Good afternoon, everyone, and thanks for joining us at the Morgan Stanley Global Healthcare Conference. I'm Mike Ault, one of the biotech analysts here, and it's my pleasure to introduce Steven Nichtberger, President and CEO of Cabaletta Bio. Just a reminder, the format for today is a fireside chat. But before we get started, I just need to read a quick disclaimer. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. With that out of the way, we can jump right into Q&A. Steven, maybe just to start, if you could provide a brief background on your CART platform.
Sure. First, thanks for investing the time to be with us today. The Cabaletta CAR platform is built on the standard CART19 platform, in fact, based on Kymriah, which was first discovered and developed at the University of Pennsylvania. Our company was spun out of the University of Pennsylvania to focus on a next iteration of a CD19 type of platform. In classic CAR T therapy for cancer, what you do is you take the antibody fragments as the targeting domain on your T cell, and that binds to the CD19 antigen on a B cell that is cancerous, as well as a normal B cell. Our scientific founders, one of whom was an inventor of Kymriah, in 2011, first identified the opportunity to put an antigen onto a CAR T cell.
In the case of pemphigus, the DSG3, desmoglein 3 antigen. That's the skin glue that is attacked when there is a pemphigus patient. That patient has B cells that make antibodies that attack the skin glue, the DSG3. If you just think it through, we know that CAR T is a very effective killing machine, killing the B cells that it targets. If the antigen that is the subject of attack in pemphigus is found on the surface of the CAR T cell, then that antigen will attract the B cells that want to make antibodies that are causing pemphigus.
It will not attack the normal B cell population, the thinking being that the patient who is treated with a CAA R, of the sort that our platform produces, will end up with no pathogenic autoreactive B cells because they are tagged by the antibody that they're going to make, and they are killed by our product in an effective use of our product. What's left behind is a normal immune system, unlike CAART 19 therapy. Our platform is really simple, really elegant, based on clinical situations that we know CAART 19 can work. We started with pemphigus vulgaris because it's really a prototypical B-cell disease.
Yep. Makes sense. Maybe since we can just start with DSG3 CAART, that's your lead program for pemphigus. Maybe you can just talk about the initial sort of phase one dose escalation data. You had some updates this weekend, but maybe just start by talking about the design of that study and then sort of the key takeaways in terms of where we are today.
Sure. In that study, it's an open label study. Any patient who has pemphigus that has been treated with and failed a single form of therapy, that could be prednisone, it could be rituximab, it could be cyclophosphamide. They've tried and failed, any form of therapy would be eligible. The vast majority of the prevalent population with pemphigus, the mucosal form, would be eligible for our study. We started out with safety coming first. These types of therapies have not been used previously in autoimmune patients, so we started out with a 20 million cell dose arm. We then went up to a 100 million cells, 500 million cells, and then 2.5 billion cells, followed by up to 7.5 billion cells. Just absorb that for a minute.
20 million cells up to 7.5 billion cells. 375-fold increase in dose. Not surprisingly, the 20 million cell dose, even through the 2.5 billion cell dose and the 7.5 billion cells, very safe, because the elegant platform predicts that we're not gonna kill every B cell. We're only gonna kill the 1% of your B cells that are causing pemphigus. Maybe it's fewer than 1%. A very small fraction of your B cells are gonna die if our therapy is effective. You would expect that our therapy should be very safe. In the first three cohorts that we reported, we saw no clear signs of biologic activity or clinical engagement.
In the fourth cohort that we just reported at the EADV meeting in Milan over the weekend, we reported out the safety data, the 29-day safety data in the first five cohorts. Remarkably, the first four cohorts. I should say, we didn't use preconditioning in the first set of cohorts. We used a monotherapy approach. Unlike rituximab, which uses very high dose of steroids, 3.5 grams in the first year to get responses in patients with pemphigus. Unlike the SDRNs, where frequently you found using concomitant patients who have steroid doses on board, some of whom can migrate off eventually. There is an attempt on our part to use monotherapy alone to figure out what our cells can do in the absence of anything else.
The 20 million cells, the 100 million cells, the 500 million cells, and the 2.5 billion cell doses, along with the 7.5 billion cell doses, up to 7.5 billion cell dose, are all very safe. We saw one low-grade fever in the highest dose cohort. It was transient, it was self-limited, and there were no other associated signs or symptoms, probably related to a lot of activated T cells being infused into the patient. The safety profile, really excellent for the platform. The efficacy or the biologic activity profile in the first three cohorts, we had reported that there was nothing that was a clear signal of activity. Prospectively, we identified four things that we would look for in order to determine if we had target engagement or biologic activity. Those four things were persistence of our product as measured by PCR or flow cytometry.
The second thing is a reduction in the ELISA for DSG3 antibodies. The cause of the disease is DSG3 antibodies. Those would go down. We would look to see a change in the concomitant medications. If you're on steroids, did you reduce or increase the dose? Fourth, whether or not the clinical symptoms improved as measured by objective scales of assessment that are standard for the use in pemphigus. In the fourth cohort, we saw one patient out of three. Two patients did not respond clinically or biologically to our observation. One patient, though, had a really interesting constellation of findings where all four of these parameters were simultaneously, or I should say in sequence, checking the box. Number one, did we have persistence in that patient?
Uniquely, we had persistence that between one month and three months stayed flat. In other words, it stayed high. It was elevated at one month, and it stayed elevated at the same level by three months. No other patient did that. Secondly, between two and three months after therapy, a time when the antibody levels should be dropping if therapy is effective, because the half-life of these antibodies is about three weeks, so you should expect to see a meaningful drop in the antibody titer if you're killing the B cells that are making the antibody. You should see that about two to three months. What do you know? At two to three months, this patient had about a 20% reduction in their antibody titer, something that no other patient in our trial had demonstrated.
Two sequential months, more than a 20% drop where there was no drop prior. We felt really good about that. More than that, this is a patient who had their steroid dose tapered between one month and four months, from 10 down to 1 mg per day. The medicines were reduced, concomitant medications were reduced. Finally, based on the PDAI, which is a more general measure of pemphigus, and the ODFS, which is more specific to the mouth and the pain that the lesions might be causing, both of those were completely resolved in the third and the fourth month in this patient's follow-up.
We had all four of the biologic indicators of activity that we felt were compellingly positive, and you didn't have to squint to figure out that there was something going on here. It was pretty obvious. We also reported at the EADV meeting that the fifth cohort, which is about two to three-fold higher dose than the fourth cohort where this one patient presented, the fifth cohort, unlike all the prior cohorts, did not give us a sequential increase in persistence. Prior to this, the first, the second, the third, the fourth cohort had a 0.96 correlation coefficient in terms of higher dose, higher persistence.
The fifth one looks like it leveled off for reasons that not enough time to go into, complicated, but it told us that monotherapy alone is unlikely to be the solution that we want to get the full efficacy profile out of this platform. As a result, we announced that we'd be prioritizing the use of an age-old drug in the treatment of pemphigus, cyclophosphamide pre-treatment, followed by use of our product. I should say also IVIG would be used in the pre-treatment period in addition to cyclophosphamide to set the patient up for a couple of things. The first is the antibody titer would be reduced by the IVIG. If the antibodies are in the way or a barrier to efficacy, they would be diminished. Just to quickly review, the antibodies that are circulating, they cause the disease.
They bind to the skin glue that is causing the problem, causing blistering the skin. Those antibodies could also bind to our product because remember, our CAR T-cell has the antigen, not the antibody on the surface. That DSG3 antigen is exactly what these soluble antibodies wanna bind to. Our preclinical data suggested that those antibodies might activate and they might inhibit, but neither to a great degree with our product. If the antibodies are a barrier to efficacy, IVIG will diminish those antibodies. Cyclophosphamide will also, by virtue of killing those B cells, eliminate some of the antibodies. In addition, the cyclophosphamide, as it does in cancer, would be expected to get rid of the cytokine sink.
By reducing the sink of cells that would absorb cytokines, the cytokines are more available to have our cells that are infused expand and ultimately engraft. We would expect. We're now enrolling that arm of the study, that sub-study, and we would expect to see data from that by the end of the first quarter, where we would present at a medical meeting the cohort that we treat, and in that dataset, we'll see the safety data. Really important in an autoimmune patient, cyclophosphamide pre-treated, what's the safety profile? Equally important, what's the persistence level? If we have dramatically higher persistence than we have seen previously with the A4 or the 2.5 billion cell dose alone, we already have that data, so we know what it should be if it's the cell dose alone.
If it's dramatically higher, we know we're onto something if persistence will correlate with efficacy. By the end of the first quarter, we should have that data. In addition, finally, I'll say our second program is myasthenia gravis. Our thinking was, let's wait for the MuSK program as long as we can to start it, and the IND was accepted the beginning of this year. Let's wait as long as we can because we can optimize how we start that study from what we learn in pemphigus because they're so similar. There's a critical difference between the two of them, and that is that if the antibodies are a barrier to efficacy, the myasthenia gravis patient on average has one to two orders of magnitude, said differently, 90%-99% fewer soluble circulating antibodies than pemphigus.
If the antibodies are in the way, the right study to do in humans is to treat patients with myasthenia gravis. That study will start enrolling this year as scheduled, and we should start to see data in the, you know, first half of next year around that patient population. Those are sort of the recent data presented, the context for the trial, and what we are doing going forward.
Maybe just quickly on the MuSK study, since that's where we sort of ended up here, but just other learnings from the DSG3 CAART program that sort of helps accelerate the MuSK program.
Yeah. We started by fractionating the dose in four fractions simply for safety reasons. Safety profile has been so good that FDA has agreed we could just infuse the product. We don't have to go through fractionation. Makes application and implementation a lot easier. The second thing is FDA said we don't need to do three patients per cohort, we can do only two. The third thing, I'm sorry, is instead of starting at 20 million cells, we'll be starting at 500 million cells in the MuSK myasthenia gravis population. The way that study will be implemented is 500 million in the first two patients, 2.5 billion cells in the next two, and then we are in the same place we are with pemphigus by the fifth patient, which is combination with 2.5 billion cells.
Just broadly thinking about the platform and different indications, do you think you'd end up in the same place in terms of the optimized dose? In other words, the 2.5 billion cells combo for sort of across the board, or could there be some reasons why you might dose differently?
It's interesting. Some of the investor discussions we've been having over the last few days, cancer CAR T is limited in how many cells you can infuse because no matter how slowly you infuse them, they get activated right away by 100% of the B cells. You had to dose back in the development of the CAR T CAART19 technology. They had to pull back on dose, add the lymphodepleting regimen so that there was a more gradual explosion of the cells. If you went to 2.5 billion cells, you would have a mess. You couldn't do that in cancer. We, on the other hand, have a very safe platform, very specifically and elegantly designed to kill only 1% of the B cells.
We can put in 2.5 billion cells and give perhaps less preconditioning. We're not using fludarabine, for example, but we're starting with fivefold more cells than you find used in typical cancer treatment, around 500 million cells versus 2.5 billion cells. It's a very interesting dynamic. You know, I have to say, we're not familiar with any data in CAART19 cancer treatment where they've dissected apart the effects of preconditioning versus the effects of the product, the cell therapy product. This is really, you know, the work we're doing in order to really understand what is the effect of our cells. And to that end, you know, cyclophosphamide does have an effect on the disease, as does IVIG.
There is on our website a slide deck with five references that talk about the limited durability of any clinical responses that you're gonna see with IVIG and with cyclophosphamide. Both of them are limited to a matter of weeks or a few months in terms of the durability of the treatment effect. By six to nine months, we should have real clarity that our product is working or not, is the point. Sorry.
You touched on dosing relative to CAR T and some of the differences, but do you know when you give CAR T CD19 and you get that expansion, how does that compare to, like, the 2.5 billion dose? Are you in that range, or do you need to sort of get high, way higher?
They're substantially higher-
Yeah.
...than the 2.5 billion. The persistence that you see in cancer, again, you can find it in the slides on our core presentations in the EADV presentation. The level of cancer persistence is higher than we've been able to achieve with our drug alone, with our product alone.
Gotcha. Maybe if we can go back to the DSG3 CAART phase I data.
Yeah.
If we look at the, specifically the A4 cohort, you had one patient that showed, you know, very promising activity. Is there anything unique about that patient that maybe explains that or maybe something that they didn't have versus the other two patients where it wasn't sustained?
Yeah. We don't yet have a specific guide for the journey, a specific marker that's gonna take us right to the answer, enroll patients who have exactly this profile. We do know that that patient started with a circulating soluble antibody level that was in the bottom tertile of those that we treated, and you can see all of them again in the data that was presented in EADV. The problem with concluding that the antibody titer is what correlates is that all the other patients in that bottom tertile who had more or less than that patient didn't respond. There's really no way yet that we've identified to prospectively predict that a patient would be a responder like that or not.
Yep. You mentioned moving to the combinations and having an update early next year, and I think the first given update, I think it's the 28-day update, and the key there is really persistence. Is there a particular level of persistence you're looking for that. You know, or target or something like that.
Yeah. Again, referring to the EADV slide that were presented this weekend or the corporate deck that we have on our website, you'll see a zone we call it a zone of efficacy in cancer, and it's the persistence level, 10 to the fourth and higher, typically, where you see clinical outcomes in cancer. More than the dose in cancer, CAR T, it's the persistence that you achieve in a patient that predicts and associates with the outcomes that you get. In that range is where we'd wanna be.
Yeah.
I think there are two things. One is, are you getting that range? Two is, are your cells activated?
Yeah. Have you thought about maybe looking at the impact of IVIG versus cyclophosphamide? In other words, instead of doing them combined, maybe do them separately and see which one has more of an impact or.
Yeah. In terms of trying to get to an effective outcome, right, we have safety to beat the band. We have great safety in the platform. We need to deliver clinical activity and efficacy, so we're going forward in this cohort. If we have positive outcomes here, I think then your suggestion, peeling away one or the other, makes a lot of sense.
Yeah.
To go at it the other way, no reason in my mind that we should separate them and then combine rather than reverse.
Yeah.
Let's go for the efficacy signal and then pull back if we need to.
Just in terms of teasing out the signal between DSG3 CAART versus the combinations, you know, how do you think about that? At what point in time is it appropriate to sort of try and figure that out or?
Yeah. You know, it's interesting. The cyclophosphamide and IVIG both have a clear published track record of transient efficacy in the first weeks or months after therapy's administered. Cyclophosphamide for pemphigus would have historically been administered on a monthly basis because the treatment effect doesn't last for 6-9 months. We're suggesting that by 6 months, maybe 9 months, we're gonna know whether or not our cells are having effective outcomes or not. It won't be because of the cyclophosphamide by that time, and the literature supports that based on, again, all of the references that we provide in our website. The same for IVIG, which lasts for weeks before the disease will resurge. I think those two things. Whether the physicians and clinicians will accept that, I think.
There was a very interesting use of CART19 therapy in lupus patients. First, in a letter to the editor, in August of 2021, and second in a EULAR presentation. CD19, when used for the treatment of lupus at this EULAR presentation in four or five patients who were presented, all of whom had a complete response, with follow-up up to one year, didn't lead to debate and discussion about whether the full-on preconditioning in that, in those patients, the cyclophosphamide and fludarabine, was responsible for the one-year remission in patients who failed every therapy prior. Instead, there was a standing ovation for what had been achieved.
There need to be a lot more data, but I believe that we physicians, business people, investors underestimate versus cancer, the pain and the suffering of the more severely affected autoimmune patient and the conditions that exist today, where we're willing to spend hundreds of thousands of dollars per year for decades on end to keep patients, not even in remission, just with more modest impact of their disease. I think that there is a healthy appetite for curative therapies, for completely remission, you know, complete remission, even with a preconditioning or a combination regimen in a whole variety of autoimmune diseases. I think the marketplace for physician use of drugs in autoimmune disease is gonna come to include cellular therapies with preconditioning for their more severely affected patients.
Makes sense. You touched on this earlier, as you went up the dose to A5, you didn't see a boost in persistence.
Yeah.
You can talk about why that is, maybe in a less complicated way.
Yeah. You know-
If it can be done.
Right. With that question, you're immediately above my pay grade. Like, that's the answer in the audience. The simple answer is we biologically don't quite know why it was limited. However, the number of T cells that you can infuse and the persistence you can achieve is something that is limited based on prior data in prior studies where T cells have been used in HIV patients and others. It's not a shocking first time ever discovery, but there's a limit to what the body wants to receive in terms of T cells and how much the cytokines can, you know, sustain the population of cells that you infuse. We think we reached that plateau with this infusion. We need to rely on the preconditioned state, which sucks out the cytokine sink. Leaving more opportunity for those cytokines to expand the infused cells.
Maybe you can talk about some of your other ways for optimizing the effect here. In the past, you talked about manufacturing as well and just your current thinking there and, you know, maybe other. Sounds like going up on the dose doesn't necessarily make sense, but, you know, maybe multi-dosing or manufacturing hand.
Yeah. Since 2018, we have always prepared for the curveball that science is gonna throw at us. For example, on the safety side, we have a gene editing partnership from 2018, excuse me, which was originally intended to circumvent any safety issues that we might see. We didn't see any, but the gene-edited version of our product was ready to push into development if it was necessary. Same thing here on the efficacy side. It's not only about gene editing. There are many partnerships that we have and many, I would say programs scientifically that we're nurturing towards the clinic if we need them.
Don't read too much into this because as I told you, on the safety side, we had a whole effort focused on mitigating safety risks that never occurred. Same thing here on the efficacy side. If more cytokine inducement is required, we have a number of strategies that we're contemplating that can be brought forward in 2023 in order to address those sorts of challenges if we need them. So there are a lot of levers to pull, and for years now, we've had the same slide in our corporate deck, and it basically says, "Look, the path to success is almost never straight in biotech. To the extent there are curveballs, we've got a lot of levers to pull." Some of them are dose related. We've been through the monotherapy dosing. Some of them are manufacturing related.
We've chosen not to enhance the manufacturing to get incremental, more activation of our cells. We think there are better ways to get more activation of our cells. The combination is one of those. We're prioritizing that. We still can enhance the manufacturing. Don't make the mistake of believing there's something wrong with the manufacturing. There's not. We're just using a basic plain vanilla manufacturing process. We can activate ourselves to a greater degree by making some minor changes, and that's what we're not doing yet because we're prioritizing the combination, which we think is a better strategy for the business. The gene-edited approaches, there are other approaches that we haven't spoken publicly about that are frankly easier and faster into the clinic for 2023.
I think all of that is within the context of a company that was built to be very capital efficient. We're a cell therapy company, and you know, there are very few cell therapy companies filled with as many deeply experienced people as we have for decades in the field. Because of that, we raised far more money than we were spending, and we continued to spend about $10 million± for the past year. We had $96.8 million in the bank at the end of June. We have enough money to do all of the things we need to do as a company in order to realize the benefit of this therapy for patients.
Okay, great. Why don't we just end it there? Thank you very much, Steven, and appreciate your time.
Thank you. Appreciate it.