we get to the cohorts filled. So I can't really say.
That's okay. Well, maybe we'll just expect initial data.
Yeah, so in terms of data, we're hoping to have our study design is a dose escalation. So 3 billion cells in cohort one, 3 billion cells, I'm sorry, 10 billion cells in cohort two. Our goal is to have that data by the end of the year.
Great. Will there be any more cohorts for that study, or is it?
Yeah, so the idea here is we establish safety at up to 10 billion cells. From there, I think it's incumbent on us to design the right protocol for a phase two program. The two things that we're looking at are, one, the dosing schedule. With 10 billion cells, it introduces a little bit of flexibility. We can have a peak dose of 10 billion cells, or we could design five cycles of 2 billion cells, for example. So we're looking at the dosing protocol for CAR-M. Then the second question is, where do we incorporate a T cell checkpoint inhibitor? If we see these late-stage patients, they can probably benefit from that. So we're looking at both of those. We recently brought on a new chief medical officer. He's working on that. We're going to nail that down.
We'll be submitting an amendment to the FDA around the IND and the protocol. The goal is to have a seamless transition from those first two cohorts into cohort likely three and four.
Maybe just touching base again on those first two cohorts, how does the cell number compare to the cohorts you established for the CAR macrophage?
Yeah, so on average with CAR macrophages, we had 1.7 billion cells, so the 3 billion would be just over a 50% increase in the number of cells. And then the 10 billion will be over 5x the number of cells.
So you'll be able to have 5x the number of cells with 10x the half-life.
Better trafficking.
Better trafficking. You could produce it in one-eighth the time.
Correct.
That sounds like a good deal to me.
It seems pretty good. I think it introduces, well, just cost of goods goes down. Sort of the throughput in facilities improves as well. So there's a lot of benefits to this approach.
Will you be shifting kind of a lot of the strategy of future programs toward monocytes or macrophages or still going to be very much in the?
Well, I think ultimately the effector cell we want in tissue and having the effect are the macrophages. But the monocytes turn into macrophages and give us all these benefits. So it's likely, I tend to think about it in terms of a prodrug. A CAR monocyte is a prodrug for a CAR macrophage. It turns into a macrophage in tissue. So likely we would go forward making CAR monocytes. And then the effector cell is when they differentiate.
Excellent. And then maybe next steps after the phase one monotherapy for CAR monocyte, would you again kind of combine it with pembrolizumab and get another phase one type study?
Yeah, so I think a lot of that depends on the outcomes we're going to be seeing with the CAR monocyte single agent. Again, if you just do 5 times the number of cells, 40 times improvement in trafficking, 10 times persistence, it's a 2,000-fold increase in the overall pharmacodynamic pressure in the system. So single agent activity is a real possibility. Now, if we are able to move upline, and I think these days we're thinking post-Herceptin-based regimen, post-Herceptin, but if we lock into a third- and fourth-line, they may not have such an exhausted T cell phenotype. In the later-line patients, we probably are going to see some exhausted T cells and probably there would be added benefit for adding Pembro to the regimen.
Now, given that you have kind of this 2,000-fold improvement in pressure, is there anything you need to be cautious of on the safety side with monocytes?
Yeah, so I did say earlier that monocytes, they are better cytokine producers, but they're better cytokine producers on CAR engagement. So the CAR engagement doesn't occur until they're in the tumor. And we also know that the cells are going to traffic more rapidly than macrophages. So rather than about 24 or 48 hours, it's probably going to be closer to 24 hours before they leave circulation. So it's unlikely that they're going to be producing a significantly higher number or higher level of cytokines in circulation, which is what you'd expect to see with CRS. So we don't think so. Of course, that's why we're doing the dose escalation study to find out.
Then maybe just broadly, I know it's very, very early, but what is kind of your thoughts on, at this stage, the registrational path for CT-0525 after phase 1?
Yeah, so one, define the regimen, define the tumor type, define the line of therapy, and then move forward. So in purely speculative, at this point, we're thinking that it's obviously a CAR-Monocyte in our minds. We'll probably be looking at maybe cycles of monocytes with or without pembrolizumab. The tumor types, there's a couple of large HER2-positive tumors. Breast is an obvious one to go after, but it's a very competitive environment. So there may be a GI cancer that makes more sense. And line of therapy, like I said, almost all patients are seeing in HER2 at this point. So it's probably locked in post-Herceptin, so third or fourth line.
We also obviously want to highlight here that these are; you're directly going for the solid tumors, not a blood-based approach.
Yeah.
Yeah, so excellent. And maybe shifting gears, there's a pipeline of other products and macrophages you have developed out of your platform. I think one of the programs I'm most interested in, I think is kind of the most underappreciated at this point, is what you've demonstrated with engineered macrophages and pre-clinical liver fibrosis and MASH models. And it's very timely as well. So can you share some of those?
Yeah, absolutely. So liver fibrosis, as everyone knows, is a high unmet medical need, especially in advanced liver fibrosis in F4 patients. And while there has been quite a bit of interest and new approaches that are in recently approved advanced programs and then others that are following on, a lot of people are looking at how do we address steatosis, how do we address fibrosis. What we know is that there is an analogous company that's working on unengineered macrophages in liver fibrosis. It's a company called Resolution Therapeutics. They have a relationship with Stuart Forbes in Scotland. And they've done some work showing that unengineered macrophages can traffic to liver and they can have a benefit in slowing the progression of fibrosis. That's great. It's the first time we've ever had read-through to another program.
So our idea is to, we have a really deep, deep understanding of how to manipulate these cells. We think that we can impart antifibrotic and anti-inflammatory properties. So as you say, we recently were able to show that we could take a macrophage and have it produce Relaxin and IL-10, an antifibrotic and an anti-inflammatory. The idea here is that we could have these cells, they will traffic to liver, they'll repopulate the liver, Kupffer cells have been eliminated. Just the macrophage alone is going to have a benefit for being there. When they engraft, they will then become micropharmacies for us, producing this hormone Relaxin, producing a cytokine IL-10.
What we've seen preclinically is that in a carbon tetrachloride model of liver toxicity, or it's a liver toxin, so it shows liver fibrosis, we were able to clear or resolve all liver fibrosis in our murine models. 100% resolution, it makes it look like a healthy liver again. Compared to an unengineered macrophage is about 50% or so. We took that one step further and said, okay, well, fibrosis, what if we look at a MASH model with sort of an exacerbated fibrosis? We had a diet-induced model where we high-fat diet for mice for six months. Concomitantly, we administered this carbon tetrachloride. These are robust mice, we'll say, with significant liver fibrosis. Similarly, we had a dose of our engineered macrophages, relaxin IL-10 macrophages, and we were able to see a 50% improvement in liver fibrosis in those mice.
So that was a really high bar and we had great activity. We also have some data in lung fibrosis, a different model that we've done. But I think the fibrosis idea takes advantage of a natural biology of a macrophage. And since we can genetically engineer them, we can add different properties to them. So we're excited about that program. And our goal is to arrive at a development candidate by Q1 next year.
Great. And in terms of kind of the liver fibrosis improvement, the 50% you saw, how does that compare with some of the kind of recently approved products in MASH and things in the pipeline like the GLPs?
Yeah, so there are so many different models of fibrosis and steatosis, MASH and NASH. It's hard to make head-to-head comparisons, especially pre-clinically.
Pre-clinically, right.
What we've done is looked at some literature. For example, in our more aggressive model with the diet-induced plus the toxin, I think we've seen some data where Madrigal's product, I think, achieved around a 15%-20% improvement versus what we saw, which was just over a 50% improvement.
Amazing. And then in terms of the macrophages in this context, what type of approach are you going to take?
Yeah, so in an advanced setting, F4 patients, I think that economically an autologous approach could work. I think there's no question that that would make sense. I think from a feasibility and just an approach that would make sense, we're leaning towards an allo approach. And we hear the stories about allo. In this case, an allo macrophage is something we can make. We can make an iPSC macrophage pretty easily. We can engineer them to whatever factors we end up finalizing. And it still has the same ability to engraft in the liver and produce and live for months. But it gives us an off-the-shelf ability. I think economics will be a little bit better. And that's one thing we're evaluating. Alternatively, we could also look at an in vivo approach. So we can do like what we're doing with Moderna in oncology.
We could identify an LNP that makes sense and mRNA that delivers, again, the genes that we want. So those are two things that we're evaluating. I think our goal, when I say the development candidate, is not just the construct, but also the modality. So like I said, we're leaning towards an allo, but there may be opportunities elsewhere.
Great. And maybe nicely segueing into kind of the Moderna program, how's the program partnership with Moderna progressing? How many candidates are you kind of currently working on with them? And maybe you could remind us of some of the potential partnership economics?
Sure. No, it was a great partnership we put in place about 2.5 years ago. It was a 5-year partnership on 12 oncology targets. It was $80 million upfront, fully funded by them. All the development and discovery work that we're doing is funded by them. And on the back end, there's about $3 billion in milestones and a good royalty as well. So we've had a great partnership to this point where we have a lead program. I can't tell you what the antigen target is or the indication, but I think when we can, it'll make a ton of sense. But we announced a lead nomination. I think it was in December we announced that. The next step is the development candidate nomination and the handoff to Moderna. So we do all the discovery and optimization. We do some of the pre-clinical work.
It passes to them. They finish the pre-clinical work, IND, manufacturing, and clinical development. So the next milestone is development candidate. And then after that, it will go to an IND filing and hopefully rapidly into the clinic. At that handoff, we have a milestone that we receive. At first patient dose, we have a milestone that we receive as well. And after the handoff, we start working on the next antigen target. And so we again do the same work. We move it forward, et cetera. So far of the 12 antigen targets, we have four that have been nominated. And so the first one, like I said, is close to development candidate. And then we'll move the next one in and so on and so forth.
Great. Maybe lastly, what's your current cash position and runway and what will that help you develop?
Yeah, we have $56 million in cash that gives us cash runway into Q3 2025 and enables us to hit the milestones that I've talked about, the clinical data on CT-0525. Hopefully, it'll be the development candidate in IND for the in vivo program and a development candidate in modality for the liver fibrosis program.
Excellent. Well, we really appreciate your time and good luck.
All right. Thank you very much.