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

Hello, everyone. Welcome to another afternoon session at the 2020 Jefferies Global Healthcare Conference. It's been fantastic so far. And up next with me, I'm really happy to introduce the CEO of Denali Therapeutics, Ryan Watts. Denali has been on a torrential pace with the pipeline over the last year or 2. And importantly, I think about 2020 as a critically important year for Denali because there's a lot of data reading out this year with all of their programs. So without further ado, Ryan, I'll turn it over to you to give us some update progress and tell us about the milestones coming up.

Yes. Great. Thanks, Mike. Excellent to be here with all of you remotely at the conference. Looking forward to highlighting some of our science and the progress in terms of the development portfolio. So just a reminder, Denali was founded with one purpose, to defeat degeneration. That goal or purpose extended across a number of therapeutic indications from rare neurodegenerative diseases to common Alzheimer's disease or sporadic Alzheimer's. We've had a number of clinical trials either initiated or completed in all 4 of the areas shown before you here. And today, I'll focus primarily on our rare disease effort as well as our effort in Parkinson's disease with LRRK2. The company has 2 platforms. So Denali is built on 2 platforms. The first is the degenogene platform, which is broken into biological areas that are implicated through human genetics in these diseases, such as lysosomal function in Parkinson's disease, glial biology in Alzheimer's disease and cellular homeostasis in, let's say, ALS. These biological areas also overlap with diseases and we built a deep biology team focused on these areas. We also have blood-brain barrier technologies, split into 3 groups: small molecule chemists enabled to engineer small molecules across the blood-brain barrier, such as our LRRK2 program for Parkinson's disease. Biotherapeutics, and I'll focus on some of that work today, to engineer large molecules to cross the blood-brain barrier. And then in the future, in fact, I'll just introduce today our first gene therapy program in the portfolio and we'll share data in the coming year or so, on that program. Again, designed around engineering brain delivery as a primary driver. So the company in the last 5 years -- we're almost exactly 5 years old. We've had a number of filings, either IND or CTA, a number of clinical trials initiated, including 2 that we filed at the end of the year and are initiated now, one in Hunter syndrome and one eIF2B activator, which we plan to develop in ALS. We also recently read out clinical data from our Parkinson's disease program, and we'll talk today a little bit about the 2 molecules, DNL201 and 151, both of which are LRRK2 inhibitors and our goal to pick one of those for Phase II/III studies going forward. The portfolio is enabled through a number of incoming partnerships, including around intellectual property as well as molecules. A good example is the Genentech partnership for LRRK2 or the F-star partnership around intellectual property and our Fc engineering approach. And then also, we have a partnership with Takeda on 3 targets using biotherapeutics that are engineered to cross the blood-brain barrier and Sanofi on our RIP kinase program. Here is our portfolio. Just a couple of notes here. We have a number of clinical stage molecules, LRRK2, IDS, RIPK and the eIF2B. And then recently, we've added our first AAV molecule targeting Parkinson's disease, and we'll discuss this in more detail in the coming year. Again, the focus being on engineering brain delivery of this molecule. And then again, we'll focus on the top 2 programs today, LRRK2 and iduronate 2-sulfatase or IDS. In terms of what we want to read out this year, I think, Mike, you made this introduction, it is a big year for Denali. The first is to select a molecule and prepare for II/III -- Phase II/III study for the LRRK2 program. This is DNL201 or 151. So completing the 151 studies, making a decision to move forward into later-stage studies. We want to establish biomarker proof-of-concept in patients for our ETV:IDS program for Hunter syndrome. And as you can see at the bottom there, that will also establish the transport vehicle platform proof-of-concept in humans. So this is our lead program for biotherapeutics. And we've also initiated 2 more IND-enabling studies on the previous slide on the portfolio, ATV:TREM2 and PTV:Progranulin. We have an ongoing DNL343 study for eIF2B activator. That particular program is paused because of the COVID pandemic, and our goal is finish it as soon as possible. It's being run in Europe. We were able to dose several arms in a healthy volunteer study and go back to dosing and then select a dose to go forward into patient studies. And then finally, we have results from our DNL747 ALS and Alzheimer's study. We're completing those studies, doing the data analysis and making a decision with our partner Sanofi, and plan to disclose that data by mid-2020. Just a note on the impact of the pandemic. In terms of operations, both on-site and remote operations are nearly at full productivity levels. We are Bay Area based and as a result, there was a strict shelter-in-place. That being said, biotech companies were largely exempt in that shelter-in-place. And of course, we've taken very careful precautions to continue operations. The enrollment of our DNL151 study, you may recall that we have the initial data that we presented at JPMorgan on healthy volunteers, and we had already initiated and had ongoing patient studies. And then based on successful data in the healthy volunteer study, decided to continue to dose escalate. And so there are several higher dose arms that are currently paused that we plan to continue to roll, but we do have a large data set on 151 and I already mentioned with DNL343 and eIF2B that, that program is currently paused. This being said, the key 2020 clinical readouts and decisions remain on track. We have a strong balance sheet with just over $600 million of cash at the end of Q1. So let's start with the first program. And the reason I'm presenting in this order is that we had 2 publications in Science Translational Medicine last week, so I'm going to highlight some of the new work that was published in those studies and establishing this as a broad platform for biotherapeutic delivery for neurodegenerative diseases and more broadly for diseases of the central nervous system. So just a reminder, the brain is protected by a blood-brain barrier. This blood-brain barrier maintains a tight microenvironment for signaling of neurons in the nervous system. And as a result, most molecules do not readily cross the blood-brain barrier. In fact, most molecules are actively transported such as glucose and iron. So our goal as the transport vehicle is to take advantage of one of these natural transport mechanisms known as the transferrin or transferrin receptor system, which is required for shuttling iron into cells. And what we've been able to achieve is engineering a portion of the IgG or the Fc portion to bind to the transferrin receptor, here in what's called the apical domain. And this binding is non-blocking. So it doesn't interfere with iron or transferrin binding to the transferrin receptor. In the first paper published last week, this highlights basically the discovery of the transport vehicle. So it's brain delivery of a therapeutic protein using an Fc fragment. Again, the Fc portion of the IgG is engineered to bind to a blood-brain barrier transport system. And we validated this in both mice and monkeys. And what I'm showing here, there's some data in monkeys on the bottom across different brain regions, and we see anywhere between 25 to 35 fold increased brain uptake across all of these brain regions. In many cases, now in the therapeutic range above the [ sublaminal ] range that standard antibodies would show. You also can see in the panel on the right here a broad distribution throughout the cortex in this nonhuman primate brain compared to a control antibody. And in the right-hand corner, we've also shared the co-crystal structure showing the actual binding site on the Fc to the apical domain of transferrin receptor. In the second study, published back-to-back in Science Translational Medicine, we basically link this Fc portion to an enzyme for Hunter syndrome, and I'll go into some detail about this data. One of the pieces of the data I'd like to share here is that we could use imaging tools to basically show the ability of ETV:IDS to eliminate substrate compared to control on the top panels versus the bottom panels. Here on the right-hand side is ETV:IDS. And this is basically comparing to simple Elaprase or vehicle. So as we've now disclosed the invention of the transport vehicle, just a comment that it's a modality-based delivery system that allows us to basically make different types of Fc fusions either with traditional antibodies, bispecific or bivalent, again using this monovalent binding transparent receptor, enzyme fusions, protein and then more recently, we've shown that we can deliver antisense oligo across the blood-brain barrier in preclinical models, achieving greater than 70% gene knockdown using the ASO fused to, what we call, the OTV. Again, we'll disclose more of this in the future as we select targets using the OTV. So let's focus in on the enzyme transport vehicle and specifically on iduronate 2-sulfatase or IDS for Hunter syndrome. So I'm going to illustrate here the ETV platform using this molecular animation. So what you see on the top is the enzyme itself, this would be in yellow and then the feet in orange. And the Fc region, again, as we've disclosed in the transport vehicle paper in Science Translational Medicine, the binding to the apical domain of the transferrin receptor here, and we've solved that structure. And then when this binds, we're able to essentially follow the path of iron across membranes. The human brain has about 400 miles worth of blood vessels. These blood vessels, again, evolved in such a way to protect the nervous system. And the idea here is that biotherapeutics can get into the brain often at subtherapeutic concentrations, relying just on passive uptake of endosomes. So the goal here is to then bind to the transferrin receptor, which is actively taken up and basically packing these endosomes, which undergo transcytosis across membranes and you get broad biodistribution throughout the brain. So this idea is not a new idea. In fact, it was proposed in the late 1980s, but it's just now that you're seeing the first clinical test of this hypothesis by our program and others using the transferrin receptor vehicle. What's unique about the enzyme approach and what we've shown in these papers is that not only do we show that we can cross the blood-brain barrier, but when you now zoom into the various cell types in the brain, you see that the enzyme is, in fact, taken up in the cell type, likely by the enzyme receptor known as mannose 6-phosphate receptor brought into cells, and this is again an endocytic event. Brought into cells, the enzyme then makes it to its site of action, which is the lysosome and then breaks down substrate there in the lysosome. So now what I'm going to do is show you some data that supports this molecular animation. So first, before I do that, I'll just introduce Hunter syndrome. It's a monogenic disease, loss of function, mutations and IDS lead to accumulation of heparan sulfate or dermatan sulfate known as the Glycosaminoglycans. These accumulations of GAGs then lead to lysosomal dysfunction and then ultimately lead to neural inflammation and neuronal cell loss. We have ability to measure biomarkers across this entire cascade and it makes it a very powerful way of basically validating the transport vehicle technology, but also building confidence that we'll have a benefit in patients. So here's one of the first experiments in which we compared ETV:IDS to Elaprase. And I think the key message here is our goal is to replace Elaprase. So we have equal efficacy in liver at reducing GAGs. You can see down to baseline levels. However, only ETV:IDS substantially reduces brain levels up to 80% reduction. This is after just 4 doses. And you can see a complete normalization of lysosome when measuring a lysosomal lipid, known as BMP, back to wild-type levels. We next assess, and as I've shown in the molecular animation, the ability of these enzymes to basically knock down substrate in the various cell types in the brain. And this is critical and a key differentiator from some of the AAV approaches in which there's a preferential infectivity of various cell types. So here, we can take entire populations of neurons, astrocytes and microglia, the various cell types in the brain. And we can see 80% to 90%, and in some cases, greater reduction of GAGs, again, after 4 weeks of dosing across these 3 cell types. This correlates with an increase in enzyme uptake in the various cell types as well. So if we then now look at -- in detail at a single neuron that's been treated with ETV:IDS, injected again systemically and asking where does it circulate, what I can show here using super resolution microscopy is the cell body in blue and green, NeuN, identifying it as a neuron; LAMP2, which is the staining for the lysosome; and then human IgG, which is the ETV:IDS. And then yellow is the colocalization of ETV:IDS with the lysosome. So in order for us to see this, we injected the animals, the molecule crossed the blood-brain barrier, entered into cells and colocalized with the lysosome. If we look more broadly across tissue types, and we show this in the nonhuman primates across various regions of the brain, we see again this 20 to 30 fold uptake in all the regions of the brain and cellular uptake across different cell types as well as shown here when looking at neurons, again in the mouse brain. So I think most importantly, the question is, how much do we need to reduce GAGs by in order to have a benefit? In other words, blocking neurons from degenerating. So in this experiment, we dosed at relatively low doses at 1 and 3 mg per kg once a week for 13 weeks. We see a robust reduction in brain GAGs, reduction in CSF GAGs, about 50% in CSF and maybe 60% to 70% in brain and a complete rescue of neurodegeneration. And in fact, in the paper that we just published, if you look at in detail, specifically at Figure 8, you can see not only do we block neurofilament elevation but also all the various markers that we identified in terms of lysosomal function. So this is the initial bar that we've set in our clinical studies, is to reduce GAG levels by about 50%. And in doing so, we would predict a protection of neurodegeneration. And then we expect after longer-term dosing that, in fact, those levels will go down even further. So with that in mind, we have submitted IND at the end of the year, that has since cleared, and we're now in the process of gearing up to enroll this study as we've approved the protocol. And the objective here is biomarker proof-of-concept of the -- in Hunter syndrome to enable our transport vehicle platform and then to enable this also in Hunter syndrome and to move towards approval. It's a 6-month study, looking at PK/PD and, of course, safety. And in terms of the success criteria, the goal here is that it's well tolerated and that we have a robust reduction in CSF GAGs. Again, the bar being 50% in short term, and this will be in our interim readout, and then we'd expect in longer-term dosing effects on other lysosomal biomarkers and ultimately on neurofilament. So I focused in on ETV:IDS and just to highlight that we have a number of other programs now using the transport vehicle and announcing today that ATV:TREM2 is officially in IND-enabling studies as well as PTV:Progranulin IND-enabling studies. You can see a number of other programs here utilizing the transport vehicle technology. So now I'm going to switch gears for the last 10 minutes and focus in on the LRRK2 program. And in particular, a summary of where we are with both molecules and what to expect going forward in terms of the decision between DNL201 and 151. I'll begin here with several very important points. The first is that LRRK2 is a very -- is an excellent drug target in Parkinson's disease for a number of reasons. I think the most compelling of which is the genetic background and the genetic support for the target. It was originally discovered there were mutations in LRRK2 that are kinase-activating that increased risk. And now we've seen a number of mutations, again, kinase-activating, even in different domains that ultimately activate the kinase that increase risk of Parkinson's disease. I just want to highlight, last week, in addition to the papers that we published on our ATV platform and on the transport vehicle platform, there was a paper published in Nature Medicine on the effect of LRRK2 loss-of-function variants in humans. And this is very important for our approach. These loss-of-function individuals are healthy. We see no deficit in lung or kidney function as published in the paper. In addition to that, we have now 9 months of chronic dosing in nonhuman primates. This is with DNL201, and we'll soon have that obviously with 151, and did not demonstrate effects on pulmonary function. There's also a recent paper in Science Translational Medicine on pulmonary function is unaffected in nonhuman primates following LRRK2 inhibition. This is with the Fox Foundation and collaborators, including us. And then finally, in terms of our clinical data, we can achieve greater than 70% inhibition at trough for up to 28 days in healthy volunteers and patients, and it's well tolerated. And this is now with 2 molecules, both with DNL201 and 151. On the left-hand side, just a general comment, there's broad therapeutic potential. So our goal in inventing our inhibitors were actually to have equal potency against the mutant form as well as the wild-type form, in part because there's growing evidence that wild-type LRRK2 is activated in idiopathic Parkinson's disease. And in fact, our development approach is going to be testing both LRRK2 mutation carriers as well as idiopathic patients, in part because of the data collected by others. And in fact, in our own experiments, we can also see improvement broadly in lysosomal function even in, for example, a Gaucher patient fibroblast as shown in the bottom. So with that in mind, just a reminder that LRRK2 mutations increase kinase activity. And as a result, lysosomes become dysfunctional, likely in neurons and other cell types, dysfunctional lysosomes lead to eventual neurodegeneration of dopaminergic neurons. Our goal is to inhibit LRRK2. And we've set basically a low bar and a high bar for the percent inhibition we want to achieve. Based on the genetics, there's roughly a twofold or 2 to 3 fold increase in kinase activity. So by inhibiting by greater than 50% at trough and maybe 60 average percent -- or 60% on average, we can basically bring LRRK2 activity levels back to normal levels. We've also set a high bar in which we've looked at basically biomarkers of the lysosome, and they're achieving greater than 70% inhibition, allows us to correct lysosomal biomarkers, both in the periphery and in the brain, as I'll show you in the next several slides, in patients. So focusing on DNL201 and then highlight the healthy volunteer data on 151 and then finish with the decision between both molecules. So this is data that we've presented previously, showing target engagement, looking at phosphoserine 935 LRRK2 pathway engagement, so the downstream Rabs. And just a note that LRRK2 phosphorylate Rabs, and Rabs regulate membrane trafficking and the Rabs that are phosphorylated by LRRK2 specifically regulate lysosomal function. And you can see on the right-hand side that both at the low dose and the high dose, we're able to modulate lysosomal function in the periphery. And in fact, these lysosomal markers are elevated by about 60% to 70% in LRRK2 carriers, and we can achieve that level of reduction with our inhibitors. But to look at markers in the brain, we decided to focus on LRRK2 specific markers. So in this particular experiment, we compare LRRK2 carriers versus idiopathic Parkinson's disease, selecting just for biomarkers that were upregulated in LRRK2 carriers. And we saw a subset of lipid-based lysosomal biomarkers that were elevated in LRRK2 carriers. And these biomarkers are all reduced with LRRK2 inhibition, so showing CNS target engagement. And also, that broadly speaking, LRRK2 again is modulating lysosomal function. So similar to what we show with 201, with 151, we're also able to achieve robust inhibition greater than 70%. Again, the high bar being reducing BMP levels equal to the amount they're increased in LRRK2 mutation carriers, and we can achieve that with DNL151 as well in healthy volunteers. The next step for us is basically to get similar data in patients and then select between these 2 molecules going forward. Just want to highlight that DNL201 is well tolerated at the low dose. And at the high dose, we saw a higher incident of moderate adverse events. These are headache and nausea, likely related to DNL201's off-target effect of phosphodiesterase inhibition. With 151, we have not identified a dose-limiting toxicity, and hence, the goal of continuing to dose escalate with 151. And in fact, that's what we've been doing in the clinic in the first half of the year. We've paused some of these studies at higher dose arms because of COVID, but have a set of data in the coming month or 2 in which we can make a decision between DNL201 and 151. In addition to this, we've now established a broad network to enable global enrollment of a Parkinson's study and specifically focusing on LRRK2 carriers. We entered into a partnership with Centogene in terms of genotyping Parkinson's patients. And our goal in the Phase II/III study is again to look at carriers as well as idiopathic Parkinson's disease. So with that, I just want to summarize what we plan to show in the coming months and year here at Denali. So the first is to select our LRRK2 inhibitors, I just mentioned between DNL201 and 151, and begin the process for basically initiating our Phase II/III study. We want to establish biomarker proof-of-concept in patients that have an interim read on the ETV:IDS potency in humans using biomarkers and specifically Glycosaminoglycans. We'd like to transition DNL343 into patient studies, and this will be dependent on our ability to reinitiate the healthy volunteer study to identify the right dose to go forward. And then we will read out data on our ALS and Alzheimer's studies and make a decision on the path for DNL747 in the RIP kinase program more broadly. Finally, our BBB platform with the publication as well as the progress on ETV:IDS, we like to establish more broadly proof-of-concept for the platform, again, using Hunter syndrome. And we've now initiated additional IND-enabling studies, including ATV:TREM2 for Alzheimer's disease and PTV:Progranulin for FTD. And with that, I thank you and look forward to meeting many of you throughout the day, and we've met several this morning. Thank you so much.

Ryan, we have 2 minutes, if I get to sneak those in, our tech people say I do. On LRRK2, even though you paused some of the high doses for COVID, would there be information enough to pick one of the high doses if you wanted to take that forward regardless just based on biomarker data?

Yes. The short answer to that is yes, because we're able to have -- we have a large number of doses in the healthy volunteers. What's actually been really interesting is that the healthy volunteers and the patients, at least for DNL201, and we'll wait to read out data in patients for 151, correlates exactly in terms of target engagement. So we don't see a shift in potency and the safety profile, we'll read that out with 151. We have been able to dose a number of patients with 151. And so our goal is that we'll -- in the totality, we'll have enough data to basically stay on critical path while, in parallel, continuing to do experiments. And that's the challenge with COVID, right? You need to basically make progress and also be the first in the queue, ready to go when these trials reinitiate.

But if the biomarker data is there, you could pick a high dose, just because you have a safe and well-tolerated [indiscernible]?

That's exactly right, Mike. And I think the other point here is we're running our trials in Europe, and it seems that some of the sites where we're running these trials are at a point where they're getting ready to get going again. And so that's great news, but we'll see how that continues over the next several months.

Makes sense. Good.