Prothena Corporation plc (PRTA) Earnings Call Transcript & Summary

Great. Good afternoon, everybody. Thanks for joining us here for the next session. I'm Matthew Harrison, one of the biotech analysts here at Morgan Stanley. Really pleased to have Prothena with us and the team from Prothena. Quickly before we get started, I just need to read a disclosure statement. Please note that all important disclosures, including personal holdings disclosures and Morgan Stanley disclosures, appear on the Morgan Stanley public website at morganstanley.com/researchdisclosure. So happy to have Gene Kinney, who's the CEO; and Wagner Zago, who's the CSO.

I guess I thought a good place to start. So you guys are obviously focused across both in AD or, I guess, I'd say neuroscience and amyloid science or amyloid pipeline. So maybe just give a little bit of background on the company and sort of the core underlying technology and how you got involved in both those sets of diseases.

Yes. No, thank you for the question. And Matt, thanks for the invite. Great to be here with you today. Yes. So for those that don't know Prothena well, we spun out from the Elan Pharmaceuticals back in 2012 as a public company. The focus of our science really is around protein dysregulation. So as we think about proteins that either misfold for genetic reasons or for wild-type proteins that misfold and aren't cleared correctly and then begin to aggregate and cause dysfunction, those are the types of proteins that we study. We look at how to intervene in that process, how best to target those protein aggregates for removal. And that takes you into a number of different disease spaces. So as you said, outside of the central nervous system, you can talk about amyloid diseases. So things like AL amyloidosis, ATTR amyloidosis where we have relatively unique and differentiated programs that don't so much target the protein production side, but rather the resident protein that's already deposited and causing dysfunction at the organ level. And we think there's a differentiated profile in targeting that resident amyloid. As we move into the central nervous system, it takes you into some very prevalent disease spaces, things like Alzheimer's disease, Parkinson's disease and a number of other disease spaces, where we have a number of active programs. Across our portfolio, you'll see representation of both peripheral amyloid disease, targeting agents as well as central nervous system disorder targeting agents. And we think a relatively balanced portfolio, where approximately half of our programs are partnered. We enjoy very productive partnerships with groups like Roche, Bristol-Myers Squibb, Novo Nordisk, and then, of course, a number of independent programs that we've elected to keep ourselves because we believe that we have an edge in terms of how to develop and ultimately commercialize those programs.

Okay. Great. Great. So maybe we could start with AD and then we'll come in to amyloid disorders. The sort of simplest question is we obviously have a handful of readouts coming out, 3 Phase III studies over the course of the next probably 12 months, depending -- or I guess how do you think about those outcomes? And how do you think about them influencing the work that you're doing in AD?

Yes. No, it's a great question. And it's a pretty exciting time if you think about how the science has progressed in terms of amyloid targeting agents. And what we're starting to see now, I think, in the literature is much more consistency that if you actually deliver enough anti-amyloid agent to remove amyloid to a sufficient degree, we're starting to see consistency in terms of that translating to clinical benefit. And we're seeing that across a number of molecules that target amyloid beta in a similar way that is targeting terminus of amyloid beta. So the way our portfolio is structured, and I'll come back to kind of how we think we're going to learn from these upcoming clinical readouts, is that we have a number of programs in the Alzheimer's disease space. So we have an amyloid-targeting agent in PRX012, which is currently in Phase I. We're excited about that molecule. We think it is a potential best-in-class molecule. And its target product profile is such that it's much more potent than the other molecules that are out there before it, which gives us an advantage, we think, in terms of how we can ultimately deliver that molecule and what that patient experience would look like. You've got a tau agent in PRX005, which is also in Phase I clinical trials. We think ultimately, tau and Abeta are on pathway in the disease process. So targeting each of these could be independently useful in terms of the treatment of Alzheimer's disease, particularly at the early stages. But ultimately, we also see combination as being a potentially interesting approach. And then we have a vaccine, an active vaccine that actually combines the best elements of both of those monoclonal approaches in PRX001, 2, 3, and we expect an IND in that next year. And we're excited about that not just from a treatment perspective, but we also think it gives us an opportunity to start to move that treatment paradigm into a secondary prevention setting. And obviously, we think an active vaccine is the right modality for that type of setting. So with these programs ahead of us, importantly, the lecanemab readout, the donanemab readout, which we expect next year, we think we're going to learn a lot about things like patient selection, is a little bit of a different approach, for example, between how Eisai is enrolling patients in the lecanemab trial versus how Eli Lilly is enrolling patients into the donanemab trial. Primary outcome measures. So again, lecanemab, the Eisai drug is actually using a CDR Sum of Boxes end point, whereas donanemab is using an iADRS end point. Thinking about that regulatory pathway, we've got both of those molecules now that have submitted BLAs, presumably for accelerated approval pathways. We'll see how the regulators actually think about that in the context of these clinical trial design readouts. And of course, importantly, reimbursement. We've seen that CMS last year had some, I think, interesting commentary around the Biogen molecule, aducanumab. As we start to see what we hope will be positive readouts on a primary basis with these additional molecules, it will be good to see how actually CMS starts to think about reimbursement in that space. And what's great for us and where we are with our PRX012 molecule, in this case, which targets amyloid beta, is we feel that it does have this differentiated profile. But we actually get to learn in each of those aspects, the clinical design aspects, the regulatory aspects, even the reimbursement aspects for, if you will, the cost of our Phase I investment, which is where we are today, which allows us, we think, to be very quick to foot to move quickly to get that therapeutic to patients as quickly as appropriate in that case. Maybe I'll ask Wagner, he's our Chief Scientific Officer, I can ask him to speak for a minute about why PRX012, we think, is differentiated relative to these first class of antibodies.

Yes. Maybe before you go, can I just ask a couple of specific questions related to the differentiation. So one, if you think about the differences between lecanemab and donanemab, gantenerumab, they're targeting different -- well, I mean, especially lecanemab and donanemab are targeting different areas, internal domain versus mid terminal. And so what impact is that? And then the second question is just with lecanemab, we know, in addition to aggregate forms, it's also targeting soluble form. So how do you think about those 2 factors?

Yes. So let me start with your last question, how we believe those 2 components can feed into disease. I think the data so far -- first, what is common among all these molecules, they clear plaques. That's very, very clear. And importantly, in fact, they can induce microglia to clear the plaques in the brain. In addition to that, they might have additional mechanisms. But the main mechanism of all these molecules given lecanemab is clearance of plaques. And we believe that the plaque is the main components driving pathology in the brain. It not only depends of the plaques and how it disrupts the axons and astrocytes and things that are around there, but it is -- all these plaques are the main centers of inflammation in the brain. When you look at microgliosis or astrocytosis, that's where the whole activity is. And we believe that by removing the plaque, it would alleviate a number of components that could be insult in the neurons. And the data is very clear. From all these different trials, there is a very strong correlation. If you cross a point of clearance of amyloid, very strong correlation between the level of reduction and the speed at which these antibodies reduce amyloid plaque versus clinical benefit as measured, for example, by CDR Sum of Boxes. So we want to maintain that mechanism as the primary mechanism. In addition to that, we do believe that soluble aggregates, that is one of the components the lecanemab binds here, it's also contributing to the disease. And we want -- when we design PRX012, we wanted a molecule that binds both, a molecule that's able to induce clearance of plaques by engaging microglia and, at the same time, neutralize the soluble aggregates. But it does that in a very important way, meaning that it would require us lower amounts of the protein, lower doses of the antibody in order to reach the same efficacy as any one of these antibodies. And that's what X012 is. It's an antibody with extremely high affinity to both pathogenic forms of beta amyloid, and it's going to induce the same mechanisms of action. With that very high potency, we open a new opportunity for us, which is the ability to dose very conveniently this antibody subcutaneously. This is what we had in mind in the very beginning. We saw that this first-generation antibodies once or twice a month, IV infusions, asking the patients to move into infusion centers once or twice a month, it would not be practical for a population like Alzheimer's. Having these antibodies dosed at home in a very convenient way with a small volume that doesn't require any specialized pump or sophisticated, that can even be administered by the patient himself or herself or even the caregiver very conveniently, that was the goal for X012. We think that we are there with a very potent antibody. We are absolutely amazed with all the quality of antibodies that we have created. And I think it goes back to what Gene was saying, how we created this molecule is also an integration of all the knowledge. We are integrating now all the knowledge that we can learn from the ongoing trials, patient selection and end points and time. But we also -- we designed X012 based on the integration of knowledge of how all these antibodies that target the end terminal portion of Abeta interact with the molecule. And the highest potency of X012 came from understanding how these antibodies interact with Abeta, with the sites in Abeta and optimize that interaction. So much so that once X012 binds to an amyloid plaque or soluble aggregates, the dissociation rate is so slow. And in fact, we cannot measure in the lab, it's almost a pseudo-irreversible bind. And that maintains, at any given point, a very high level of opsonization of the target. What that does is optimal neutralization of soluble aggregates and optimal signaling to the microglia are clear. So we're extremely proud of the molecule that we designed. And -- but again, taking advantage of all the learnings from all these different molecules that we consider them as the first generation. Extremely important for the patients. Again, we think that there is an opportunity to improve even more on top of those.

Okay. Great. Great. So maybe just last question on this. I think you've guided towards Phase I data next year. What would we expect to see out of that Phase I study?

Yes. So we expect the full Phase I data set next year and be able to start speaking about that. So that includes kind of the common things you expect from a Phase I study. So safety tolerability, PK, immunogenicity, with an antibody. But also importantly, we'd start looking at amyloid patent. And we'd expect to have some of that data next year as well.

Okay. Great. Great. So maybe just talk about tau. And I guess the biggest question on tau is we've obviously seen other tau antibodies in the clinic. They haven't had an impact. There's clearly a debate around extracellular versus intracellular tau. So maybe just give us your thoughts around tau and your approach to tau.

Yes. So maybe I'll start, and Wagner can come in with some of the science here. But I think one of the things, if you look at the class of antibodies that have failed to date, they've all been very consistent in 2 ways. First, they target the immunoterminus of tau. And second, their effector -- reduced effector function, so their IgG4 isotype. We think in both of those cases, that's probably from our data not optimal in terms of how to interact with tau. We take a little bit different approach at Prothena. We call it empirical epitope mapping. And so what we mean by that is if you kind of take tau, it's a big molecule. It's about 440 plus or minus amino acid. There's 6 splice variants. There's truncated forms. There's phosphor-related forms. So I think the idea of prespecifying which is the toxic form that causes all the problems, for us, is a bit challenging. So we take a more empirical approach. What we do is we make antibodies across the entire protein and we -- including post-translationally modified forms of protein. And then through a series of in vitro studies up to and including in vivo map studies, we ask the question of, if you interact with the protein in this way, does it give you more consistent and more robust effect in terms of ameliorating the negative biology across these different end points relative to targeting it in a different way? And it's a very empirical approach. But what it tells you is that if you're knocking down the biological effect that, that epitope, that way of interacting with the protein, first of all, it's available to interact with in its negative confirmation. And second, that it's sufficient to interact with it there in order to disrupt that negative biology. And so that's kind of how we approach it. And we ended up at a very different place than these first generation of tau-targeting antibodies. And we think this epitope selection process makes our approach a fair bit differentiated from what's come before. And then we can talk a little bit about the effector function as well. But maybe, Wagner, do you want to talk about the MTBR targeting?

Yes. Just to complement, we -- as many others, we also started testing antibodies in the end terminal portion of tau because that's the most immunogenic portion of the human tau if you are immunizing a mouse. And we were not convinced. So for us to move in to [ buy ] forward, as Gene said, it has to be a very robust effect and very reproducible, very consistent in different models. And that's not what the end terminal antibody showed us. And it was only when we walked along tau and hit the MTBR region of tau that we saw extremely high efficacy in multiple models with 100% blockade of cell transmission of tau. that it's only when we saw that we were convinced that immunotherapy against tau was a viable way. And that's where PRX005 came from, from a campaign of PRX005 is the best antibody MTBR tau targeting antibody that we designed. It specifically blocks 100% of that transmission in vitro. And we have demonstrated and presented at external, a very robust in vivo efficacy multiplying models. The selection of the epitope was the very first critical component, but we also demonstrated that the presence of effector function was also an important component. We had to have both. If you remove effector function from a very good antibody, you can, in fact, decrease the efficacy in vivo. We think that what's happening in -- with these antibodies is, first, binding to tau is preventing the cell-to-cell transmission. If it binds to a right epitope, I'm going to park this for a moment. I'm going to go back. But also you need fast clearance of that complex after it binds. Just binding itself is not sufficient. While we came to the conclusion that MTBR was the best epitope in tau to prevent this pathology from propagating, there was more evidence, external to Prothena, indicating that the binding of them to be a region of tau to a post-synaptic and sometimes pre-synaptic receptor called heparin sulfate is critical for the uptake of tau. So things then start making sense in our mind that PRX005 is probably competing for that binding. And by blocking that site in the tau, it's preventing tau for being uptaken by the target neurons. They could prevent a propagation of the disease by that way. But we -- again, we also think that the clearance process, once it binds, is not only the inhibition, but the clearance, very fast clearance of this complex is important for the mechanism of action.

Great. Great. Maybe we should switch to amyloid diseases, just to make sure we cover them as well. So maybe just start with AL amyloidosis and just remind people sort of what you're working on there and what the most recent updates are.

Yes. So AL amyloidosis is actually our most advanced program. We currently have a confirmatory Phase III study underway that's enrolling. We expect top line data there in 2024. So AL amyloidosis, for those that aren't familiar with it, is an orphan disease. It's a disease caused by overproduction of light chain by clonal plasma cells. So it's somewhat distantly related to multiple myeloma. Both of them are plasma cell dyscrasia. The main difference is in multiple myeloma, you're kind of worried about clonal expansion. Here, the clonal size can stay relatively small. It's about protein that's being produced. So you get this overproduction of light chain. It's not combining with heavy chain appropriately to make immunoglobulin. And it starts to misfold, bind on itself and then it can deposit in organs like the kidney, the heart. Heart is definitely the worst-case scenario in that with significant cardiac involvement, you'll see progressive restrictive cardiomyopathy that often leads to mortality in these patients. So what we had done was something a little differentiated. The current standard of care in this space is to use cytotoxic chemotherapeutic agents that target those plasma cells, try to kill off that clone, reduce the new protein being produced, which I think, to some extent, if you look at different biomarker indicators, is a useful approach, particularly more in mild patients. But in the more advantaged patients that have a significant amount of resident amyloid, it's akin to turning off the bathtub faucet once the bathtub is already full. And so where birtamimab, our molecule, is differentiated, it's designed to go after that resident amyloid, drain the bathtub, if you will. And notably, if you look at these more advanced patients, there's a categorization system in AL amyloidosis known as the Mayo Staging system. And the 2012 version of that goes Mayo Stage 1 to 4, with 4 being the most advanced patients. And so with those most advanced Mayo Stage 4 patients, what we routinely see is turning off the faucet is too late for those patients. Median survival is between about 4 to 6 months. And there's no plasma cell-targeting agent that's really ever shown survival benefit in that patient population. So we were conducting a Phase III study with an all-comers study previously with this molecule, it's called the VITAL amyloidosis study. And we stopped that study early. And unfortunately, it was at an early enough time frame. The more mild patients hadn't really -- we haven't really seen a lot of mortality events, not unfortunately, but unfortunately, from a data perspective. In the more advanced patients, however, and fortunately, we had stratified based on this Mayo Staging system. We noted a median survival in the control arm of just over 8 months. And in our treatment arm, we saw a hazard ratio across the first 9 months of 0.413. So approaching a 60% relative risk benefit on all-cause mortality. Obviously, that's notable. We saw that across other clinical domains as well in nominal p-values less than 0.05 on 6-minute walk test, which is a measure of function on short form 36, which is the measure of patient reported outcome. And so we did an analysis of these data. We wanted to be sure this wasn't an artifact of a non-alpha controlled analysis. So we brought some external statistical folks in that really took a hard look at the data to make sure that, that effect could not be explained by anything other than the drug. Once we were satisfied there, we entered into dialogue with FDA. That dialogue ended in a special protocol assessment agreement where it was agreed that at an alpha of 0.10 that, that would be sufficient with the prior data to register the drug in a confirmatory study. So that confirmatory study is the AFFIRM-AL study. As I said, that's underway. We're -- it's a global footprint on that study, we're looking to enroll about 150 patients, and we expect to have top line on that in 2024.

Okay. Great. Maybe we should talk about ATTR as well. I guess just given news in ATTR, maybe you can just talk about how the -- how HELIOS-B has any impact or not on what you're doing.

Yes. I think you're seeing it play out actually very similarly in TTR. So ATTR amyloidosis, different proteins, different place of production, but deposits in a somewhat similar way. So here, you're talking about transthyretin as the offending protein that's made in liver, choroid plexus, a couple of other places. But it typically assembles into a tetrameric structure, so homotetramer, 4 units of transthyretin is its normal form and underlies its normal function. So if you will, there's 2 ways to turn off the faucet in this disease. One is to stop de novo production. So these are the siRNA-based approaches, your antisense approach is where folks like Alnylam and Ionis come in. And then, of course, you can stabilize the homotetromeric normal form, which is where the stabilizers come in. So this would be like Tafamidis, which Pfizer now has. What we saw in terms of the first real focus on cardiomyopathy in this space was a 0.7 hazard ratio in a 30-month study with Tafamidis. And so they showed about a 30% relative risk benefit on overall survival. Interestingly, though, if you kind of look into that data set, you see 2 things that are very interesting. First of all, for the first 18 months, those curves don't really separate. It's at the later time points that they start to separate. The second thing you saw, which is probably related to the first, is that if you break down New York Heart Association Class I, II, III, and they excluded IV in that study, you saw that most of the effect was really built into the more mild patients, New York Heart Association Class I and II. P-value on the New York Heart Association Class III patients was 0.78. And so I think what we've seen is that as the newer studies have come on, you've seen kind of a little bit of a push towards more mild patients, which makes a lot of sense to us because if you kind of follow the discussion from AL amyloidosis, patients with higher resident amyloid that may be more at risk of early mortality, it's probably going to be harder to get that impact by turning off the faucet at least from a buyout -- from a conceptual perspective. And so our molecule in this space is a bit like birtamimab. It targets the resident protein for removal, drain the bathtub. And the idea there is that, obviously, it could work in parallel with turning off the faucet. But in these more advanced patients, where the treatments that are a little more upstream may be less effective, we think PRX004 may have a place, which could be unique in terms of how it's applied. And we entered into an agreement with Novo Nordisk in terms of further development of that molecule. They actually now have that molecule to develop. What they've said is they actually have already initiated a Phase II there. That's focused on New York Heart Association Class II and III patients. About 99 patients, 3-arm study, 2 different dose levels in placebo. And they're looking at, in the first instance, 6-minute walk, NT-proBNP, which is a biomarker of cardiac function. And so I think we're going to see -- we're going to have a good read in terms of these more advanced patients, maybe a little less advanced patients in New York Heart Association Class II, although they have an inclusion criteria that requires at least 6 50 nanograms per liter in terms of the NT-proBNP. So that would argue again that these patients are going to have pretty good cardiac involvement. So we're excited that, that study is kicking off. We would expect to see data in that probably in the 2024 time line.

Okay. Okay. Great. And then maybe in the last couple of minutes here, we should talk about alpha-synuclein.

Yes. Love that. So alpha-synuclein is a program we have partnered with Roche. We have a molecule there by the name of prasinezumab, which targets in a very specific way this protein. For those who don't know the biology here, alpha-synuclein is kind of the smoking gun with respect to cause and progression of Parkinson's disease. Genetic modifications that increase production of alpha-synuclein or increase the aggregation propensity of alpha-synuclein tend to lead to Parkinson's disease oftentimes with an earlier age of onset. And so we had, with Roche, run a Phase II study called the PASADENA study, where we ask patients to stay off symptomatic drugs so that we can get a good, clean look at what was happening on clinical scale. It was a signal-finding study. So we included a number of different scales as a composite end point. It turned out some of those weren't sensitive. They actually didn't really progress all that well in these early patients, these patients with early Parkinson's disease. But one that did progress over that 1-year period was something called the MDS-UPDRS Part III, which measures the motor dysfunction of Parkinson's disease. And in that, if you actually look across the 12-month period, depending on whether you did the reading at the site, the physician at the site level or did the reading or whether you did a central read, it saw a 25% to 35% slowing of progression over that 12-month period. We saw concurrent data, supporting data with smartphone-based applications that did kind of continuous monitoring of patients. We saw some very interesting signals on MRI looking at blood flow and presumably activation of key areas of the brain involved in Parkinson's. So we're very excited about those data. What ultimately was decided was to move into a Phase IIb study here, allowing patients to take L-DOPA as a concomitant treatment and, in essence, that more or less quadruples the potential patient population at launch. Interestingly, what Roche worked on doing was really looking at how to measure that primary outcome, measure using UPDRS Part III and ultimately settled on a time-to-event analysis. So using a 5-point worsening of UPDRS Part III as the event, they can actually look at time-to-event and see -- and ask the clinical question of whether prasinezumab slows the time-to-event or time-to-worsening in that regard. And of course, retrospectively, that data looks very interesting in the PASADENA study that's been published. And so we're excited about that in part because it brings UPDRS Part III, which is the key piece that really moves in that first year. We expect most patients would likely have an event before they required an adjustment of their L-DOPA dosage. So we think that, that's promising just in terms of the period of observation. And ultimately, a 5-point change in UPDRS Part III matters to patients. It tracks through to patient-reported outcomes, how they feel, measures of whether this is meaningful for them from just a daily life functioning perspective. So we think it brings in that feel function, which is very, very important.

Okay. Great. Maybe some last minute here, remind us from a financial standpoint, cash, cash runway and how you're financing the company?

Yes. So appreciate the question. I mean I think the kind of best way to say it is we can -- we have the cash to actually be able to afford any of the future is available to us, right, which is to say, we finished last quarter with $510 million. We're projecting, at a midpoint, to finish the year with $454 million. We have access to $365 million in partner payments over the next several years. And as we look forward to the different parts of our portfolio, what that, in essence, allows you to do is: get your top line data on your AL amyloidosis program; get your top line data on your Parkinson's program; your ATTR program; get to amyloid PET readouts on your PRX012 program; CSF MTBR readouts on your PRX005 program; and to get our Alzheimer's vaccine into an IND and into Phase I. So we think there are a lot of potential value inflection points that we can reach with our current cash. And obviously, we're always thinking about allocation and where we are from a cost of capital perspective.

Well, perfect. Great. Thanks for being here. Appreciate the time.

Appreciate the invite. Thank you for the questions.

Thank you.