Cabaletta Bio, Inc. (CABA) Earnings Call Transcript & Summary

Good afternoon, everybody. Thanks for joining us for the next session. I'm Matthew Harrison, one of the biotech analysts here at Morgan Stanley. Pleased to have Cabaletta here with me for the next session. Just quickly before we get started, I need to read a disclosure statement. Please note that this webcast is for Morgan Stanley's clients and appropriate Morgan Stanley employees only. This webcast is not for members of the press. If you are a member of the press, please disconnect and reach out separately. For important disclosures, please see the Morgan Stanley's research disclosure website at morganstanley.com/research disclosures. And if you have any questions, please reach out to your Morgan Stanley sales representative. So with that, I'm pleased to have the CEO of Cabaletta, Steve Nichtberger.

And Steve, I thought to start off. Just talk a little bit about the CAART platform and how that's different from maybe what people are a little bit more familiar with, which is CAR T and oncology.

Yes. Thanks for having us, Matthew. And thanks for investing your time to hear our story today. So the CAAR platform was launched out of the University of Pennsylvania as the basis for Cabaletta's creation. And it is really built around the CAR T, the CART19 success that was originally seen with the product that ultimately was discovered and developed at Penn, which became Kymriah. The work of Carl June and his team, among them, the lead in his lab, Mike Milone, who is one of our scientific founders and a dermatologist, named Aimee Payne, our Co-Scientific Founder, they got together right after the first CART19 paper was published, and they looked at the possibility of starting with the original CART19 construct and keeping everything the same, but changing only one thing, the targeting domain in order to specifically target the autoreactive pathogenic B cells that are causing autoimmune diseases. Now the CART19 products were developed for cancer. And as you know, they wipe out all B cells when therapy is effective. They have a really remarkable ability to completely eliminate B cells. And the thought of Aimee and Mike, when they cofounded the company with me, was that the specific use of the identical signaling chain, the identical approach to CART19 that was used with Penn modified only to change that targeting domain from the CD19 antibody fragments to the autoantigen that is being attacked in an autoimmune disease might allow the CAAR T to destroy only the autoreactive B cells that are causing an autoimmune disease. So pretty much everything across the platform of our portfolio is known to be safe and effective in the form of Kymriah. The only thing that we're studying is not a new manufacturing process, it's not a new vector or a new approach to manipulating T cells, but rather whether a specific change to the targeting domain can allow these CAAR T cells to do their work in a limited way, only for the approximately 1% or fewer of your B cells that are causing certain B cell-mediated autoimmune diseases.

Great. That's a great start, Steve. So maybe we can talk about how you're planning to prosecute that. So now you've got the idea, you've made the change, and I think you looked at PV as sort of the first potential indication. So maybe just tell people, why did you pick PV? And then I think maybe more importantly talk about what we know about the B cell population of PV?

Yes. So PV was really the obvious choice and in a trite way, I would say it was the obvious choice because the science paper published in 2016, which was the first effort, fruit of the collaboration between Aimee and Mike, demonstrating in vivo targeted and specific efficacy of these CAAR that were designed for pemphigus, that was the basis for believing that we can do this in patients. And so the tried answer is that we have the data, and we're moving ahead with it. But the real answer lies in the science like everything with our company. So the anti-DSG3 desmoglein is skin glue, the antibodies that are directed to your skin glue, which is otherwise known as desmoglein 3. The antibodies of desmoglein 3 are 98% to 100% sensitive and specific for pemphigus vulgaris, the mucosal form. And therefore, if you have the antibodies, you have the disease, and if you have the disease, you're going to have the antibodies. Depletion of these B cells that make the DSG3 antibodies has already been shown to improve clinical symptoms at least transiently, while those B cells are eliminated. So the incomplete B cell depletion by rituximab, which is both transient and fraught with complications and side effects, has demonstrated the proof-of-concept that if you eliminate all B cells, you can improve the clinical outcome transiently for patients with pemphigus vulgaris. In addition, when we thought about designing the first CAAR, it turns out that you need to understand the many different epitopes on an antigen, which are triggering the antibodies to form in the first place. And in the case of pemphigus vulgaris and about 2 dozen other diseases that are pure B cell-mediated diseases, those epitopes either are known or can be known. And so in the case of pemphigus, they've been published. So knowing the exact target that the antibodies want to attack, we designed our CAAR to include all of those targets, so no matter which patient you are in the pemphigus vulgaris population that has mucosal pemphigus vulgaris, we've got you covered with one product. That product has the epitopes that are going to be attacked in your body, but when our therapy is administered, the targeting domain is what contains those epitopes. And that targeting domain is going to be attacked by the B cells that have the copy of the antibody on their surface, which is called the B cell receptor. And that B cell receptor is going to bind to the CAAR T that we've administered and it will be destroyed if the preclinical models are an indicator of what will happen in patients. So because the disease is so well described and defined because the desmoglein 3 antibodies and the B cells that they create are indeed agreed to be the cause of the disease, this is not simply a surrogate marker, this is actually the cause of the disease, and because we understand very well, scientifically, exactly which epitopes can cause the formation of these bad antibodies for all of these reasons, the pemphigus vulgaris population was the obvious place to get into man with our first-in-man study, the DesCAART trial, which is currently screening patients for our Phase I study.

And can you just touch on safety? I mean is there any reason to believe that any of these epitopes will be present on something other than desmoglein 3 and any other safety risks that you've had to look for?

Yes. Thanks for the question. So it's interesting. We are using the same signaling chain. And let me just talk about sort of in a nutshell, efficacy and then spend a lot of time on safety because efficacy is target engagement for us. And it is clear that durable and deep responses have been achieved with exactly the same signaling chain and the same manufacturing process that we are using in our IND, in our ongoing clinical study. So we know that if the target is exposed based on in vivo preclinical models, a number of them that have recently actually been published in JCI, including an active autoimmune model, we know that target engagement should occur. So really, the only important outstanding question that we focus on with real intensity is the one you've asked Matthew, it's the safety question. So there are a number of ways to look at this. One is, when you choose a target, are there other unintended consequences of on-target toxicity? And the answer for the desmoglein 3 antigen is no. We've tested a thorough battery of Fc gamma receptor containing cells and there were no interactions. We've tested a membrane protein array of over 5,300 membrane proteins, which I'm told essentially represents all of the human membrane proteins that exist. There were no confirmed interactions. In fact, our preclinical program, when it comes to the safety profile of our product, demonstrated no signals whatsoever that would cause us or did cause us to put anything different into the design of our clinical program other than the normal prudent things that you would do when you're using a CAAR T therapy for the first time in an autoimmune disease population. There's another side to the safety question, which is something where we've spent years of thought with all of the greatest minds that we could bring around our table, ranging from Carl June to Jay Siegel, previously Head of CBER and Senior Biotech Officer at Johnson & Johnson, many, many others. And the key question that we and others ask is, these are patients who are not cancer patients. Will you see the acute toxicity that CART19 therapies have shown us? And if so, why? And if not, why not? And the core of that answer is that cancer patients are very different from autoimmune patients. Oftentimes, people think of them as 2 sides of deploying. In the case of cancer patients, you've got a burden of all of your B cell population and perhaps orders of magnitude more in the form of B cell tumor, leukemia or lymphoma, where you are going to try to destroy all of those simultaneously with an infusion of CART19 therapeutic modality. And when you destroy all those cells simultaneously, I think it's widely accepted that the burden of B cells that you are destroying is best predictor of whether you're going to see acute toxicity, particularly CRS or neurotoxicity. Let's contrast that with the autoimmune population. The patients who we're treating will have no tumor. So those multiples of those orders of magnitude, more B cells just don't exist in our patients. And when we talk about the normals, we don't have products that are designed to destroy 100% of your B cell normal population. We have products that are very specifically designed and have been demonstrated to selectively and specifically destroy only the targeted cells. And in the case of pemphigus vulgaris or the other diseases where we are seeking to cure patients, provide deep and durable responses. We're going to destroy, if we are successful, somewhere between 0.1% of your normal B cell population and 1.0%. And so if the tumor burden that is being destroyed, if the population of B cells, the quantity of B cells is the best predictor of the risk of acute safety toxicity, then our risk should be logs lower than what we see with the currently approved CAAR T products.

Okay. Great. That's I think that's an important distinction. So maybe that's a good place to turn to the study, which you're obviously currently enrolling and talk a little bit about the COVID impact there and then how you're thinking about getting that acute safety data which I believe you're going to look at it in 8 days now in the study?

Yes. So we have to answer your question right upfront here. So COVID, we were very fortunate as a company. We're spending on the order of $10 million a quarter, pretty stable burn rate. We have about $123 million on our balance sheet as of the close of the last quarter. So we've got plenty of runway to prosecute the opportunity. The current estimate is that our cash will last us at least until the end of the third quarter of 2022. And if we are not moving multiple products into the clinic and actively progressing the entire pipeline as we planned, in other words, things went sideways in some form or another, that cash runway has incredible flexibility because of our ability to have partnered with our manufacturing and all of its associated fixed costs in our first-in-man study. And as a result, the flexibility we have is to take our cash runway well into 2020. So that's sort of the first point. Second point. COVID caused us to have about a 6 months delay in pretty much everything we were doing. So we're very fortunate, though, because we had that cash on hand, and we were not spending at a very high rate when COVID hit in March, we shut down as did our partners at Penn and that shutdown lasted for about 4.5 months. The preclinical studies that we were undertaking and the discovery work we were doing, were all put on hold. The good news is we didn't have any long-term studies that needed to be stopped halfway through, for example. So every day that we lost, we lost a day, so about a 6 months delay. So we expect for our clinical program, we got back up and running as soon as the University of Pennsylvania started advancing clinical programs again after the surge of COVID. And the reason is we had fast track designation earned during the March-April time frame while we were shut down. And because that signifies the advance of an important therapy for an important unmet need, the folks at the University of Pennsylvania recognized that and put us into their first cohort, let's say, of studies that would move ahead. And so we have. We expect to deliver acute safety data from the -- at least the low-dose cohort of patients in the first half of 2021. You'll notice that that is 6 months delayed from what we said a year ago, and it's simply because of COVID. We've been, I would say, thrilled with a number of things. The ability to get the study started in the time line that we anticipated and to get our first site open. The ability now to anticipate that we will have at least 2 additional sites open and starting to be able to enroll patients by the end of this year. And we're very much on track such that in the first half of next year, we'll be able to deliver the acute safety data on those first cohort of patients. Now we plan to talk about patient data, only top line except in scientific sessions. However, we think it's really important that when we jump up and down, we should be sharing that with investors. And when -- what will make us jump up and down is a full cohort, dosing cohort, where we see target engagement, where we see a safety profile, that's information that we want to share in a timely way. But one patient doing particularly well or one patient having a side effect that does not change the time lines over the trial, those are things that are to us making of the sausage. And we're not going to make decisions on those sorts of data nor should, in our view, we'd be telling investors that they should be making investments on those sorts of smaller data sets. So we intend to share dose cohort level data as we see it on the safety and the efficacy or target engagement side. First half of '21, at least the low-dose cohort. Originally, we said 10-day data. It changed to 8-day. It is a trivial change. It is a distinction without a difference. And the reason we changed it is the folks who ended up finalizing the protocol wanted to do it at 8 days instead of 10, and so we changed to 8 days instead of 10.

Okay. Perfect. I guess 2 other things that come up a lot, obviously, at the time of acute safety data, how should we think about efficacy? And I guess what I'm really asking about here is what are the kinetics of depletion and your ability to understand if you're having an effect or not?

Yes. It's -- you, of course, know that we think about that a lot. So the target engagement is going to be best demonstrated by the measurement of the desmoglein 3 ELISA. That is a CLIA-approved ELISA, which is used in clinical practice. We also have a research-based one that has dilution protocol associated with it, so that we can get a better view of exactly what the levels are. There's a 3-week half-life to the desmoglein 3 antibody. And if we take 5 half-lives, we're at about 15 weeks, so we're about 4 months, let's say. So 4 months after the last patient is dosed in a dosing cohort, we should be in a position to have our first view into the dramatic reduction or not of the desmoglein 3 antibody titer. We have said publicly that we need 6 months to be able to have investors expect us to speak about it rather than 4 months. And the reason is quite technical, and we don't have time to talk about today. But we would expect, if we do see things at 4 months that are clear cut and uncomplicated, we will talk about it if it takes another month or 2 to parse apart some of the details, that's our base case, is that if we need to do that, we'll do that. So mid-next year, after we see the safety data, we can add a few months and expect to start seeing some target engagement data from the low-dose cohort at least.

Okay. And then the second thing is, we're all familiar with preconditioning chemotherapy in CAAR. Do you need to have that here? And if you don't, why not?

So I firmly believe that preconditioning is a prerequisite for the patient who has cancer being treated with CART19. With equal depth of belief, I don't believe that it would be wise for us to start our trial with preconditioning. And so we are not using preconditioning our first-to-man study. We're prepared to use it if we need to use, and we're going to follow the data. But here's why we don't think we need it. Preconditioning is used, it's theoretically understood to either debulk the tumor and get rid of a cytokine sink. Well, our patients don't have tumor. So we're not worried about that. We're to clear out the bone marrows to allow for engraftment of the CAAR T therapy, perhaps. Again, our bone marrow is not overwhelmed with tumor cells. And ultimately, you want your infused product to multiply and to engraft. Well, let's look at the autoimmune patient who doesn't have any tumor, has no need for cytotoxic chemotherapy. Perhaps they have a need for cytokines that would cause the cells to replicate and to expand after they're infused. Well, it turns out that in these autoimmune patients, you've got soluble circulating antibody. On the one hand, the pessimistic view is that these soluble circulating antibodies will bind our CAAR and will prevent it from becoming effective or worse yet, activate our CAAR and cause a CRS-like syndrome. The JCI paper that was published about 2 weeks ago contains an active autoimmune model, in which soluble circulating antibody is seen, is present at levels that are consistent with what you see in patients according to the experts who I've spoken with and to our scientific co-founder, the author of the paper Aimee Payne. And in that model, we were able to see target engagement with the appropriate cells overcoming the soluble antibody. But furthermore, and more importantly, maybe the soluble circulating antibody could be somewhat activating or somewhat inhibitory. But net-net, we see them being modestly activating and causing our infused products to expand, which is exactly why you use preconditioning. And so the autoimmune patient is a setup for success when it comes to infusion of the CAAR T product, CAAR product, which we know that the cells expand when they're exposed to PV patient serum. And as a result, we don't think we need preconditioning. We are prepared to use either preconditioning regimens that are cytokine enhancing or that are cytotoxic chemotherapy oriented, depending on the data. We don't think we're going to need either, and we need the data in order to decide.