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

Well, great. Good afternoon, everybody, and thanks for joining us for the next session. I'm Matthew Harrison, one of the biotech analysts here at Morgan Stanley. Very pleased to have Denali with us for this session. Before we get started, I 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/research disclosures. And with that, happy to turn it over to Ryan Watts, the CEO of Denali to make some opening comments, and then we can jump right into it.

Thank you, Matthew. Excellent to be here with you. Again, a little unfortunate that we're here by Zoom, but hey, that's the way that it is. It's been an amazing last year or so at Denali. I just want to highlight a few points, and then we'll dive into the Q&A. So first, we have 5 clinical stage assets and 2 of which will be advancing to late-stage clinical development in early 2022 and this is an exciting time for us. Our first small molecule to advance into late-stage clinical development as well as our first large molecule using the transport vehicle technology. So along those lines, our program in Hunter syndrome has laid the foundation in terms of validating the transport vehicle, getting robust drug across the blood-brain barrier and allowing us to advance other enzymes, antibodies as well as ASOs. And our LRRK2 program, which is in collaboration with Biogen, will also be advancing the late-stage clinical trials. And so it's an exciting time, I think the way one should look at Denali is we have 2 platforms. We have the degenogene platform, which is essentially the underlying mechanisms of neurodegeneration. We, of course, are biased towards monogenic diseases or diseases in which we clearly know the genetic underpinnings, but also we've founded the company essentially to develop treatments for Alzheimer's and Parkinson's and ALS, more complex genetic diseases as well. In addition to that, the second platform is our blood-brain barrier crossing technology or platform. In that case, it's really split between small molecules and large molecules. And I'm sure today Matthew will talk quite a bit about our transport vehicle technology as well as some of our small molecule programs. There are some recent milestones or near-term milestones coming. And I think I'll just highlight a small molecule program targeting ALS. We just announced initiation of ALS clinical trial for EIF2B activator program. It's DNL343. We'll be sharing clinical data early October at NEALS and ALS conference. We also have 2 more transport vehicle molecules that will be entering the clinic will be filing INDs or CTAs by end of the year or early next year. One is PTV progranulin for FTD, and we actually just had a paper on the mechanism of that approach, both understanding the mechanism of programming, but also crossing the blood-brain barrier with a progranulin approach that was published in Cell in early September. And then the second program that will be entering clinic is TREM2 that will make 7 clinical stage programs, 3 of what's using our transport vehicle technology. So in addition to that, with the advancement of our biotherapeutics platform, we are building clinical manufacturing. We realized that now is the time to basically accelerate the build of clinical manufacturing so that we can bring more enzymes, antibodies and ASOs across the blood-brain barrier and have also began to build out, I'd say, stay tuned around our commercial organization in which we plan to go after rare diseases followed by a more complex and larger diseases. So an exciting time at Denali and Matthew I look forward to questions, and let's dive in.

Okay. Great. Thank you for that Ryan. So you're right. Why don't we start with blood-brain barrier. And I think look, I think it will be helpful for me to review briefly sort of the data that you've generated there. But I think more importantly, there's obviously some controversy in the Hunter's program just around which biomarkers are best describe the activity of that drug? And in particular, obviously, neurofilament is one of those that I think investors are very focused on, but yet that sort of contradicts some of the direct evidence around GAGs and other things that you have there. So maybe you could just comment broadly on the blood-brain barrier program but focus in on some of those points and your viewpoint on them.

Right. So about 6 years ago, when we founded the company, our goal was to invent a platform that allowed us to get antibodies, enzymes and now ASOs across the blood-brain barrier with systemic delivery. And the lead program is our Hunter program, which is essentially a iduronate 2-sulfatase or IDS engineered across the blood-brain barrier using the transferrin receptor, which is highly expressed in blood vessels in the brain. So the recent data, which is 6-month data from Cohort C, in addition to that 12 or 10 additional patients in Cohort B., really, there are 3 major areas that we focused on. First was safety and here, we see that it's well tolerated and it's consistent with standard of care. So very similar to enzyme replacement therapies in this case, Elaprase or idursulfase, okay? The second was peripheral activity. And in fact, we saw superior peripheral activity when switching from idursulfase to DNL310. Part of that, we think, is related to the fact that we can give a higher dose because it's tolerated at higher doses. But the second part is that transferrin receptor could provide better biodistribution throughout the body. And then the third area is CNS activity. And here, we think about the primary biomarker, which is heparan sulfate and heparan sulfate actually across the MPS diseases is both necessary and sufficient to drive CNS involvement. In other words, if heparan sulfate is elevated, there is a neurological component. And so that's our primary biomarker for decision-making. And what we observed in this data set, which included 15 patients in the biomarker data set is normalization across all doses, 3, 7.5, 15 and 30 mg per kg. And I think importantly, at the lowest dose of 3 mg per kg, the fact that, that normalized, that was actually more robust than what we've observed in animal models. And we think that's because in the human brain, you have a much larger vascular surface area. So highly validating for the transport vehicle and for transferrin receptor. And importantly, when you measure heparan sulfate in CSF, what we've shown is a 1:1 correlation with heparan sulfate in the brain. And in fact, all these patients were on Elaprase, they would have reduced heparan sulfate in the choroid plexus, which produces the CSF and also in brain capillaries because they're first order cells. However, when you get across the blood-brain barrier, that's where you start to see this robust reduction in heparan sulfate. That subsequently led to a reduction in various lysosomal biomarkers, GM3, BMP, GlcCer. And these actually are brought down to normal levels as well. So this is critical. So not only are you reducing the primary substrate, but you're seeing this correlation with lysosomal rescue. So the next step for us was the assessing clinical benefit as well as neurofilament. And what we observed in all 5 patients with really an age between 5 and 8 years of age where they should be declining, we see that, in fact, they're improving and global impression of change as assessed by both clinicians as well as a parent or caregiver. And this is actually -- this was surprising because these are more advanced patients that enrolled in this first cohort in the Hunter study. In case of neurofilament, we saw a lot of variability. We are the first ever to look at neurofilament in Hunter syndrome. We noticed that just in cross sectionally, there's a lot of overlap between non-MPS and MPS. And then in our natural history study, 3 of our patients that we were able to then subsequently enroll in cohort A, had a lot of variability, including elevation one about eightfold or 800% elevation before going on to DNL310. And then at that point, again, about 2 or 3 patients decline, 1 or 2 go up, and we see about a 15% elevation after switching to DNL310. Now we understand the controversy here. We're the first ones to really explore neurofilament. We don't know if it's going to be informative. Certainly, it doesn't appear to be correlating as far as we can tell with the clinical benefit we're observing in these 5 patients.

And maybe if you can just touch on that a little bit more because I think what we've -- there's not a lot of data on neurofilament in different diseases. But what you have sort of observed is there seems to be a pretty good correlation in multiple sclerosis and then as you sort of work out in some other diseases, even if you look at, say, some Battens replacement therapies, it -- neurofilament does decline, but it takes 3 or 4 years for it, it takes a substantial amount of time for it to happen. So I guess time course is also important in terms of what happens. So I guess, maybe just give us your outlook on using neurofilament as a biomarker broadly across CNS diseases as a company focus there versus some of the more proximal biomarkers like heparan sulfate here.

Yes. So I mean one of the advantages of Hunter syndrome is there's an approved therapy, where there's a correlation between heparan sulfate reduction and clinical benefit in the periphery. So for us, that's the primary driver for decision-making. But you have a really good point, which is we are seeing a pretty significant heterogeneity across diseases like MS like Batten disease. And maybe I'll just make 1 or 2 comments related to that. I mean, MS of course, relapse remitting, you would see an elevation in neurofilament going to decline naturally. So it's probably more naturally increasing and decreasing over time. That disease is probably the best example, at least relationship. It's a similar lysosomal storage disease, enzyme replacement therapy, actually given directly to the brain. And what's observed here is that clinical benefit actually far precedes neurofilament. So they observed a clinical benefit in the first year of direct delivery of Cerliponase Alfa. However, neurofilament itself doesn't decline as you mentioned, until 2 or 3 years. So I think what we've learned is that the -- for us, the correlation we're drawing is between heparan sulfate lysosomal rescue clinical benefit and then it may be that these all precede changes in neurofilament. And remember, we're also working with young children who naturally have elevated neurofilament. There's natural remodeling taking place early on in development, you have excess connections, which are naturally pruned. Now with that being said, we're actually very excited to sort of blaze the trail here. We're okay with the uncertainty, and we'll continue to look at various biomarkers as we go forward. But in terms of decision-making, really heparan sulfate, the ability to both rapidly as well and robustly, but have a sustained normalization is important for us, in terms of dose selection and advancing this program into Phase II/III.

Okay. Great. So then I guess that leads to the second question, which is -- how do you get this drug approved? What is the potential pathway for doing that? And do you need to do any more substantive work before you can engage in a pivotal study?

Yes. So we're ready to go on the pivotal study in terms of the data that we have in hand. We now have dose ranging from 3 to 30 mg per kg. We're selecting the dose. We've engaged regulators on design of that clinical study. So that's going to be very important. But I'd say, in parallel to that, we're exploring also the peripheral benefit that we're observing as well as really robust clinical benefit by enrolling a cohort that's focused on younger children. So I mean, obviously, what we would ideally do here is just use natural history to show that we're seeing a clinical benefit and all patients would go on to DNL310. However regulators are pretty keen on a comparison head-to-head with Elaprase. And so we're wasting no time. We're gearing up for that Phase II/III as we continue to expand the Phase I/II, which already has a number of patients in both Cohort A and B and beginning to enroll Cohort C now as well.

And then just to remind people, and I'm sorry to say I don't remember this off the top of my head, but Elaprase is a fully approved drug. So there's no accelerated approval strategy open to you. Is that right?

Yes, it's a great question. I cannot comment specifically on how regulators do that. I will say we're not working with the neuro division, right? So we are working with rare disease as we think through this. And so for us, an accelerated approval based on biomarkers is not likely. However, obviously, seeing clinical benefit as we observed in the first 5 patients in cohort A and now going to younger patients, that's a path we would take.

Got it. Perfect. So then I guess a follow-up to all of this discussion is -- and you mentioned some of the other work you're doing using transport vehicle. If we sort of take -- and I'm sure we can debate this. But if we take that, these initial studies have demonstrated that TV is safe and TV gets a substantial amount of drug across the blood-brain barrier, how are you sort of accelerating that broad strategy for looking at a variety of antibodies or ASOs or enzymes that you'd want to get into the brain.

Yes. So I think that the evidence that's irrefutable is for us, heparan sulfate and CSF correlates 1:1 with brain reduction. We've shown that. We've shown that at the cellular level. We can reduce heparan sulfate in astrocytes, microglia and neurons at basically purifying single cells and correlating that in animal models of what we see in humans. As I already mentioned before, Elaprase is pretty good at knocking down heparan sulfate in the periphery, which it would do in brain capillaries and in the choroid plexus. So for us, we have definitive proof that the transport vehicle works. Now the question is across what other modalities? And so the next 2 molecules to enter the clinic, 1 will be an antibody for TREM2. It's an agonist antibody. The second will be a progranulin molecule crossing the blood-brain barrier acts very similarly in fact, the Cell paper, which we published earlier this month, really highlights its role in lysosomal function. And we've now accelerated another 6 enzymes. We have SGSH in IND-enabling stage now, so preparing to enter that in the clinic. And I think probably one of the most exciting things that unlocking the blood-brain barrier can do for us and which we've recently shown is that we can actually take a full antibody, tag an ASO, inject it systemically and knock down gene expression in brain. And I think importantly, we see that it's distributed broadly unlike the intrathecal delivery of enzymes and ASOs, which has limited biodistribution, especially in humans where you're traveling a large distance and relying on basically diffusion. Here, we can see crossing capillaries and knocking down gene expression in astrocytes, microglia and neurons. And so we've very rapidly advanced the ETV by adding additional enzymes. We're expanding our antibodies and then the OTV will be, I think, a really important platform or subplatform of the transport vehicle to be able to knock down gene expression or modulate gene expression.

Okay. Okay. Great. Good. Well, we'll look forward to seeing all of that as it progresses. Maybe, I guess, one last question. You have a deal with Biogen, which covers LRRK2 in addition to blood-brain barrier. Maybe just to remind people of what Biogen's rights are for blood-brain barrier programs?

Yes. So they basically have the ability to opt into 2 programs that are named. We haven't disclosed the second program, which is a Parkinson's program. The first program is Abeta and so the idea here is an antibody using the transport vehicle technology to show basically more robust brain uptake at a lower dose and plaque immunodecoration as well as plaque reduction. And so we've presented some of this Abeta data last year in BBB R&D Day, and that was a big part of the Biogen collaboration is essentially thinking about the next generation of Abeta antibodies using these blood-brain barrier technologies. It's going great. We really enjoy collaborating with Biogen. They obviously know an enormous amount around Alzheimer's and antibodies for Alzheimer's disease. And so that program is moving forward.

Okay. Great. Good. Why don't we sort of take the pipeline in reverse order to maybe normal and talk about EIF2B because I think it's a target people don't talk a lot about, but you obviously are making progress there. So maybe people remind people about the mechanism why you're investing in there and what we're likely to see.

Yes, it's a timely -- it may not be in order because it will be the next data that we present. So we have clinical data that we're presenting at the beginning of October at the NEALS conference as mentioned before. This is the first EIF2B activator to enter clinical studies. And just a reminder of the pathway itself. So when cells are in a stressed environment, they lock down their translation basically as a transient protective mechanism. But interestingly, in ALS, many of the genes that are linked to ALS are these RNA/DNA binding proteins. And what is observed is that in ALS when a cell becomes stressed, it locks these -- you create these RNA stress granules, and they're not released. The cell starts to starve and then it dies. And in the case of motor neurons, if motor neurons are dying of course, that leads to motor neuron disease and motor dysfunction. And so essentially, the mechanism here is to release these RNA stress granules by activating EIF2B. It's also worth noting that there's a genetic link to a leukodystrophy known as vanishing white matter disease which is a direct genetic link to EIF2B, EIF2 alpha as well as sort of a mechanistic link. Now we've done an enormous amount of work on it. We have only presented like 1 or 2 pieces of data, and we look forward to sharing more of that data at NEALS coming in October. And the goal here is, obviously, to be first-in-class with a molecule that can basically release these RNA stress granules. ALS is the primary indication. Obviously, this rare disease vanishing white matter disease, there's hope that we could take a molecule forward there. It's obviously an ultra-rare disease. And then in addition to that, a subset of patients in Alzheimer's have TDP-43 pathology, which are found in these RNA stress granules, it's about 30% of Alzheimer's disease. We don't yet have a biomarker to determine what that 30% of Alzheimer's is. So our focus right now is ALS, and we announced last week basically the initiation and beginning enrollment of that clinical trial.

Okay. Okay. Great. And so I guess what I want to ask about ALS, like maybe just also remind people what kind of clinical data is meaningful in ALS and how much data out of a Phase Ib study you can get in terms of demonstrating clinical effect?

Yes. So this is, of course, focused on -- in ALS, we're focused on biomarkers that are related to the integrated stress response, right? So that would be the first is really looking proximal to EIF2B. But as is the case with a lot of these rare diseases, you would have really an open-label extension where you would stay on the drug for a long period of time, and that's ideally what we would do as well for DNL343. There you can look at other endpoints. And ALSFRS, basically assessment of motor function, is ultimately what you'd be looking for halting disease progression with EIF2B activation.

Okay. Perfect. Good. Maybe we can tackle RIPK next. Obviously, you got 2 programs there, 1 that's partnered for peripheral disease, which probably gets a little less airtime than the 1 that's centrally acting, but maybe just an update on 788 to start off and then we can touch on sort of what's happening with 758.

That's right. So we have 2 molecules currently in the clinic for RIPK, the blood-brain barrier penetrant molecule, DNL788, is in a healthy volunteer study, and we're now gearing up for ALS study and we being Sanofi. So this program is largely led by Sanofi in the partnership, including the healthy volunteer study. So the goal here is robust exposure in brain, similar to what we show with the LRRK2-DNL151, what we're -- we can show at DNL343, basically, that broad sort of distribution and robust brain uptake and then a correlating biomarker to show target engagement. In this case, it's phospho RIPK. And the idea there is essentially to block this pathway, which is downstream of TNF receptors. So receptor interacting protein kinase 1 or RIPK1 is downstream specifically of TNF receptor 1. And so obviously, in diseases that relate to peripheral inflammation like lupus, where DNL758 is being tested, there's validation of the pathway. It's a TNF pathway. However, in CNS diseases like ALS, MS, and Alzheimer's disease, there's also substantial elevation around this deleterious inflammation and activation of the RIP kinase program. So we'll be transitioning it from healthy volunteer into patient studies, again, focused on biomarkers before we enter larger studies looking at clinical endpoints.

And maybe just for everybody's benefit, remind everybody, I mean, RIPK, you've been working on for a while and trying to get the right molecule. So how confident do you feel like you've solved most of those issues? And the data here, I guess, will eliminate whether you solved some of those issues?

Yes. And I think that's the path for small molecules. We know at the very beginning bringing multiple molecules into the clinic and every molecule iterates on what you've learned from the first molecule and make a better and better version of that. So obviously, we learned a lot from DNL104, which is one of the first molecules we took into the clinic as well as 747, and the idea is to engineer around specificity. What we've observed has essentially been off-target related with those previous molecules, right? And now I think the other thing we learned, 747 was actually very robust inhibition in humans, but it did not sustain inhibition above 90%. And you may recall that we were able to correct 1 of 2 biomarkers in the Alzheimer's study. And we knew that the next step was to run a much larger study, clinical proof of concept in Alzheimer's. We were not willing to take a risk with what we thought was probably too low of a dose. And in order to elevate that dose, we'd have to elevate it in the clinic, and it was actually faster to bring 788 forward which didn't have this off-target liability, which played out in preclinical models. So we actually saw no toxicity in the clinic for DNL747. Now we're lucky because we also now have 758 data. It's already in Phase II. And that, in many ways, sort of validates the safety of inhibiting RIPK, and that's an extremely robust inhibitor in the periphery, but it isn't designed to cross the blood-brain barrier. So it's polarity limits its exposure into the brain. So we're now building up a pretty substantial data package around RIP kinase and tolerability and 788 will be the first molecule that we think we can sustain the levels of inhibition we want to sustain to then test the hypothesis in ALS and Alzheimer's and MS.

Okay. Great. And then I guess last question, just remind people about peripheral RIPK, why you think that target is interesting. And if other people maybe not directly looking at RIPK, but have demonstrated that peripheral activity could be important.

Yes. So I think it gets back to this original mechanism I highlighted, which is the TNF receptor pathway and TNF and TNF receptor 1. Now interestingly, RIP kinase is specifically the downstream of TNFR1, not TNFR2. So it's more selective than, say, like an anti-TNF. And the idea here is that essentially any disease where anti-TNF has shown efficacy, it's worthy of assessing RIP kinase inhibition in those diseases. So really an oral approach to inhibiting the pathway specific to TNF receptor 1.

Okay. Great. Good. So then maybe in the last few minutes here, we can touch on LRRK2, which is obviously another 1 of your small molecule key programs. And I guess, first, just remind -- I think this is an area that we're starting to see either more competition or others trying to bring LRRK2 inhibitors or next-generation LRRK2 inhibitors to the clinic. So maybe just for sake of everybody's benefit here, there was obviously a lot of LRRK2 inhibitors to begin with and then only a few survived. And just sort of remind people of trials and tribulations of the pathway and then why you think you have a strong inhibitor in your hands?

Yes. I'll start with the reminder of the rationale around LRRK2. So LRRK2 is mutated and a large -- actually about 3% of Parkinson's disease, the most common mutation, it's the G2019S mutation, is in the kinase domain of LRRK2, and it's hyperactivated. So it increases LRRK2 activity by about twofold. And this was discovered in 2004. We actually began working on it in 2006 and discovered a lot about the biology around LRRK2 and its role in disease, including the fact that it's activated in idiopathic Parkinson's disease, so in broader Parkinson's disease. And so as we begin developing inhibitors, many of the inhibitors we worked on were at Genentech and then, of course, Denali, we license those inhibitors. But DNL151, our lead program was invented at Denali, blood-brain barrier penetrant, very robust inhibitor. And the goal here is to bring LRRK2 kinase activity back to normal levels and to normalize LRRK2 function. Now interestingly, when LRRK2 is hyperactivated, what we see is a coalescent of lysosomes. Lysosomes become dysfunctional and there's this link between lysosomal dysfunction in Parkinson's disease. So for example, GBA, heterozygous mutation carriers have a much higher risk of developing Parkinson's disease, GALC, other endolysosomal proteins have been linked to Parkinson's disease and in homozygous mutant form cause basically a lysosomal storage disease. And what we've observed is that when we inhibit LRRK2, we increase the size and function of the lysosome. So the idea is that we can go in a broader idiopathic population as well as specifically in LRRK2 mutation carriers. And that's exactly what the late-stage trials are going to look like. We're going to focus on LRRK2 carriers for one of our clinical trials and the second will be an idiopathic Parkinson's in part because many of the genetic variants actually point to lysosomal dysfunction. In fact, we have data where we can inhibit, for example, Gaucher patient fibroblast, LRRK2 and improve lysosomal function by fourfold in this otherwise like very dysfunctional lysosome in these mutation carriers. So that's the path. I think we have -- we're not aware of another LRRK2 inhibitor in the clinic there may be in the last month or so, but we have 2 LRRK2 inhibitors in the clinic DNL151 being the lead program. And there, we can basically robustly inhibit LRRK2 in both mutation carriers as well as idiopathic Parkinson's.

Okay. Good. Well, Ryan, thanks very much for being here. Thanks for the comments. I very much enjoyed it.

Yes. Likewise. Take care, Matthew.

Bye.