Cogstate Limited (CGS.AX) Earnings Call Transcript & Summary

Yes. Perfect. So that said that up perfectly, Case to talk about trial, the [indiscernible] 345 trial and Lilly's trials Trailblazer-Alz-3, which is taking those same drugs into an earlier population, right?

Maybe I'll start with why are we even doing this if someone doesn't have symptoms. And the reason is, as Akash mentioned, there are already -- there is already pathology in the brain. So amyloid is a really sticky protein. And it starts clumping together and laying down plaque in the brain maybe 20 years before you ever have any symptoms. By the time you have symptoms, you've already damaged neurons and lost a lot of neurons. And there's this whole cascade of stuff that happens. -- including how starts building up and killing neurons. And so the thought is, and you've seen in all of the state, just like Akash mentioned, once you already have symptoms and you try to put the drugs on, we haven't seen that much efficacy. I mean what is 25% or 35% slowing. And as I mentioned before, people are still declining. So -- but what if you could get to it and get it out of the brain before someone started having symptoms. Could you slow down the amount of time it takes them to get to Stage 3, the symptomatic stage or maybe could you even prevent them from ever getting there in their lifetime. And that would be the ultimate hope. So both Lilly and Eisai have ongoing clinical trials right now where patients are in this state. They've got -- and we can't really call them patients because they're not sick yet. But participants are -- they have amyloid in their brain. So they have the disease just like, as Akash was mentioning, is dads prostate surgeon, if you have a high PSA, you have and you get a biopsy and there's prostate cancer there. You still have prostate cancer even if it's not causing you any problems, even if you didn't have any symptoms, you don't wait until someone gets symptoms to go in in there and get it out. So I think of Alzheimer's disease just like this. So these people already have the disease in their brain. They just don't know it because we haven't been doing TSAs every year on the brain or some other biomarker. So you identify folks who've got the plaques building up in the brain, but they're still cognitively normal. And those are the folks that have been enrolled in these trials. The trials are similar and that they're trying to get at the same idea, right? Can we delay progression in these patients, but they're designed very differently. So the ASI trial with Leqembi basically enrolled these participants that we're talking about, these preclinical participants, and everyone is going to get 4 years of treatment. And their outcome is a cognitive composite basically, anytime you take a whole bunch of different cognitive tests and turn them into one number, that's a cognitive composite. So this trial is really similar in design to the standard Alzheimer's disease trial of all the ones before. It's a change in a score from baseline, whether that is a change in a composite score or a change in CDR cell on box score. It's just the usual traditional methodology. The Lilly trial Trailblazer-Alz-3 is very different because in this trial, they are going to treat participants with donanemab or placebo until a prespecified number of participants progress from non-impaired to impaired or from cognitively normal to MCI or dementia. And that's going to be measured by a CDR score, but it will be a global score. And so in this kind of outcome. So it's a time to event outcome. They're basically going to watch all of these participants and assess them every few months and then count up the outcomes as they start progressing. And so some people will be exposed to the drug for a really long time, maybe the ones you were first in and others might have been randomized relatively late when they hit the number of endpoints and could have only a little bit of exposure. So not everyone is going to get the same amount of time on the drug. It just depends on how fast people are progressing in the theoretically, hopefully, in the placebo arm in the treatment arm. So those are 2 really different approaches to the same question.

Thanks, Kaycee. And so Akash, this is where we start to get into the juice of this story. So for people who have been listening for 30 minutes, this is the time to pick up your ears because isn't where it gets interesting because you've been following Lilly closely, you've met with the management team. And you have some thoughts in terms of when you think you're going to see data from the Trailblazer-Alz-3 trial? And the implications for the industry if that is to read out positively. So when do you share some of those thoughts?

Yes, absolutely. And look, I don't -- we have published a deep dive on this. So -- we've done our own kind of separate analysis. So I'm going to first kind of communicate what we've heard with Lilly, I wouldn't overlever on that and then also what our own analysis as [indiscernible], but here's the way that I would kind of think about it. If you talk to that management team, this -- both of these studies, by the way, so the Hersey, I think, started rolling Troll 3 started to enroll, I think, 2 or 3 years later. Both of those studies on clinicaltrials.gov are set to read out in 2027. Now I think the hints that we've got from publicly available comments we take from Lilly are the following: -- number one, we have the chance to actually host the management team last year and asked them like, hey, if you think about opportunities that could diversify meaningfully away from GLP-1s, they have talked about Alzheimer's as don't give up on this indication. And the analogy they basically said is when we talk about obesity, the dogma was obesity is never market. That's what everyone told us. And that's Lilly's initial therapeutic window for [indiscernible] appetite was just not very compelling. You had high rates of vomiting, et cetera, et cetera. So you got have a clear good window and a huge theoretical population, but was not able to actually gain reached. And then with the right titration, we all know what's happening with GLP-1s in terms of the market. The management team said think about metamalogy with Alzheimer's. And we can think about treating earlier-stage patients in the right place at the right time, the risk benefit might be meaningfully different. So I think that was the first time that, hey, something might be really interesting here. And frankly, I think we missed it because I'll be honest with you, when I don't think -- and Kaycee and Rachel, tell me if you agree or disagree, is there a real need to measure to [indiscernible] in symptomatic patients right now. I didn't see a compelling difference in terms of high to versus low town in their initial data set. Remteririb, right, like they're back up a beta drug. The first note we wrote about it was, why do patients need this? It seems to have similarly high rates of RGA, but it just is subcu is that really what we're looking for is brain shuttle really where the headed. So I don't think I really understood Lilly's long-term approach here. And when you actually start applying it to preclinical, things make sense a bit more. Why start measuring talent because you're trying to establish a link between A-beta positivity and centralize plaque and [indiscernible] levels. And you want to do it in the late-stage patients, but eventually want to do at earlier stage. And that means [indiscernible] is not competing against Abeta, they're actually 2 sides of the same coin. That's a very interesting paradigm shift different than what I think we've thought about historically. Number two, it's thinking of maybe to this year and then like what's comments we've been hearing, the added commentary that you've gotten from earlier this year is that a study that's supposed to read out at latest end of 2027 might read out earlier than expected. And some people have even speculated including us, that we could get a readout on this in the first half of 2026. Now what hard data points can we use to point to that. Well, what Lilly pointed to when we hosted them recently in London was that we actually have like let's put this way. Imagine you and I have an equation, total number of events, and then we have what we expect a placebo and then what we theoretically expect in the treatment arm, right? Now these are usually dangerous analysis to do unless you have a certain set of circumstances. Lilly has pointed to the [indiscernible] study. That is with their prior [indiscernible] in a similar population. It was in a preclinical Alzheimer's setting. Now that study basically accrued events about 2x more quickly than what Lilly had initially expected. Now let's go back to that equation analogy. If you think about the underlying baseline characteristics of the Trailblazer-Alz-3 study, it's actually quite similar to that solanizumab A4 study. So let's go back to the equation. You probably have a very good sense on what your event accrual is for placebo. And this study is going on a blended basis. So you have total number of events, what you would expect at that point in time for placebo, and you probably have a sense from indirectly on what the treatment arm is doing. I think that may be fueling some of their excitement in this population. But I also want to be clear, and we should be humble here, there's literally no data in this earlier stage population. We don't know. Like there's cuts that we've seen and Brad and Kaycee and Rachel and I were talking about the webinar started, early due to the Eisai published with low [indiscernible] patients ectopartive improvement with Lamb like we see in that we can have a different therapeutic window here, but we don't have solid information on what the cognitive profile will be in this less advanced cohort of patients at this point. Now -- what do we have on efficacy? Well, if we actually do like kind of a retrospective analysis in the studies in mild patients with mild dementia, what we have seen is that the lower plate you have, usually the better you actually do on these drugs. And also, you tend to have lower REA rates as well. So [indiscernible], you actually do have a better risk-benefit profile. But again, we don't know about that in the trailer also repopulation, we have to basically wait and see. On safety, this is one of the great minds I think our team found was that actually Lilly has presented on a blinded basis what they [indiscernible] rates on. So remember, in their current population of these mild cognitive patients, they have an RO rate of about 25%. Lonbea little directionally lower. What we know from Talara based on Lilly's owned data disclosures is that the ARIA rate is -- we backed out about 11% for guessing in the treatment arm. And they're also enrolling another arm with slower titration, which could theoretically get that REO rate down to 5% or lower. But what I can tell you is, again, based on publicly available information, Lilly is talking about the study reading out sooner than expected. They are setting the stage for this preclinical Alzheimer's population. When you think about return correct or next-gen drugs and the fact that they're establishing Tau as a biomarker, not only in the preclinical setting, but actually in the late-stage setting. You and I can make some educated guesses on what they're seeing on a blinded basis that can help us inform what they're seeing from an efficacy perspective. And then on safety, they've actually disclosed their blinded [indiscernible] rate. And you and I can basically say, that in this earlier stage population, they're getting about 1/3 of the [indiscernible] rates that they had in the 1 trial. So I think those are the reasons for excitement. Building on what Casey said, and I think is really important. What Lilly is doing is kind of threading the needle where I think there are going to be 3 really critical questions that need to be answered here, and we're not all going to get the same -- these [indiscernible] at the same time. Answer number 1 -- actually, maybe 4. Number 1 is like really what is the safety profile? And can you substantially get the [indiscernible] rates down to the range where someone who is healthy and maybe what you're now introducing risk, can you get into -- you're already a year rate at a rate where a 50-year-old healthy male or women feels like, hey, I'm motivated to do this despite the theoretical side effects that I might be able to get here. So therapeutic window is, I think, critical. But I think you're -- I think what you also need to answer is the Lilly study has 2 questions. One, as Kaycee mentioned, is a time-to-event analysis. That's the primary analysis, which basically means if you have true back-to-back measurements of even mild cognitive decline, you're considered having that. So when Lilly [indiscernible] my suspicion is when Lilly releases their first kind of the data, the headline will be donanemab lowers -- prevents Alzheimer's progression in 70% of patients, 30% of patients, whatever that number is, but you're going to get the first cut of, hey, x percent -- 70% of patients had none of their Alzheimer's progress. 50% of their patients are done in Alzheimer's progress in this period of time. What I would argue patients also need to know is, well, what if you launch new we follow these patients over an extended period of time, how would their CDR sum of the boxes, how does their cognitive decline change if they had their platform moves in such an early stage. That is the other just as critical question, frankly, it might be even more important from a payer perspective, right? What is the progression? And do you put these patients on to a different kind of trajectory if you treat them in this early stage? I think the last question, and I think this is just a provocative is can you move into a treat and extend paradigm, right, where perhaps you clear plaque and then you stop. And how does that change, I think payer adoption becomes really fascinating, right? Like I cover Lilly, a huge debate with these GLP-1s is, well, if I'm an insurance company, cost might lose 30% of his weight, but then he's going to change jobs and now Cigna is going to be getting the benefits not Anthem, right? But adherence, for these drugs for any of these outpatient drugs, including GLP-1 is a massive issue, and it totally changes the risk benefit for filing for insurers. Well, what if Lilly or Eisai can basically say, like, hey, you don't have to be on these drugs forever. In fact, we will get you to a point of plaque clearance and then we'll watch what we do, similar, again, to prostate cancer, where we're going to look at your PSA velocity and then if we need to treat you, we'll treat you. But otherwise, you will have no meaningful amount of disease progression. We don't know that answer yet, but I suspect that in both of these studies, there will be maintenance portions of these trials, which will test that question of patients who are dosed more on a continuous basis versus those that basically get to flat clearance and then they don't -- they watch how they progress on a go-forward basis. These are enormously important questions. And I think what's really critical is that initial top line data set for Lilly will only tell us 2. One, which is the general risk benefit profile, right, your REA rates? And then number 2 is the percentage of patients who have no Alzheimer's progression, but again, these other 2 questions, which I think payers are going to need are going to probably be a year or 2 away from that top line press release. And what I would just say, again, I don't cover Eisai, I can't give you any stock [indiscernible], but I think Kaycee is absolutely right. If you think about how those companies have enrolled patients, the way that Lilly has enrolled patients, about 40% of their patients are arguably -- they have high [indiscernible] levels that they don't have cognitive decline, but they are probably patients who are about to progress into cognitive decline or maybe already show early stages. Why would you do that? You want to accrue events in a timely period of time in a very quick time line versus taking maybe 7 years, which potentially the Eisai study might take? And then also -- so if you look at the Lilly versus Eisai studies, the baseline severity, both from a simple of plaque and tau levels are a bit different in both populations, right? And that can lead to differences in event rate accrual that I think are important. And again, those are really, really critical studies, but they are answering suddenly different questions.

Yes. So thank you, Akash, and Kaycee. That's -- so to summarize, what we're saying here is that there's a potential short-term catalyst in respect of Lilly data, we'll wait and see. But there will be a catalyst at some point. So there's a question around time line. But if that reads out positively, we're talking about a complete change in the way that we're treating Alzheimer's disease. So Rachel, this is where we're going to lean on you to sort of bring this home for everybody because I think the question from an investor point of view is that's all very interesting. What does that mean for Cogstate. So -- and for those who are unaware, Cogstate heavily involved in both the running of a head 3, 4, 5 and the Trailblazer-Alz-3 study. And these are complex studies to run, right? There's some complexities that we've had to solve on the art as it were.

No. No, it's absolutely true. I think over the years, we've gained a really unique set of experience in managing endpoint data quality in these programs. And I think as we see more trials move into the earlier stages of disease or we see these early-stage trials start to go after combination therapies and new -- answering those additional questions that you point to Akash it's an exciting area where we really get to apply some of that understanding about how to conduct these trials and how to manage the endpoints. And I think one of the main reasons that our Cogstate really shines in the space, is because of how we were born out of digital endpoints. And so it really creates the need to have more sensitive measures to measure more subtle change. And so we're seeing sponsors now adopting digital endpoints in a way that they had not previously in some of the trials. So we're seeing this not only in the prescreening area, where we're using them to identify potential candidates for trial participation. but also using them as endpoints in the trials. And so this is something that have been done in a more limited way previously. But certainly, when the drugs in trials are addressing a later stage of disease, the potential for the digital end points. was diminished. And now it's really a time where the adoption of those is a focus area. The other place is taking the clinical measures and really trying to ensure the highest data quality possible so that we can increase signal detection by decreasing noise and decreasing variance. And so that's another area where Cogstate has been able to add a lot of value. So we've been able to work with the sponsors to ensure that the readers are trained to a really high degree and then that they're monitored ongoing throughout the trial to ensure that there we're managing that variance of any readers throughout the life of the trial. And so we're doing this on a global basis. We've got clinicians in over 30 countries where we have clinicians that act as that second set of eyes and oversee the work of the [indiscernible] is the rating -- the clinical progression of the patients in the trials. So those are 2 of the ways that we really help support this -- the unique needs at this stage of the trial. And then the third one is with respect to the the decentralized opportunity. So the opportunity to support these trials in a new way. When we go into earlier stages of disease where the -- the participants are well and younger. They have different needs. And so it's -- and more capability. And so we have this unique opportunity to make the trials a bit more patient-centric by offering the ability to participate from afar. And so I think that's something that's really opened up because of this this dynamic of going earlier in the disease. And so [indiscernible] has unique experience providing that what we call central rating services, where we are acting as the clinical experts conducting the clinical assessments and the psychometric assessments, the [indiscernible] assessments. So by combining sort of the Cogstate digital endpoints with this global capability to provide centralized administration of the endpoints. I think it really allows us to add a unique and meaningful impact to overall endpoint data quality so that we can help ensure these trials are as conclusive as possible.

And Rachel, as we -- just to pull on that signal over noise thread a little bit, is we if we do see success in these trials and therefore, an expectation that, that's where R&D heads into a much earlier stage patient population. The role that we think the technology can play in terms of that aero detection and increasing signal and decreasing noise. Do you want to just 30 seconds on some of the things we're doing there to sort of try to advance the technology.

Yes, absolutely. So I think there's 2 ways that we really will continue to innovate to offer advantages with respect to endpoint data quality and signal detection. One is through new digital endpoints. And so developing endpoints that are sensitive to even the earliest changes of cognition going even earlier because we believe that these trials will keep going earlier, first into stage 2 and then into stage 1. And so as we go earlier in the disease, there's going to be the need to have more and more sensitive end points. And so that's one area where we're investing in new digital assessments that are tapping into things like learning over memory. And so that's, I think, a really exciting area of the business where we see a lot of possibility. And then another is in terms of the conventional assessments, the clinical assessments. You'll still need a clinical assessment of someone's functioning in addition to some of the performance-based outcome measures or the cognitive measures. And so in that area, we're developing into quality monitoring capabilities using advanced algorithms in AI in order to detect errors. And so this is the whole bolster what we're doing with our clinician network the one that I described doing central ratings around the world. We really want to make that more tech enabled. And so we are developing capabilities to pick up on error by looking at data monitoring and also looking at the administrations themselves. So we take the audio recordings of the sessions, and we are able to analyze those with human reviewers. And -- and what we're working on now is the capacity to add AI to that to help strengthen what we're doing with our human reviewers. So maintaining that human in the loop, but adding adding a sense of automation to that and ability to scale that. Because what we see with these trials, the sponsors have to make a trade-off decision between how many assessments can be reviewed by that second set of eyes usually focusing on the primary endpoints mostly and usually employing some sort of sampling strategies. They can't really look at every single data point for that second set of eyes. And so by employing new technologies and new innovations, we're able to do a 100% data quality oversight. And that's really the goal is keeping that that human in the loop, but allowing them to do it in a much more efficient way so that we have the ability to look at every data point and ensure that the administration was done in the right way and school in the right way.

Yes, it's really important. And so just to contextualize this, we're going to open up to the Q&A in a second. But just to contextualize this investors. I think that Alzheimer's disease represents depending on which year, but it's roughly about half of our business is focused on Alzheimer's disease. We've been long Alzheimer's disease for years. We've been saying that whilst we do want to expand our offering into other indications, we fundamentally believe that the R&D spend in Alzheimer's disease is going to increase. And we have been positioning ourselves for a number of years to be at the forefront of these prevention trials. And we invested heavily in the A4 study that Akash mentioned earlier and really lost quite a bit of money on that, but to position ourselves to be the the dominant player, if you like, to support companies like Eisai and Lilly as they moved into the secondary revention trials. And I think we're really well placed there -- should the data readout as we all hope it will, that I think we're going to see -- as the commercial opportunity increases for the pharma companies, we are going to see an increase in R&D in this space, and I think we're really well positioned to support companies as they do that. So with that, we're going to open up to Q&A.

Brad, I just want to add 1 thing that we're internally we're actually really struggled with this. And I'm curious, maybe for your team, how do you think about it. Tau seems to be relatively -- like the correlation between central is a flat and tau levels with BTA217, which is the biomarker that I think will make it an established [indiscernible] studies work is relatively sensitive. It's not perfect, but there's a pretty solid correlation. What I struggle with is, okay, once you've removed the plaques, does your tau levels actually decline, not necessarily, and that actually becomes really interesting. And obviously, I can't make any stock opinion or anything to your company. But what I would find quite interesting with your software would be as we move into a treatment paradigm or let's say, a patient is 2 years off. And -- they don't want to get a pet scan every single year, right, that because really were some sure payers are like, well, look, the burden of actually requiring a PET scan annually for some of these treatment extend patients is going to be very hard to do. But Tau [indiscernible] might not be a very sensitive marker once your plaque has already gotten removed, Could the Cogstate test be actually a great way to monitor those patients for evidence of cognitive decline in a way that you can do at home via your software. And I think that's actually a really interesting question. Because, again, it's great. I think there's a good correlation when you're looking to treat patients initially but we actually don't see [indiscernible] levels change once the plaque is actually removed. And that is, I think, a big question we're internally struggling the way. So I actually hand it off to you guys. How do you think about that emerging opportunity.

Kaycee.

Yes. Maybe it's an opportunity to clarify confusion around PTL 217. It actually has nothing to do. It's really not a marker of tow at all. So [indiscernible] is a marker of amyloid. And as you said, Akash it correlates really well with amyloid plaque. So you can basically look at a pTau217 level and estimate what the panel of amyloid plaque would be. So pTau217 is actually a marker of amyloid. It correlates very importantly with Talpa signal. So when you -- so we -- even though it's called -- it is a Tau molecule, it's a fragment of Tau, but we shouldn't think of it as a marker of Tau. -- it's a marker of amyloid. And actually, once you've taken out the amyloid, you'll see pTau217 levels fall. So it may be that in the future, we will monitor people with pTau217 to know when they need to get their booster. So I see a future where -- and one of the reasons why I'm so excited about trontunumab is that I see a future where you could get 1 or 2 doses, especially in the preclinical space, where you don't have 200 centiloids of amyloid, you have 60. One dose of that will probably drive you down to below normal. And then maybe you just come in once a year and get your booster when you get your flu shot. So I see a day or people will track pTau217, and Eisai also released some data at this AAIC conference showing that the pTau217 level drops to normal threshold levels and then off treatment, it starts climbing up. And they're using that as their rationale for needing continued dosing. Anyway. So I would -- and I'll stop here because I know they want things, but the only way we can really measure Tau right now essentially is with a [indiscernible].

Yes. But a role -- and we're guessing at this stage, but we think perhaps a old to pay in terms of ongoing monitoring of cognition, especially using digital assessments that can be done, self-administered via smartphone and those kind of things that we keep monitoring that. Let's open up really quickly to questions because we're coming up to the top of the hour, and we're going to run out of time.

Yes, Brad. So I'm going to actually combine the 3 questions that have come through, and I think it will give you an opportunity to sort of summarize So -- and because some of this has already been covered in the last sort of 15 minutes. The first is around how Cogstate is positioned to capitalize on the growing investment and regulatory support for AD drugs. Particularly in light of the FDA approvals. The second is around -- you've talked about diversifying your clinical trial portfolio and just what percentage of your clinical trial contracts are still AD, and if you've seen the acceleration in opportunities. But in the context of the third question, which is just around what happens if the Trailblazer-Alz-3 doesn't read out like expected and what the implications on Cogstate could be. So I think that's a nice way for you to round it out.

Yes. So I think -- I mean, just to bring all that together. I mean, so as I said before, Alzheimer's still represents around half of our business. It's -- I think there's no doubt that I think with [indiscernible] or the outcome of Trailblazer-Alz-3 that the focus from all of the pharmaceutical companies is to go left to screen to go early in the disease phase. I think it's fair to say that that there's a desire and a hypothesis the earlier treatment will lead to better health outcomes for patients. And I don't think that should Trailblazer-Alz-3 readout negatively or not statistically significant. I wouldn't believe that you'd see a major departure from -- for that investment. And I think that this is where the Roche data, the case you talked about with the shuttle technology, which is leading to faster clearance of amyloid with a better safety profile. I think that's where and Roche already announcing that they're going to move into that same earlier-stage patient population with that drug and that technology, I think, just reinforces what I'm saying is that this is an area where people are going to continue to invest in. I think the final thing I'd say is just in respect of the -- what is the opportunity outside of clinical trials for Cogstate. And I think that's something we're still trying to solve that question. And I think -- and it's really going to be depend on the data and what do we see. But fundamentally, we believe that there's a role for measurement of cognition. If that can be if that can be sort of friction free, right? So if it's low cost, if it's self-administered, if it's delivered via smartphone and it provides meaning for we believe that, that's -- there's a real opportunity for us there. And I think -- and we're really well placed. As Rachel said, we're looking at different indications, not just around cognition but around learning and developing assessments in that area that are really sensitive to that very early stage of decline. And I think that's the critical aspect. It's not just what we're seeking to do here is not just technology, but is to bring the very best science to answer the questions of how do we measure those patients in that very early-stage disease. And I think we're really well placed there in terms of the excellent scientists that we have on our team. So we're at the top of the hour. So with that, I'm going to thank everyone for your attention. I really want to thank Akash especially for joining us from Jefferies; Kaycee Sink, Chief Medical Officer of Cogstate; and Rachel Colite, who's our Executive Vice President. I want to remind everyone that our Cogstate's annual results will be released Friday, 22nd of August. There will be a webinar there. The registration is now open on our Investor page of our website. Our Annual General Meeting of Shareholders is scheduled for the 16th of October virtually and in person in Melbourne. And then finally, Friday [indiscernible] November, we're going to have an in-person Investor Day in Melbourne. The hole of the Cogstate executive team will be there. We're going to have some special guests, including some people who can talk from a customer point of view to really allow our investors to dig deeper into the Cogstate story and what we're trying to build here. So those registrations are also open. So that's Friday, 7th of November in Melbourne. Thank you, everyone, for your attendance, and we hope this has been helpful for you.

Thank you.

Thanks so much.

Thank you.