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

All right. Good morning. Looking forward to today's discussion around the work we're doing here at Denali Therapeutics. I'm Ryan Watts. I'm the CEO of Denali Therapeutics. And for those of you that have joined the call today, there will be a slide deck distributed on the webcast, and I ask that you look to the bottom left-hand corner for the slide numbers. I will reference those slide numbers as I go through today's presentation. So as a reminder, Denali -- starting with Slide 2, Denali was founded to work on neurodegenerative diseases, and our purpose is to defeat degeneration. You'll see on this slide, basically 4 major areas: rare neurodegenerative diseases, ALS, Parkinson's and Alzheimer's. And in fact, we have either ongoing or completed clinical trials in all 4 of these areas. I will share with you today some of the data that we've generated in Parkinson's disease as well as an emphasis on our approach for engineering large molecules for rare neurodegenerative diseases and also beyond in Alzheimer's disease and frontotemporal dementia. If you go to Slide 2 -- sorry, to Slide 3. Denali is built on 2 platforms. We have the degenogene biology platform. This basically is using human genetics to identify targets and pathways that we go after our therapeutic approaches, and we have 3 pathways: lysosomal function, glial biology and cellular homeostasis. These pathways are represented in Parkinson's, Alzheimer's, ALS and rare neurodegenerative diseases. We've built biology teams around these 3 pathways and have developed biomarker efforts also to delineate these pathways in the clinical settings. The second platform is our blood-brain barrier technology platform shown on the right-hand side of Slide 3. Here, it's broken into 3 areas. One is small molecule engineering to cross the blood-brain barrier, and that will be represented by our LRRK2 program, which I'll present today as well as large molecule engineering, and this is being antibodies, enzymes and proteins across the blood-brain barrier. The third area is AAV. That's an area that we began working on about 1.5 years ago and have made significant progress in that area, but we'll not share that data until we near clinical studies. But our goal there basically is to engineer proteins that are expressed in the liver that can cross the blood-brain barrier and engineer AAV capsids to cross the blood-brain barrier directly. We now turn to Slide 4, which is our journey from the founding of Denali. And in fact, on -- tomorrow, Denali was founded exactly 5 years ago. And as you can see at the top is a number of our filings, including very recently, towards the end of last year, filing ETV:IDS, our first large molecule engineered to cross the blood-brain barrier for Hunter syndrome. As well as filing IND and CK filings for eIF2B, which I'll discuss in some -- later on in the presentation. We also recently presented data on DNL151 and DNL201, which are small molecules targeting the LRRK2 kinase in Parkinson's disease, and I'll share that data today. The company is also built on a number of partnerships. We've entered into over 30 partnerships. Highlighted here is a handful of partnerships, both incoming and outgoing, including a partnership on our blood-brain barrier technology with Takeda, on 3 targets that are enabled by our transport vehicle technology, and a partnership with Sanofi on RIP kinase inhibitors for both central and peripheral inflammatory diseases. We now turn to Slide 5, the pipeline. And here, I just want to emphasize that we have 5 -- 6 clinical-stage molecules in 5 programs, including 2 molecules for LRRK2 as well as 2 molecules for RIPK, 1 targeting central nervous system diseases and 1 targeting peripheral diseases. Today, I'll focus on the 2 programs at the top here, the LRRK2 program for Parkinson's disease and IDS for Hunter syndrome, both of which are either beginning clinical testing in the case of DNL310 or ETV:IDS or nearing now late-stage clinical testing for the LRRK2 program. In terms of our goals for this year, I'll highlight the goals and the impact of the COVID pandemic on these goals. So let me begin here by saying that it's our goal in terms of LRRK2 to select either DNL201 or 151 to advance into late-stage studies. We will make that decision midyear, and that has not been delayed. However, higher doses of DNL151 have been paused for both healthy volunteer as well as patient studies. And that's essentially exploring the window -- the safety window more broadly. That being said, we'll have sufficient data midyear to make a decision between DNL201 and 151 for LRRK2. In terms of ETV:IDS, our goal is by year-end to basically establish biomarker proof-of-concept, and at this point in time, we're on track to hit that goal. Here, we need a handful of patients that have been treated with DNL310 looking at CNS biomarkers showing target engagement and validation of the BBB platform. Sort of the goal represented at the bottom here will also be completed. In terms of EIF2B, that program has been also paused, the healthy volunteer study. And our hopes is to get that up and running as soon as possible with the ultimate goal of enabling a decision to move into patient studies, that will be dependent on our ability to reinitiate that healthy volunteer study. And then in terms of RIP kinase, we have completed enrollment in both the ALS and Alzheimer's studies, and we'll be completing those studies and ongoing data announcement decision with Sanofi also expected in midyear. So Slide 7 is basically summarizes -- the 2 top bullet points here summarize the impact that I mentioned in the previous slide in terms of our goals. I'll just emphasize on these 2 bottom bullet points here that Denali is based in South San Francisco. So we are adhering to the shelter in place with the majority of our employees working remotely. However, all lab-based employees and activities continue on-site in South San Francisco. Thus, our clinical biomarker efforts as well as our animal model and biology efforts are ongoing. The financial impact is under assessment. Denali has a strong balance sheet with $614 million in cash and investments at the end of February and a cash runway to last at least through 2022. So now we'll turn our attention to our programs, starting first with the LRRK2 program for Parkinson's disease. So we're now on Slide 9, and I will give an overview of the rationale for LRRK2 as it relates to Parkinson's disease. So LRRK2 is what we term a degenogene or gene when mutated that causes -- or is a risk factor for neurodegenerative diseases. And in this case, it's Parkinson's disease. And the mutations in LRRK2, uniformly, there are a number of coding mutations all result in increase in LRRK2 kinase activity. As shown in the middle panel here under the pathway, when LRRK2 is hyperphosphorylated, we see a -- hyperactive, we see a hyperphosphorylation of both LRRK2 itself as well as its downstream substrate or pathway modulators, the Rabs. When Rabs are hyperphosphorylated, they're dysfunctional. And as a result, lysosomes accumulate and become dysfunctional. You see there, BMP, which is a phospholipid represented in lysosomes. And our observation is when LRRK2 is hyperactivated that BMP levels also increase, giving us a measure of pathway target engagement. There are also a number of other lipid-based biomarkers that are modulated by LRRK2, and that will become relevant as I present the clinical data. When this happens within cells, we see an accumulation of proteins, and ultimately, degeneration in the case of Parkinson's disease of dopaminergic neurons. Human genetic variance linked to LRRK2 show, again, an increase in kinase activity. However, there are also genetic variants that are broadly protective, which point directly to a role in sporadic Parkinson's disease. And in fact, in sporadic Parkinson's disease, what's observed is activation of LRRK2 in a region-specific way as -- which is in contrast to the mutation carriers, which will have hyperactivated in LRRK2 in all cells. Moving to Slide 10. Using this as a rationale, our approach is to target both the LRRK2 carriers as well as sporadic Parkinson's disease. Just represent 1 of a number of publications. On the right-hand side showing LRRK2 activation in sporadic or idiopathic Parkinson's disease. I've mentioned previously that LRRK2 -- there are variants in LRRK2 that are also protected. These variants result in decreased LRRK2 protein levels. And ultimately, what we see is when LRRK2 is hyperphosphorylated is lysosomal dysfunction, and we see this as a central pathophysiology in Parkinson's disease, including in sporadic Parkinson's disease. So our therapeutic hypothesis is that inhibiting LRRK2 should be beneficial, obviously, in the mutation carriers that have hyperactivated LRRK2, but potential for efficacy as well in Idiopathic Parkinson's disease. Moving to Slide 11 is an outline of the clinical studies to date and where we are. We have completed the DNL201 Phase I healthy volunteer as well as the Phase Ib study, and I will show some of the data today for those studies. We are also nearing completion of the DNL151 Phase I as well as the 1b patient study. And as noted before, the high doses, which were added at the beginning of the year, have been paused due to the COVID pandemic. However, we have -- we will soon have a significant amount of data for DNL151. Our goal is to make a selection mid-2020 between 201 and 151 and then continue towards the Phase II/III study with 1 of these molecules. So now I'll take a few minutes to highlight some of the clinical data for each of these compounds. So DNL201 Phase Ib, this is the first study conducted with a LRRK2 inhibitor in Parkinson's patients. This is Slide 12. And we have, in this study, both mutation carriers as well as sporadic Parkinson's disease. We selected 2 doses, a low dose and high dose. And what you can see is that with both doses on the left-hand side, we have robust target engagement. The minimum bar for inhibition that we have set is greater than 50% inhibition at trough. And the reason for that is that mutations in LRRK2, result in about 2 to threefold increase in kinase activity. And our goal minimally is to bring LRRK2 kinase activity levels back to normal levels. That being said, the average inhibition that we achieved even at the low dose is greater than 65% inhibition. And at the high dose, we have greater than 75% to 80% inhibition throughout the dosing window. And you can see here both robust target engagement when looking at LRRK2 in terms of pathway engagement. The downstream Rabs, also greater than 75% inhibition at both the low and the high dose. And this correlates with a reduction in BMP, which is a downstream lysosomal biomarker. The reason why this data is relevant on the right-hand side, the lysosomal function data is that LRRK2 mutation carriers in urine have about a two to threefold increase in BMP levels. And you can see that we basically can normalize BMP levels on the right-hand side with inhibiting LRRK2. So this data demonstrates strong LRRK2 inhibition and improvement in lysosomal biomarkers. On Slide 13, we ran an experiment in which we basically compared the mutation carriers to the sporadic Parkinson's disease patients within the same study. So before dosing, we have CSF from sporadic versus mutation carriers. And we looked for a set of lysosomal biomarkers, lysosomal lipid biomarkers that were elevated specifically in mutation carriers. Again, a reminder that all cells in these mutation carriers will have hyperactivated LRRK2. However, in sporadic, it will just be a subset of cells or regional specific activation. So here, we identified a set of LRRK2 specific lysosomal biomarkers that are elevated. And you can see that on the right-hand side, a handful of these biomarkers, which were selected, we can see a robust inhibition bringing them back down to wild-type levels. You'll also note that the sporadic Parkinson's patients do not have elevation in these lysosomal biomarkers when measuring them from CSF. Now moving to Slide 14. I'm shifting here to DNL151. And this molecule was developed in part because DNL201, we had evidence that we needed at least b.i.d. dosing, and we developed the molecule here that we believe we could dose q.d. In addition to that, DNL151, we engineered out some of the off-target effects of DNL201, specifically phosphodiesterases, where we see effects on blood pressure, not surprisingly, knowing the biology of phosphodiesterases. And indeed, we were able to achieve that with DNL151. And you can see here a number of dose levels in the healthy volunteer study, which we continue to dose escalate and a greater than 75% inhibition in target engagement as well as pathway engagement, and similar to DNL201, robust reduction in BMP as we increase dose with DNL151. Now moving to Slide 15. I just want to emphasize here that DNL201 was well tolerated at the low dose. At the high dose, we had a higher incident of headache and nausea, which we believe, again, is related to PDE. And in terms of 151, we have not identified any dose-limiting toxicity. And as a result, we continue to dose escalate both in the Phase I as well as the Phase Ib. In terms of the next step, moving to Slide 16, we're able to achieve both our target engagement as well as pathway engagement and downstream lysosomal biomarker goals for both DNL201 and 151. What's pending now is basically clinical data in Parkinson's patients for DNL151 in order for us to make a selection between these 2 molecules and advance into later-stage trials. We've actively established a global network of -- for our patient enrollment, specifically with Centogene in terms of sequencing potential Parkinson's patients to enroll LRRK2 carriers as well as a broad network of collaborators, including the launch of our own website, Engage Parkinson's, to work with Parkinson's patients and caregivers to enroll these studies. I'll now shift gears to our transport vehicle technology using a large molecule technology to get antibodies, enzymes and proteins across the blood-brain barrier. So turning to Slide 18. This is a diagram of the blood-brain barrier. So the human brain has roughly 400 miles' worth of blood vessels. These blood vessels evolved in such a way that most molecules need to be actively transported across the blood-brain barrier. And in this case, what we've worked on is basically an approach to engineer the Fc region of an antibody, shown on the right-hand side to bind to a blood-brain barrier receptor that is responsible for transporting iron. This binding is in a nonblocking way, and we're able to integrate receptor binding into the Fc and bind to, as you can see here, the apical domain of transferrin receptor. This technology is called the transport vehicle. And in fact, both the discovery engineering as well as the co-crystal structure of this interaction, which I'll show in several slides, will soon be published as well. It's the application of the transport vehicle technology to Hunter syndrome, and I'll share some of that data now. So moving to Slide 19. The transport vehicle enables a broad base of potential modalities, including antibodies, proteins, enzymes. And we have recent data showing that we can transport ASOs across the blood-brain barrier linked to the OTV, achieving greater than 70% reduction in RNA expression for these targets. So I'm going to focus first on the upper right-hand corner, which is the enzyme transport vehicle with the goal of delivering enzymes across the blood-brain barrier. Moving now to Slide 20. This is an example, a crystal structure of an enzyme Fc fusion on the top is Iduronate 2-sulfatase for Hunter syndrome. And then the Fc region, highlighted in the bottom there, is the transferrin receptor binding site for the ETV. On our website, you can see a video of the transport vehicle technology. This is basically a snapshot from that video describing its mechanism for crossing the blood-brain barrier. Now moving to Slide 21. Just a reminder of the background around Hunter syndrome or MPS II, which is caused by mutations. It's a monogenic disease in the Iduronate 2-sulfatase enzyme. When IDS is mutated, you see an increase in heparin sulfate and dermatan sulfate, which collectively are termed the glycosaminoglycans or GAGs shown in the middle panel. This is the downstream pathway. So GAGs are a critical readout of activity of IDS. When GAGs accumulate, lysosomes become dysfunctional, and we see an increase in both inflammation as well as neuronal cell loss. And this will become relevant. I will show data using a biomarker for neuronal cell loss or neurofilament towards the end, showing that we can rescue basically neurodegeneration with ETV:IDS. Existing ERTs are not effective at crossing the blood-brain barrier. And as a result, there are large a number of indications, including Hunter syndrome, where 70% of patients are not effectively treated by approved enzyme replacement therapies. In addition, using the ETV:IDS technologies could be superior to AAV approaches because you'll be able to rescue all cell types, and I'll actually show data here as opposed to a preference towards various cell type rescue with AAVs. So now moving to Slide 22. This is a proof-of-concept. The initial experiment in which we compared ETV:IDS to an activity equivalent level of IDS or Elaprase. This is actually purchase Elaprase. And what you can see on the left-hand panel is that in liver, both IDS and ETB:IDS are both equally efficacious at reducing GAGs or these glycosaminoglycans. However, in the central panel, we see an increase in brain GAGs with no treatment. And in terms of Elaprase, we also see no significant reduction in brain GAGs. However, we see greater than 75% to 80% reduction with ETV:IDS in brain GAGs. And looking on the right-hand side, similar to what I shared with LRRK2 using this phospholipid biomarker BMP, we can see a complete rescue of lysosomal dysfunction. This is after 1 month of dosing of ETV:IDS. Now moving to Slide 23. As I mentioned before, 1 of the benefits of using enzyme replacement therapy that can be engineered across the blood-brain barrier is the ability to rescue across all cell types. And I'm showing you here is ETV:IDS concentration in the various cell types after dosing on the left-hand side. And the way that this experiment was run, we injected animals, allowed the ECV:IDS to circulate and then we harvest brains. This is after 4 weeks of dosing, dissociated neurons, astrocytes and microglia, what you can see is greater than 80% to 90% reduction in GAGs in all 3 cell types. So a cell autonomous rescue of IDS loss of function. So taking this one step further, we can then look at single cells within the brain through this treatment paradigm. And here using super resolution microscopy, this is now on Slide 24, we can mark neuronal cell bodies in green and the nuclei in blue are DAPI. And look for colocalization of basically EPV:IDS as measured with human IgG, and lysosomes as measured with LAMP2. And as you can see in yellow, basically is a broad colocalization of ETV:IDS. And just a reminder that this enzyme was delivered in a tail vein injection, crosses the vasculature, is taken up in-brain and rescues at the lysosomal level. Now going to Slide 25, just showing rather than a single cell across a population of neurons. And again, you can see this colocalization of ETV:IDS and lysosomes across various brain regions. And in fact, we've looked at nonhuman primate as well as the Hunter syndrome mouse model, and we see rescue across various brain regions. This is critical when comparing to, for example, a direct injection of IDS intrathecally, where you have to rely on broad biodistribution and Brownian motion. Here, we can rescue in a capillary-specific way. So this basically shows that systemic injection can effectively cross the blood-brain barrier and then rescue the cell type-specific region. How about rescue of neurodegeneration? So moving now to Slide 26. Here, we treated animals for 13 weeks, and looked at brain CSF -- brain GAGs, CSF GAGs as well as neurofilament. And what we can see is greater than 50% to 60% reduction. This is now at therapeutic dose levels at 1 and 3 mg per kg, anywhere between 50%, 60% or 70% reduction in brain and CSF GAGs. This correlates to complete rescue of neurodegeneration, as you can see at the 3 mg per kg dose on the right-hand side as well as robust rescue at the 1 mg per kg dose. This has set the bar for our acute pharmacodynamic readout in the clinic, in which we're -- within 1 month of dosing, our target is to achieve greater than 50% reduction in GAGs, which would be roughly twofold greater than other approaches that are being taken. We believe that this, in terms of long-term dosing would result in block of neurodegeneration. So with that in mind, I'm now turning to Slide 27 and giving an overview of the Phase I/II patient study, design and objectives. This is a 6-month study to evaluate safety, PK and pharmacodynamic readouts. In terms of the interim readout, we will be looking at acute changes in CSF biomarkers and GAGs with a target of greater than 50% reduction. Our goal is to obtain this data before the end of the year, and this data is critical in terms of read-through for the broader blood-brain barrier technology platform for the transport vehicle technology. With that in mind, let's go back now to slide -- we'll go to Slide 28 and look at the other approaches using the transport vehicle technology, including the antibody transport vehicle first. So our goal here is to deliver antibodies in either bivalent or bispecific format, either targeting a single target or 2 targets. And because of the binding to transferrin receptor being monovalent, we can have sustained exposure and acceptable pharmacokinetic profile. So our lead program in this, moving now to Slide 29 is TREM2. And at the end of 2019, we published a paper in Neuron highlighting a critical mechanism for TREM2 regulating microglial cholesterol metabolism. This is an important insight. And in fact, when Alzheimer -- Alois Alzheimer first described Alzheimer's disease in 1906, he highlighted 3 pathological features. The senile plaques, known as the amyloid beta plaques, neurofibrillary tangles, and fat accumulation in microglial cells. And what we have seen in our study, this is our glial biology pathway team, is that TREM2 loss of function mice when presented with an insult resulting in demyelination of neurons, normal mice will have an upregulation of cholesterol metabolites and enzyme, allowing microglia to turn over the cholesterol that accumulates within the microglial cells. However, TREM2 mice are incapable -- TREM2 non-GAG mice are incapable turning over these lipid accumulations. Now interestingly, the mutations in TREM2, the risk factors for Alzheimer's disease are loss of function mutations. And as a result, our goal is to activate TREM2 in the clinical setting. So shown on the bottom is a comparison of a control ATV, antibody transport vehicle; a standard TREM2 antibody given at 50 mg per kg and ATV TREM2 given at 1/5 the dose. And we can see an upregulation of a homeostatic microglial marker TMEM119, and on the right-hand side, a robust upregulation of cholesterol metabolites and enzyme here in wild-type mice. This data shows that ATV, not only is able to cross the blood-brain barrier and activate TREM2, but we get a robust downstream activation of the pathway, which we believe will be protective in Alzheimer's disease. Also in terms of comparing to others that are using standard antibody approaches, both the increased uptake as well as binding to transferrin receptor with TREM2 shows that we can more robustly activate this pathway. Now moving to Slide 30. I'll highlight briefly the protein transport vehicle and then end with a summary. Here, basically, we've used progranulin to link to the Fc, making an Fc fusion. Moving now to Slide 31. And what we can see, similar to what we saw with IDS is rescue in liver with both a standard progranulin Fc fusion as well as PTV progranulin. But on the right-hand side, only PTV progranulin rescues robustly in brain. And here, we're using normalization again of a lysosomal BMP biomarker. So moving to Slide 32. In summary, we have a number of ongoing programs, and highlighted today is ETV:IDS, ATV, TREM2 as well as PTV progranulin. I've commented on the utility for using the transport vehicle for oligonucleotide transport. And our summary here is that basically, the data for ETV:IDS will be critical in validating the transport vehicle, and now, its application across a number of targets, including an ongoing partnership with Takeda in which we're -- they have selected both TREM2 progranulin as well as Tau as target in that partnership. So I can end now on Slide 31, summarizing that even in the context of the COVID pandemic, we do have some delays in programs. However, our ability to make decisions and generate data, either with studies, which have completed or ongoing studies, we'll be able to hit the majority of our goals in 2020. And we look forward specifically to validation of the BBB platform as well as advancing our LRRK2 inhibitor Phase II/III studies. And with that, I thank you for your time.

Thank you, Ryan. Thank you for presenting. Have a good day.

Great. Take care.