Denali Therapeutics Inc. ($DNLI)

We're going to get started on our next session. I'm Andrew Tsai, senior biotech analyst at Jefferies. Thanks for joining me. And it's my pleasure to have Ryan Watts, CEO of Denali, joining me today. Welcome, Ryan.

Yes. Great to be here, Andrew.

Of course. So for those in the audience less familiar with the story or revisiting the story, can you maybe help us level set where Denali is, what programs you might be -- there's a lot, but what programs you're working on and the milestones over the next 12 to 18 months could be very helpful.

Yes. Thanks, Andrew. It's great to be here and definitely a lot has happened at Denali in the last year, in particular, especially with our recent approval of AVLAYAH a little more than 2 months ago. And so just as a little bit of a background, we set out to solve, I think, one of the bigger challenges in drug development, which is the blood-brain barrier, inventing medicines that readily cross the blood-brain barrier, especially biologic medicines. It's an area where 11 years ago in the founding of Denali, not many companies were working. And in fact, I think it was in the late 1980s when the first idea of engineering a molecule to cross the blood-brain barrier using some type of natural transporter was proposed, but it's taken from that point till now to actually get the first FDA-approved medicine. So over the last 11 years, we've invented a platform, which we call the TransportVehicle platform. It's one of these broader categorically brain shuttle technologies using a natural transporter known as transferrin receptor to cross the blood-brain barrier. And looking back now just 11 years ago in the founding of Denali and 20 years ago when we first started working on blood-brain barrier technologies, it's amazing to see how this space has so rapidly evolved and is quickly becoming, I think, the next really hot area in biotherapeutics. For us, the proof of concept was in enzyme replacement therapy. And I think the reason we chose that as the first place is that enzymes have had a very high success rate. There are a number of enzymes that have been approved, I think, about 85% success rate in developing enzymes for enzyme replacement therapy, but those enzymes do not treat the neurologic manifestation of disease. And Hunter syndrome was the first of now a number of areas that we're going to test this hypothesis that we can take an existing enzyme, engineer it to cross the blood-brain barrier and effectively treat neurologic manifestations, but also whole body. And that's going to be important as we talk about AVLAYAH and its differentiation from potential standard of care. And so we set out to do that 11 years ago. And for me, it's been a 20-year journey working on blood-brain barrier technologies. And so the approval on March 24 was, I think, a really key step for us. But that approval is just the beginning, obviously, especially for the patient community. But within 2 weeks, we had filled the channel as we call it. And within 3.5 weeks, the first patient was dosed in the commercial setting. And so we are ready to go and have been very excited about the launch of AVLAYAH. I think the reaction from the patient community has been twofold. The first is great enthusiasm. It's the first medicine in 20 years. It's ironic. 20 years ago, we started working on blood-brain barrier. And then it took 20 years to get the first approved medicine that treats neurologic manifestations and the whole body. But also a little bit of disappointment because there is an adult segment in which the label does not cover. And so a lot of enthusiasm broadly throughout the patient community. So what we've seen is a pretty significant engagement from prescribers and obviously, key opinion leaders and families, and it's been a really exciting time at Denali. So that's where we are today. So the question is what's next? And I think the next 6 to 12 months is also very data rich. We have -- we'll have our second BLA filing for Sanfilippo. We also have begun studies in Pompe disease. So this is another enzyme replacement therapy. And then we have a progranulin replacement therapy. So very similar to enzyme replacement therapy. So a number of medicines. Those are moving forward. I'll put those categorically in the enzyme franchise. But also what we have to think about is the next big wave. And for us, it's Alzheimer's disease. We're now dosing patients with a MAPT oligonucleotide, which suppresses the expression of MAPT, which leads to reduction in tau protein expression. And we'll soon begin testing in the clinic our own Abeta antibody using the TransportVehicle technology. And so if you think about the next 6 to 12 months, you're going to be seeing data across the enzyme platform, but also now in the broader neurodegenerative diseases and specifically Alzheimer's disease. And it's just -- worth mentioning that, that is the ultimate goal, I think, of Denali is to defeat degeneration to focus on the neurodegenerative diseases ultimately, but we also believe the enzyme franchise is a great foundation to begin building our commercial footprint.

Great. So again, congratulations on the approval. It's a huge milestone. And so you mentioned you're differentiated because you can cross the brain, BBB benefit clinical manifestation. So as we think about the standard of care starting with Hunter with this launch, ELAPRASE, I think they're doing $700 million worldwide sales. So how are you superior long story short to that product?

So we obviously have a head-to-head study ongoing now. That's the COMPASS study, which is the Phase III confirmatory study for accelerated approval. But I think it's worth noting, and this data was published in the New England Journal at the beginning of the year. The vast majority of patients in that Phase I/II study switched from ELAPRASE to AVLAYAH or tividenofusp alfa at the time. And basically, what we see is, obviously, the most striking is the normalization of cerebrospinal fluid heparan sulfate. That's evidence of crossing the blood-brain barrier. We subsequently see the normalization of neurofilament, which is a marker of neurodegeneration. So that's highly differentiated. And so recall that most of these patients are on standard of care, but they have elevated heparan sulfate in CSF and in brain. But what's also striking to us is that we can also see a robust reduction in peripheral heparan sulfate. So I think that basically tells you that we're -- based on our dose and also biodistribution that the biomarkers of both the central nervous system, but also the periphery are enhanced with the medicine. And so this is important. And I think we had an important decision to make early on with AVLAYAH, which was around dose. And part of it was, I think, most enzyme replacement therapies experience what are called antidrug antibodies that recognizes foreign -- this is well known. Infusion-related reactions are the most common adverse effect for all of these medicines and often, I think, are limiting for dose, but we made the decision, especially because of our architecture, which is an Fc fusion to really push dose. And the goal here was to get not just superior brain exposure, but also ideally superior reduction in peripheral biomarkers as well, which we believe will ultimately lead to a broader clinical benefit.

Great. And so you've launched dose the first patient in April, we're in June. I must ask how many -- to the extent you can share how many patients have you dosed, I guess, with AVLAYAH? And are they coming from ELAPRASE? Or are they naive patients?

Yes. So I was thinking about this particular question, which we get often. I mean, it's so early in the launch. It's like 9 weeks since approval, although it feels like a year ago now, immediately becoming a commercial company. What we can say is that we -- the interest has exceeded our expectations. So maybe that shouldn't be surprising. We are debating what are the right metrics to share at the right time to basically provide meaningful data to be able to predict the future of this medicine. I think for now, what we've seen is just great enthusiasm across all patient segments. So there are those who are newly diagnosed, which come in, as you know, there are about 500 patients in the U.S. and 2,000 worldwide, but especially switch. So in many ways, this is a switch market. And all patients are, generally speaking, known, at least by age 3 or 4. And what we're seeing is a pretty significant interest in switching to AVLAYAH. But I'd say stay tuned on exact numbers. I think we're just -- we're gathering the information now and looking forward to hopefully a very successful launch.

Great. And you mentioned earlier in the community, there's a slight disappointment that the label does not cover all 500 patients in the U.S. So what does your -- what proportion of those 500 U.S. patients does your label cover?

Yes. So it covers 70% to 75% of the population today. So it's the pediatric population. And if you think about actually Hunter syndrome, it's about 70-plus percent that have severe mutations that invariably lead to severe neurologic impairment. But what's been interesting and I think very striking is that neurologic manifestations is seen throughout the entire spectrum of disease, even in what we call attenuated patients. A good example of that is hearing loss. And that's one of our most striking, I think, data points that was published in the New England Journal was improvement in hearing. And so what the community is generally understands is that this in the label where it states neurologic manifestations, which is linked to the CSF heparan sulfate as the biomarker for accelerated approval that, that basically represents essentially all patients at some point in disease will have a neurologic manifestation if it's behavioral, sleep disturbances, hearing. And so -- but the label itself covers roughly 70% -- 75% of the existing population, so very significant.

And in the meantime, like you said, you're running a confirmatory study called COMPASS. I think the data reads out second half 2027. That's a head-to-head study versus ELAPRASE. That should also -- not only can you prove superiority, but I think -- can you confirm it opens the door to all 100% of the U.S. population.

That is the goal of COMPASS ultimately. And there really -- there are 2 cohorts. There are those that are sort of, I would say, categorically severe neurologic and then those that are -- have been categorized as attenuated, again, all of which experiencing neurologic manifestations. The attenuated patients are generally adult patients. So in the Phase I/II, we did not enroll any adult patients, although now there are patients who are, by definition, adult that are on AVLAYAH because they began treatment at, let's say, age 14 or 15 and now they're 17 or beyond. So that is the goal of -- ultimately of COMPASS. But I think really where COMPASS has its greatest value is likely ex U.S. is in Europe and unlocking Europe, where it's not just biomarkers that drive approval, but also the correlated clinical benefit that's observed.

Right. And totally understood your sales reps cannot market superiority, but this -- any NEJM article is quite compelling. Is there any way some personnel on your team can flag that to people to convince physicians, patients that you have a differentiated product. I don't know how it works.

Yes. So I think the most important thing for us is when you look at the label is the fact that 9 out of 10 patients have normalized CSF heparan sulfate. I always worry about the tenth patient. In this case, they also have like 85% reduced CSF heparan sulfate. So it's just near normalization. That's a really important piece is that there's no other medicine that has shown that level of heparan sulfate reduction and clearly normalization in cerebrospinal fluid. And that also points to the broader validation of the TransportVehicle technology. So for us, we can just focus on really the data that we have, and it's a very distinct profile from standard of care.

And in terms of -- so there's one volume and then there's price as it relates to sales. So on the pricing side, can you give us a sense of just based on the average patient -- this is weight-based dosing. So the average patient on AVLAYAH, how much it could cost per year and how that would compare to the average price per year for ELAPRASE?

Yes. So the wholesale acquisition cost is $5,200 per 150 mg vial. So this is the vial. Most patients will need 2 to 3 vials per infusion depending on weight. So if you do the math, a 10-kilogram patient will be about $270,000 per year and a 30-kilogram patient will be roughly $800,000 per year. So that's -- obviously, that's the exact pricing dynamic. The medicine is priced at a premium, recognizing the invention and the fact that we're treating not just whole body, but also the brain, which is, I think, a key step here. It's also priced in line with more recent enzyme replacement therapies that have launched. And just if you're doing the math, the roughly 30-kilogram price for ELAPRASE is about $500,000 per year.

Okay. Very clear. And so -- gross to net, any color around that inventory build for Q2? Any color around that?

Yes. So again, I think we -- it's very early days. It depends on the payer, obviously, Medicaid versus private payers. We've -- we're just -- policies are being put in place now. Just reminding myself that we got approval 9 weeks or so ago. And so just stay tuned. But the math is the math. If you know the 340B institutions and the discount that's required, you can do that math. It depends on if it's pediatric only or adult, but you can start -- I mean that will make sense over time, definitely as we launch and you see more data. I would not over-index on 2026. This is a foundation building year for the launch. We're excited and we're enthusiastic certainly by the interest in the community, but it's going to take time, especially as you get payer policies in place.

Okay. Well, things seem to be in the right direction. So congratulations. So then shifting to Sanfilippo then the next leading program that you have, you're filing next year, that's contingent upon full 20 patient worth of data. And my understanding is all those patients should complete around September of this year. When would the data be released though?

Yes. So the data cut is not too dissimilar from when we had the data cut for the filing for AVLAYAH in the sense that we got data sometime in September, October. Obviously go through the database lock cut evaluation. We're aiming towards a WORLD presentation would be February of next year and then filing next year. Right now, the rate-limiting step for us is manufacturing. We made the decision to onshore this manufacturing. So AVLAYAH is manufactured by Lonza and DNL126 for Sanfilippo, we will manufacture ourselves, even commercial product here in the U.S., which we're very excited about. That's the key there. The -- as we've reported previously, and we'll probably get into some detail, on average, we see about 80% reduction of heparan sulfate. We also can achieve normalization both in heparan sulfate, but also downstream biomarkers of lysosomal dysfunction. What's distinct about Sanfilippo is that it is largely a CNS disease more than Hunter syndrome. So there is components of peripheral disease, but it's mainly CNS. And it is a very rapidly and aggressive degenerative disease, especially with the complete loss of function mutations. And so even like at the volumetric level, when you look at MRI, you already see patients before you begin dosing have pretty significant volumetric loss. And so I think with Sanfilippo, it's going to -- we need to get a drug approved and then it's going to be about newborn screening to just get there as early as possible to see a huge benefit in these patients. Hunter is -- doesn't have the same volumetric loss. The reason I just state that is that in addition, heparan sulfate is actually 2x higher in Sanfilippo in brain than it is in Hunter syndrome and probably is just where SGSH is expressed versus idursulfase. Interestingly, they're both -- we're both looking at heparan sulfate as this class of MPS diseases. And so we're really enthusiastic about the initial data. The 20 patients will complete the 49-week point in September, and then we'll begin looking at that data and preparing for filing and presentation in 2027.

Okay. And we've discussed before the next batch of -- technically, the remaining 12 patients to be presented at WORLD are dosed at a higher dose, we should get better efficacy. So to me, the likelihood of success is very high to support an accelerated approval.

And I think just a point related to that, if you look at our presentation from this year, it's what we did with the first 8 patients, it was all about dose finding. It was dose frequency, dose finding, dose levels. And so we started with every other week dosing. And I think what's really striking is that with enzyme replacement therapies, specifically with SGSH, the drug is better tolerated with weekly dosing. It tells you something about building tolerance. And so having more frequent dosing and having drug on board was an essential learning from that initial study. We then kind of went to what we thought would be the optimal dose to weekly dosing, but many of the patients weren't really at that dose when we took the data cut it and presented it. So at this point, obviously, with the 12 additional patients, we'll be using that optimized dose regimen.

Great. And help us frame the patient sample size in the U.S., the peak sales opportunity relative to Hunter?

Yes. I think it's similar to Hunter. The difference is there is no standard of care. And as a result, not all patients are identified like you have with Hunter syndrome. And so there's no existing enzyme replacement therapy in part because it's a largely neurologic disease. And so standard ERTs are not treating the sort of very rapidly progressing degeneration that is observed in Sanfilippo. But we think the numbers will be roughly similar, maybe a bit smaller. It depends on whose epidemiology you're reading, but obviously, having an approved medicine will be really important in this disease area.

And so because there's no approved medicine as we think about the confirmatory study that I believe you said you're going to start later this year. Is that still the plan?

That's right. So we're in start-up mode right now. We've been used -- using the word start, we were selected as -- for the START program by the FDA, which is quarterly engagement with the FDA around designing the Phase III. We think that's very important. Obviously, again, with no standard of care, the dynamic is different than what we've designed for COMPASS. And so I'd just say with that, stay tuned. We continue to engage with the FDA, but we're definitely in study start-up mode for the Phase III for Sanfilippo.

Yes. And so because of no standard of care, there is this potential maybe to be an open label rather than a placebo controlled, which maybe means you can enroll pretty fast and generate the data pretty fast. Is that the guide?

I think what's implied in that question is that patients are excited to join trials where they know they're going to get active drug. And that's generally true. And certainly, that's been the case with the Phase I/IIs for both of these studies for both of these medicines for both AVLAYAH and for 126.

Okay. Very good. And so I guess, later this year, you'll have a progranulin readout shifting gears in FTD dementia. And it's -- I would just like to know what exactly are you seeing with -- what do you want to see to move forward basically in the study?

So our sort of hypothesis with FTD granulin is that it's a progranulin replacement therapy. So take what we've just talked about with Hunter and Sanfilippo and now apply that to FTD where you have a loss of function, although it's a single copy loss of function. And so we published a paper in 2021 in cell describing the mechanism of progranulin in the lysosome and associated lysosomal biomarkers. And so here, we'll be looking at a panel of biomarkers around lysosomal function, both those are what I call the proximal biomarkers and then also distal biomarkers like NfL. Our experience with NfL, you can see it in the New England Journal paper is that at least -- and actually for most likely lysosomal storage disease, it takes some time. So at 6 months, we didn't see a reduction, but by 12 months, we are clearly seeing a reduction. And then by 18 months, we're starting to see patients that normalize. So those are the distal biomarkers, both of which are going to be important also in FTD granulin. I think one -- if you look at that paper, the only cautionary note is that the signal to noise for biomarkers with a single loss of -- single copy loss of function is not as great as with the enzyme replacement therapies in some of these MPS diseases. And so we don't have huge patient numbers, but we do have more than, let's say, like 8 patients or so, which we had in the original LSD studies. And so anyway, so that's -- those are the biomarkers that's what we'll be looking at end of year in Cohort B3.

Okay. Great. And so I do want to spend some time on tau. But before that, you recently had a LRRK2 update for Parkinson's oral LRRK2. Unfortunately, it failed to many people, maybe not too surprising because it did evaluate a broad Parkinson's population, not necessarily just LRRK2 specific. By chance, did you happen to see a signal though in that LRRK2 specific population? Or else, why continue on this other study that you have BEACON to completion, which is evaluating LRRK2 specifically?

So it's interesting. We talked about the last 20 years working on blood-brain barrier. We also in parallel for 20 years, have been working on LRRK2. It was first described in 2004, so I guess, 22 years ago. This is the first medicine to test LRRK2 in idiopathic Parkinson's disease. This is the like, let's call it, the home run study. What we saw is robust inhibition in periphery and brain. And I think it's important to look at the biomarkers that we use to assess that. So one is phospho-serine-1292 or, I guess, it's 935 or something. We -- that particular biomarker, we saw greater than 90% inhibition in the periphery. In brain, we use phospho-RAB and the reason we use phospho-RAB is the same biomarkers you can't actually measure in CSF. And in a LRRK2 knockout, phospho-RAB is reduced 40%. So 100% inhibition of phospho-RAB is roughly a 40% inhibition. So we're achieving greater than 30% inhibition of phospho-RAB. So again, probably about 90% inhibition. So we think it's a good test of the LRRK2 hypothesis in idiopathic Parkinson's. The study itself only enrolled 11 LRRK2 carriers. So it's too few. In fact, really only 10 that have full data sets. One was from the original LIGHTHOUSE study. And so we don't have enough data to be able to determine if there's a signal or not in LRRK2. So that's a very specific answer to your question. I think 5 placebo, 5 treated with data. The BEACON study is only LRRK2 carriers, 50 patients enrolled. But I think importantly there, it's focused on LRRK2 specific biomarkers. And so we want to read -- that data will read out by the end of the year, and we'll understand where we're at from really the target engagement perspective with the medicine. It is not designed to look at UPDRS, unlike this 650-patient study, the LUMA study, which was powered for clinical efficacy.

I see. And then finally, then shifting gears to your Alzheimer's portfolio of tau and Abeta like you said, starting with the tau, it's pretty interesting. Biogen just shared their own tau intrathecal, given intrathecally. I think you have an IV version. And so what is going to be the exact dosing frequency of your IV? And is there a possibility to even go to subcu over time with...

I mean we could have started here. There's so much to talk about in Alzheimer's disease. And certainly, if you look at our portfolio, the most competitive area is tau and Abeta not surprisingly using the transport vehicle technology. So DNL628 is what we call OTV MAPT, and OTV is Oligo Transport Vehicle. It's delivering an antisense oligo to MAPT, delivered IV and/or subcu. The ultimate goal will be subcu delivery for these medicines. We're obviously, again, optimizing dose. Therefore, IV is the best place to start. And our goal there is to see tau reduction with a systemic delivery of an oligonucleotide. And I think the recent data from Biogen is very encouraging. I think it's like early days Abeta, where you're looking for a correlation between reduction in some pathological marker. In this case, it would be tau or tau imaging, leading to some correlation with clinical benefit. I think the numbers are small, so probably looking at all treated patients pooled. The biggest difference between the intrathecal delivery and the systemic delivery is the difference in biodistribution. So with intrathecal, this is our own data, but also data presented by Biogen and others, you have basically pretty significant heterogeneity from patient to patient, but also within the brain. So a lot of intrathecal delivery, you get great delivery around areas that are adjacent to cerebrospinal fluid, but not into deeper brain regions. When you deliver with TransportVehicle technologies, you can actually cross capillary beds, and we've shown this even distribution and even knockdown of expression. So we think that will dramatically reduce the heterogeneity. And also, we don't -- like when we look at monkeys that are treated with OTV, we have even from monkey to monkey, the same distribution in the brain, whereas when we do intrathecal, some monkeys will get good brain distribution, some will get less, some will get none. And even when you're injecting it intrathecally, there's pretty significant variability.

I see. So aside from the dosing convenience, you saturate the brain better, I guess that could lead to better efficacy overall, more patients response...

Yes. When you think of where tau pathology exists in entorhinal cortex, hippocampus, ultimately cortex, there is an unevenness about the adjacency to cerebrospinal fluid. And so it's better to have a technology that gives you that even distribution.

Yes. I guess there's a debate in the community, does -- is it your view stronger tau reduction leads to greater cognitive benefit? Is there too much tau...

Well, I mean, if anything, what you saw from the Biogen data is that, again, taking it at face value, the lowest dose had the greatest benefit. That's probably because the lowest dose had the fewest number of patients and subject to the most heterogeneity would be kind of my explanation, but let's wait until we see the totality of the data. But the reality is that there doesn't seem to be a dose response. So if the higher dose is giving you more tau reduction, it's not necessarily leading to better cognitive benefit. So I just think it's too early.

We'll just wait for the data in July...

Yes and probably need more data to...

And so then Abeta also starting Phase I, this is interesting. So the tau, I think you're starting -- you're doing Phase I in actual patients. And then the Abeta, I think you're starting SAD doses in healthy volunteers first, then MAD dosing in actual patients. Data -- all the data is in 2027. But in healthy, what exactly do you want to see? Why is that important?

Yes. So I think we published a paper in science last year in August that describes our approach to Abeta. And there are 2 major differentiators. One is, again, the biodistribution that leads to even plaque reduction and less ARIA-like effects in mice and hopefully in humans, so less vasogenic edema, which is one of the limiting factors, right? The other is we engineered the molecule to be immune silent when bound to transferrin receptor. And as a result, it preserves any reticulocytes that express very high levels of transferrin receptor. So the advantage of doing a healthy volunteer study is we can very rapidly test that first hypothesis or the second hypothesis around [ RTIX ] in particular, in hematology, and then we can rapidly move to patient studies and look at plaque reduction. So the ultimate goal for differentiation is a safer, ultimately subcu delivery with robust plaque reduction. That's what we're trying to achieve with ATV-Abeta DNL921.

Right. Okay. I think that's all the time we have, but I look forward to connecting with you in a few months, and then we'll talk again. But thank you, everyone, for listening.