Denali Therapeutics Inc. (DNLI) Earnings Call Transcript & Summary

Great. I think we'll get started with the next session. I'm Matthew Harrison, one of the biotech analysts here at Morgan Stanley. Really pleased to have Denali with us for the next session. Briefly before we get started, I need to read a disclosure statement. So 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 with that, really pleased to have Ryan Watts, the CEO of Denali with us. Ryan, maybe just like an overarching question to start out, which is, I think, since Denali came public, we've seen a lot more companies come into the neurodegeneration space. I think everybody is not necessarily focused on the same targets or even focused on the same disease areas, but just broadly, as you think about the scope of development in neurodegeneration, what do you think are the key features that differentiate Denali and your approach?

Yes. Matthew, thanks for having me here, and thanks for that question. It was a really interesting time 7 years ago when we founded Denali in neurodegeneration. There was a large number of companies that were -- especially big companies that were exiting the neuro space. And we've noticed really in the last 2 to 3 years that, that started to change. And I think that probably the biggest driver behind that change is success in some rare neurological diseases where the genetic understanding is stronger, where there's, for example, monogenic diseases. And when we launched Denali 7 years ago, we focused on 3 major principles. The first is what we call the degenogene pathway. So genes when mutated that cause neurodegeneration. The second is brain delivery, all about getting medicines across the blood-brain barrier. And then the third is biomarker-driven development. And what we've seen over the last 7 years is, we've taken 10 molecules into the clinic. We have 7 active programs this year alone. We now have 3 late-stage programs with potentially registrational data kicking off those programs now. So it's been pretty incredible over the last 7 years to make that progress. And we've seen, as you pointed out, now more companies being founded focusing either on broader neurodegeneration, rare neurological disease and even blood-brain barrier technologies. And I think when a field starts to mature, that's what you should expect to see is this proliferation of approaches. And I think for us, in particular, it's been very exciting to pioneer a lot of the blood-brain barrier work, both before Denali and now here. And I think the clinical data that we've -- over the last 2 years that we've seen with our Hunter program has laid that foundation at least for large molecules. And the same can be said for small molecules, for example, our LRRK2 program in Parkinson's disease. So I used to get the question a lot, 7 years ago, how are you different? Everyone's failed in this space. And my thought back then was that we're going to differentiate based on data, and I think today, we've now generated a lot of data. We have a number of partnerships, and there's a lot of hope and -- either from rare neurological disease where we can get, for example, enzymes across the blood-brain barrier or for broader diseases like Parkinson's where we're going after a specific genetic targets. And so that's -- now it's about that translation. It's now about seeing the clinical benefit from some of these discoveries.

And I mean you've got small molecules. You've got large molecules. In your mind, does it matter? Or is it just fit for purpose, and you'll use what works?

Yes. I think, again, going back, historically, we wanted basically a part of our portfolio that was like codependent on success. So building a platform that would feed to multiple programs, that ends up being our transport vehicle technology. But we also felt, there was a number of targets that were ideally targeted with small molecules. That was actually the first wave of discovery at Denali were these small molecule programs, and we continue to invest in our small molecule portfolio. It's roughly split half and half, and that -- the concept of fit for purpose really comes down to the molecular nature of the target. Is it targetable with a small molecule so we can -- that can be given orally once a day or twice a day? Probably, ideal over an IV therapy. And so we see that balance between small molecule and large molecule as being a key differentiator for Denali as opposed to being focused on, let's say, a single platform or a single target or therapeutic area.

Okay. Great. Great. So I'm going to maybe take the portfolio in a -- I don't know if I call it backwards or frontwards approach, but I want to talk about ETV first, and then maybe we can talk about some of the small molecules. So I see ETV as one of the sort of the most interesting things you have right now, just given you have an initial program that I would say is derisked, and now you're starting to invest behind that in many more enzymes. So can you maybe, just for everybody's benefit, talk about how derisked do you see Hunters? And what you see the risks are there and then your philosophy for investing in maybe MPS or other diseases beyond that?

Yes. So for those of you that are new to the Denali approach, the TV stands for transport vehicle, and it's a technology that utilizes natural iron transport at the blood-brain barrier to basically hitch a ride. The E is for enzymes. So we have enzyme transport vehicle. We have antibody transport vehicle, oligonucleotide transport vehicle, basically utilizing this transferrin receptor approach to get across the blood-brain barrier. And when we first started working on this technology, we wanted a clinical proof of concept in an area where we had a robust set of biomarkers and the potential for clinical translatability. And so really the MPSs or lysosomal storage disease was really the first area that we approached because we -- there were approved enzyme replacement therapies that we believe if we could cross the blood-brain barrier, we could essentially solve that tissue. They were already working in the periphery, but they needed to cross the blood-brain barrier. And so one of the first programs we began working on this and now a handful of other MPSs is MPS II or Hunter syndrome. And the primary substrate for that enzyme is heparan sulfate, and it had already been shown that reducing heparan sulfate peripherally actually measured either in urine or in blood was translating to a clinical benefit with peripheral manifestations. However, about 70% of these patients are developed neurological disease and essentially, neurodegeneration. In fact, they develop normally until about age 2 and then start to fall off the developmental curve. Again, about a hand -- about 70%, so a vast majority of these patients. And so our idea is essentially to get the enzyme across the blood-brain barrier using this transport vehicle technology. And the first data we generated was actually, at this point, still unmatched in the clinical landscape, which was basically a normalization of heparan sulfate just after short-term dosing. So the magnitude of response was very robust. Now what we've shown and very recently is that sustained reduction. Now over a year in 27 patients, which is correlating with what we see at least in the open-label study is clinical benefit. And I think really importantly, we've done a lot of work to show that heparan sulfate in CSF is essentially 1:1 related to heparan sulfate in cells. And in fact, the enzyme is only active inside a cell. It has to be in an acidic pH. So when this enzyme crosses a blood-brain barrier, in order for it to reduce heparan sulfate, it has to be taken up into a cell to reduce the heparan sulfate levels. And so there's a lot and we'll get into more detail that we've learned from this program, but it ends up being an excellent pioneering program for the transport vehicle because of the acute biomarker and that translation, we believe, to clinical benefit.

Okay. Great. So let's talk about what the Phase II/III program looks like? And then maybe you could just compare and contrast levels of activity in the brain versus level of activity in the periphery. First is peripheral activity of the naked enzyme.

Yes. Yes. So it's interesting to go back in time. I mean idursulfase or, I guess, known as Elaprase, but idursulfase has been around for a while, has been approved. And it's -- the dose that's given seems to, at least in blood and urine, reduced heparan sulfate almost to normal levels, but not exactly the normal levels. In fact, it's probably about still twofold elevated. On average, heparan sulfate in Hunter patients is increased by about tenfold in the periphery and in the brain. So on idursulfase, what you see is that patients will get a pretty robust reduction in the periphery, but essentially, no reduction in brain as measured by CSF. So in our trials, we -- basically patients switched directly from idursulfase to DNL310 or ETV:IDS with no washout period. And what we see is, again, that normalization, but notably, we also see an improvement in peripheral biomarkers as well. And we think, again, it's probably because it -- idursulfase generally under dose. The other major factor is that patients develop antidrug antibodies. So it's very common to see antidrug antibodies, and in our study, almost all patients have pre-existing antidrug antibodies to idursulfase before they go on to DNL310. And remember, DNL310 is idursulfase fused to an Fc. And so we think that, in part, maybe the lack of normalization of the periphery may be through some of these antidrug antibodies. And what's -- I think the first clue that we got is that essentially all of our patients have normalized or near-normalized heparan sulfate, so anywhere between 85% to 90% reduction in heparan sulfate. But there were 2 patients that did not -- we did not see immediate normalization. And in those 2 patients, the pre-existing antidrug antibodies to idursulfase were about six to tenfold higher than any of the other patients, and it affected the PK of DNL310. So we'd actually like calculate the minimally efficacious dose. So it's actually about 1.5 mg per kg to normalize, which is really robust. So what those patients needed was either a higher dose or a longer duration of dosing. And some of the data that we just shared at SSIM in Europe was actually that those 2 patients that had these high ADAs, they've been on idursulfase for a long period of time that with DNL310, their ADA levels after about 6 months dramatically dropped. So in other words, DNL310 is helping to essentially establish tolerance. We think this is probably related to the Fc fusion. This is not the first example of Fc-fusion proteins allowing you to build tolerance in part because Fc is recognized itself, right? But we think this is really important data around the transport vehicle and the future MPSs that we'll be going after such as [indiscernible], and I think that's critically important. So ultimately, we selected a clinical dose in the COMPASS study. The question is, what is the COMPASS study of that Phase II/III look like? It's basically a 2 cohorts: a neuronopathic cohort and a non-neuronopathic cohort. And our goal ultimately is to improve cognition and behavioral endpoints at co-primary with CSF heparan sulfate, but we selected a dose that allows us to capture the entire patient population because we have this ability to have a large range what we believe is the right dose. In this case, 15 mg per kg because it allows you to dose over the ADAs and establish tolerance, which I think is a very exciting possibility for the platform.

So how should we think about enrollment there? And I guess because we have an established and effective treatment for the neurological effects of the disease, what is an effective treatment look like there from a clinical standpoint?

Right. So we're actively enrolling now. It is a competitive program. We have -- we're enrolling in the U.S. and in Europe, and ultimately, what we're looking for is a head-to-head comparison with idursulfase. And the study is powered to see a reduction in the rate of decline when looking at either Bailey or Vineland or essentially, the various endpoints that we're looking at. Now the data that we have looking at this is essentially from our natural history study, whatever we can glean from the literature, but also our ongoing Phase I/II in terms of how we power that study. So essentially, power to see a reduction in rate of decline, but the data we've shared so far is showing improvement in the open-label study for the majority of patients in the Phase I/II study.

Okay. Okay. I guess the follow-up here is investors, I think, have struggled with how to think about reductions in biomarkers, which I think there's no question about that you've seen significant reductions in heparan sulfate in the CSF. And I think there are 2 pieces here that I get reflected to me. The first one is, you've seen certain programs which were dosed directly intrathecally. And when they're -- and I think we'll discuss this, but how they were measured probably influenced their ability to see a CSF decline in substrate even though it was an actual decline. So I think the first question is, how can you be sure that what you're seeing is not something like we've seen with other of these enzymes that have actually been dose directly in the CSF?

Right. Yes. That's a great question. And I think the transport vehicle and the approach of using transferrin receptor is all about biodistribution. It's a systemic delivery, right? So it's a given IV and then you're measuring in the brain or in CSF, specifically the effects, right? So in order to have an effect, the idea is that you most certainly have to cross the blood-brain barrier and have this cellular effect. The intrathecal idursulfase, that's a very interesting -- if you look historically at the data, the dose frequency and the dose levels. And without getting into too much detail, they do not see normalization, but they're also measuring heparan sulfate at the same region of injection. So the maximum impact you should have should be right where you inject. And we've seen this actually. Probably the best example for us is comparing an intrathecal ASO versus an ASO delivered with the oligo transport vehicle and looking at the vast difference in biodistribution. And if you look at the spinal cord, you have a very high concentrations of ASO. In fact, it's dose limiting. We see sometimes hind limb paralysis with ASO delivery. We don't see that with the OTV. What we see is this broad biodistribution, right? So that being said, also patients are dosed once a month at 10 mg per kg, at least in the trial that was reported out for idursulfase. And so it's very different than what we're seeing, which is this broad distribution. We see a reduction with the systemic delivery. That being said, there are -- I mean at least what we've heard from clinicians is that if started early enough, they are seeing some benefit in these patients. Now not big enough to have a statistical significant change, not meaningful enough for there to be approval with ITL [ price ], but I think it goes back to sort of the questions around dose level and dose frequency and most importantly, biodistribution.

Good. And then the second follow-up, which we've discussed even over time, which is, there's been a significant investor focus on neurofilament as somehow a catch-all biomarker. And this is a broader question than Hunters really, which is, I think we don't really understand neurofilament either from a temporal standpoint in disease or even in certain diseases that may not be as good a biomarker it may be in other places. So just broadly, how do you think about neurofilament at this point? And then maybe the direct question is, why should people have confidence that -- even though you haven't seen a significant reduction in neurofilament that you have seen a significant reduction in the substrate and that should lead to clinical outcomes?

Yes. So let's start first with the substrate and sort of the cascade of cellular deficiency, right? So when heparan sulfate is elevated, what you see is over lysosomal dysfunction. And that's measured by other lysosomal biomarkers like GM3, GM2, any of which, by the way, when elevated also cause neurological damage. And so basically, what we see as a rescue of all these lysosomal biomarkers. The question then is, what do we observe with neurofilament? And let's talk about neurofilament more broadly. So I think our confidence that we're having a biological effect is related to the fact that heparan sulfate is both necessary and sufficient to cause neurological damage across all MPSs and then we rescue the lysosome. When we started to explore neurofilament, it had never been reported -- actually to this date, the only data presented on neurofilament and Hunter syndrome is what we've published and what we've shown subsequently. And what we observed, and I think in particular, our natural history study was incredible variability. We had one patient that went up 800%. So eightfold elevation in it and like basically 1 year or 6-month time point. Others go up twofold. And interestingly, when we looked at the Batten disease data, another lysosomal storage disease, we also saw this pretty incredible heterogeneity where patients would go up and down and up, which is very different than what you see in ALS. This kind of gets back to like disease-specific behavior of neurofilament. The other thing that was really fascinating is that another cytoskeletal associated protein, tau, is not elevated in Hunter syndrome. So the original hypothesis of neurofilament is really simple. A cell dies, it spills it's guts and you measure it in the form of neurofilament because it's a cytoskeletal protein. You would expect that you'd see elevation in other cytoskeletal associated proteins. So certainly, in SMA and in Huntington, you do see that neurofilament and tau follow each other exactly. They both are elevated, but in Hunter syndrome, we don't see that. In fact, we also see patients that don't have neuro -- elevated neurofilament, but they have the clinical manifestation of neurological disease, right? So we already -- we learned from that, that there's -- it may not be this linear relationship, right? And I think the most striking evidence is in that disease in cerliponase alfa, where clinical benefit is observed in the first year of direct delivery of cerliponase alfa. This, by the way, is directed to the brain at 600 mg. So a massive dose given every other week, but neurofilament doesn't decline until 1.5 to 3 years. So basically, clinical benefit precedes neurofilament. Now in ALS, we may have an example where neurofilament -- and the other way around, neurofilament precedes potential clinical benefit. And then we have cases where there may be a benefit in ALS and no changes in neurofilament. So it's -- we of all people would love to have a surrogate biomarker that predicts clinical benefit. But in Hunter syndrome, it's not neurofilament, right? And I think that's -- and when you pioneer that space that we have to try to figure that out. Now do we expect over time that neurofilament may decline similar to Batten disease? That may be likely. And again, I just point to other cytoskeletal proteins not elevated. So I think we're going to have to approach neurofilament on a disease-by-disease basis. And the question is, can you show that correlation in ALS or an SMA or in lysosomal storage diseases? And that's the way that we'll approach it.

Okay. Great. Great. So I haven't managed time well and I've chewed up a lot of our time. But now we're going to go to a quick hits across a lot of other stuff. So can we talk about FTD, GRN, CTA filed? What's the outlook there? And we're familiar with other FTD programs, so why have you chosen to go after -- so the other program that I think most people are familiar with is basically blocking degradation so to raise for granular levels, right? You've taken a different approach. So why have you taken the different approach, and how to think about updates for that program?

Yes. So our goal is to have quick hits here. So I'll be efficient. So PTV Progranulin is protein transport vehicle -- progranulin, and it's simple. It's like an enzyme replacement therapy. So we're just restoring progranulin. So if progranulin loss of function that causes FTD, it's usually a loss of one of the gene copies, and similar to IDS, that's our approach. And basically, if you want to know how we're approaching it, we published a paper in Cell in 2021, September, that sort of highlights the rescue of these lysosomal defects and microglial cells. So I think probably the biggest differentiator is that we're restoring the protein that is deficient in FTD granulins.

And timeline or outlook for first...

Yes. So we basically are in healthy volunteer studies now. We plan to transition this year into FTD granulin studies. And what's different with this program than some of our other programs is that we need to be in patient studies in order to see the desired biomarker effects and rescuing lysosomal function. So we're not guiding on that timing right now, but simply said, we're partnering -- we partner with Takeda on it, full speed ahead to get through the healthy volunteer study, kick off the FTD granulin and then look at biomarkers.

Great. You'll have more soon. TREM2. You went outside the U.S. because of the issues in starting in the U.S. So how should people think about TREM2? And just broadly given what we've seen from the TREM2 landscape, what's your excitement level or lack thereof on TREM2?

So TREM2, loss of function risk in Alzheimer's disease. The approach here is to use an antibody that activates TREM2. It's an agonist antibody. What we see is, when we combine our antibody with transferrin receptor, we have this shift in potency, probably anywhere between like 20 and 40-fold increased potency to the standard antibody. And part of that is the avidity effect of having transferrin receptor, right? So we -- Takeda opted into that program at the end of last year. We filed the IND, unexpectedly received a clinical hold letter from the FDA. Immediately made the decision to go to Europe after receiving that letter and initiate healthy volunteer studies. It's notable that, that letter did not suggest any additional studies, it's just that we wanted to stay as quick on the timeline as the best we could. And we know that running these studies in a clinical center where we run a number of our studies would allow us to move forward. So I think it's an interesting target. It's in Alzheimer's, which has a lot of complication. It's a really exciting area with the biology being unknown. What I think is most interesting about TREM2 is that when Alzheimer's disease was first described, we know about the plaques, we know about the neurofibrillary tangles, but also there were these fat bodies that accumulate in microglial cells that were described. It's like the 3 hallmark features, and no one really talks about the microglial deficiencies. But what we showed in 2019 is that, that's TREM2 dependent. So basically, TREM2 loss of function in -- not even in an Alzheimer's model, but when challenged, will create these fat accumulations inside the microglial cells. They essentially become unable to turn over like triglycerides and cholesterol esters and activating TREM2 can reverse that. So I think the genetics is compelling. The pathology is compelling. The clinical path is what it is. It's a little challenging in Alzheimer's. I think TREM2 is even a little bit more complicated based on predecessors who have gone forward in this field.

Okay. Great. Why don't we flip to small molecules? Why don't we talk about ALS 343? You have some updates there for us, I think, later this year. So just maybe remind us what to think about that program and that target?

Yes. So closing the door on our large molecule portfolio, obviously, PTV Progranulin, ATV:TREM2. We talked about the Hunter program. Maybe one last point. We have this oligo transport vehicle technology with broad biodistribution of ASOs. Obviously, preclinical, but nonhuman primate data that we think is really exciting. So we've picked several targets, 2 targets to rapidly advance with the OTV and that's basically the transport vehicle portfolio. In terms of small molecules, we have a number of small molecule programs as well, some going into late-stage studies. And one of those is the DNL343 program, which is eIF2B agonist. It essentially activates eIF2B. And the simplest way to think about it is, when a cell is in a stressed environment, it stops translating protein. And if you add DNL343, you can reactivate that translation. And in fact, what happens is, you form these RNA stress granules, which apparently in ALS, they basically get locked into place. Many of the mutations in ALS are linked to these RNA stress granules, right? So based on sort of midyear data -- blinded data, we made the decision to advance that program into late-stage development. So we're basically designing those studies now. We look forward to sharing additional data. I think the expectation is going to be similar to what we saw in healthy volunteers. Where we can look at the ISR pathway, we can activate it and then, we can basically inhibit the ISR pathway by activating eIF2B. And most of this is done in human PBMCs. We've done that in healthy volunteers. We'll do the same in ALS. We have a number of candidate biomarkers, but I would say there, stay tuned. We don't see a sense of urgency. I mean this is a highly competitive space, and we want to continue to mature those data.

Okay. Okay. Great. LRRK2, obviously, a program that you're partnered with Biogen. You pushed into pivotal studies. Just how to think about the broader Parkinson's population versus the highly mutated population?

Right. So there, we have 2 studies. We have the LUMA study, which is basically focused on idiopathic Parkinson's disease. This is a large study focused primarily on early patients in that study. And there, it's essentially all comers with the exception of LRRK2 carriers. And the second study, the LUMA study is the LRRK2 carrier study. LUMA has kicked off. We're enrolling that study now, and Lighthouse will soon kick off with the LRRK2 carrier study. And it's all based on functional benefit basically for these patients. So these are design studies to see robust clinical benefit. So they're powered for clinical benefit, and we're looking forward to testing the LRRK2 hypothesis, both in sporadic Parkinson's as well as LRRK2 carriers. And I think the data around sporadic is that there's a number of genetic lesions that point to lysosomal defects and inhibiting LRRK2 improves lysosomal function. So we think there's a broader mechanism here.

Okay. Cool. RIPK to round us out. There's obviously peripheral studies, centrally acting studies. Maybe just comment briefly on ALS. What's the potential there? And then how to think about AD and the potential path forward in AD?

I always get excited talking about our portfolio. There's a lot of programs, and we could spend a half hour talking about RIPK and TNF receptor 1 and being downstream of TNF receptor 1. This is actually one of the first programs we ever started working on at Denali. And now there are 2 clinical stage programs. One is a molecule that crosses a blood-brain barrier. It's a HIMALAYA study. It's an ALS study being operationalized by Sanofi. They're leading that. It's a global study in U.S. and Europe, powered for clinical efficacy. That's the goal. And then in terms of the peripherally restricted compound, that's moving forward in lupus. And other -- they're picking additional peripheral inflammatory diseases, and so this is a great partnership. Sanofi is very committed to RIPK as a pathway. Obviously, being downstream of TNF receptor 1, and we look forward to getting additional data from those studies. I'll just also comment that the program will move forward also in MS. The -- actually, the CNS crossing molecule. So...

Great. So then maybe just to finish off, remind people where you are from a cash position? How you can prosecute the portfolio? And really, you mentioned the handful of partnerships that you have, so how you think about non-dilutive capital as being an important driver of the business?

Yes. I mean our ultimate goal is to discover, develop and market our medicines. In fact, very recently, we have Chief Commercial Officer join -- actually, not that recent. Now it's been 9 months, helping design our late-stage studies and getting us ready to launch in lysosomal storage diseases and in ALS. And we're in a strong cash position, $1.16 billion in cash. Partnerships are fantastic. Great collaborations with Takeda, Biogen and Sanofi. The portfolio is about split between partnered assets and unpartnered in the development, but obviously, in preclinical, most of it is unpartnered. We continue to be opportunistic around partnerships. It's not necessary right now, but with our platform, we have the ability to go so broad that there is certainly opportunity around partnering.

Great. Ryan, thanks for being here. I appreciate it.

Thank you.