Arvinas, Inc. (ARVN) Earnings Call Transcript & Summary

Everyone, thanks for joining us. I'm Ellie Merle, one of the biotech analysts here at UBS. Very happy to have Arvinas here with us today to talk all things CNS. Joining us from Arvinas is Ian Taylor, Chief Scientific Officer; and Angela Cacace, SVP of Neuroscience Platform Biology. Thank you, guys, so much for joining us. Just a quick disclaimer before we begin. As research analysts, we are required to provide certain disclosures relating to the nature of our own relationships at UBS with any company on which we express a view on this call today. You can find these disclosures at ubs.com/disclosures or reach out to us, and we can provide them to you.

With that being said, maybe just a high-level question for you, Ian. Can you give a high-level overview of Arvinas' strategy for development in CNS indications? And what's the company's philosophy in general when considering indication selection and target selection in this area?

Thanks, Ellie, and thanks for the invitation. It's great to be here. Yes. I think our general strategy is looking at high unmet medical need. And I think the indications that we go into are hand-in-hand with target selection. And we use the same philosophy for target selection that we do on oncology even, which is looking for proteins that are driving the disease. And if you look at our portfolio, at least the main portfolio in neuroscience, talking LRRK2, which we're going to talk a lot about today, mutant Huntington, Tau, synuclein, those are clearly proteins either by dysfunction or overexpression or aggregation that are driving neurodegenerative diseases like PSP or Alzheimer's, Parkinson's or Huntington. And so we're looking for proteins that drive the disease, degrading them inside the cell at the root cause of the neuronal cell dysfunction or death will lead to symptom improvement and really disease modification because there's so few disease-modifying therapies. And we think that the degradation modality, especially with an oral degrader getting into the brain, crossing the blood-brain barrier, getting into deep regions of the brain, and Angela will talk to you about that with respect to LRRK2 in particular, will have a significant benefit over other modalities like antibodies can't get to cell. Like ASOs can get into the cell, but it's not oral. You have to have intrathecal injection. So from a biology, from a delivery perspective, we think degraders really have a high potential in this area. And it really all starts with the targets, and that's our driving philosophy.

Absolutely. And maybe just a high-level platform question. I guess you alluded to this a bit, but I think it's important to dive deeper into this. I guess what are the potential advantages of using the PROTAC modality in the CNS versus other more conventional modalities?

Yes. Again, I think when you compare to the modalities really being able to get inside the cell, degrade the protein, especially aggregated proteins, which antibodies can't get inside the cell, they're only dealing with an outside the cell with -- as the protein is being propagated from one neuron to the other. It might be too late at that point. ASOs can get inside the cell as we mentioned, but it's an intrathecal injection, and it's not so friendly for patients. So being able to use the ubiquitin-proteasome system to remove those proteins hopefully early on, and that's where sort of the biomarker approach catching these patients earlier and earlier, probably earlier than we are now, will really have a disease-modifying effect. So I think it's a combination of degrading that protein that's driving the disease, the root administration, the iterative activity of the PROTAC, you don't need much to get in. That's, I think, really important. Not much needs to get in to cell, which is in the cell is just kind of catalytic or interactivity can chew through that protein. I think that's going to be a big advantage as well for PROTAC.

Absolutely. And maybe for both of you, I guess, what do you think makes a good target in the CNS? And what are you looking for when selecting targets?

I gave my view. Angela can give hers.

Sure, sure. So one of the things that Ian mentioned is we also rely heavily on human data. So human genetics really weighs into our view on the target. He mentioned the potential if there's overexpression of the target. So in this case, LRRK2, there's some evidence for that. And also if there's a way for us to potentially modify the disease by reducing a target that had scaffolding activity. So also in this case, LRRK2 is a large multidomain scaffolding kinase. And so because of that, we believe and most preclinical literature support that reducing the target would be beneficial. So those are just some of the approaches that we take. And just the belief that it's really the protein that's driving the disease in the pathophysiology.

Absolutely. And being able, I guess, to get both the intracellular and extracellular, I think, will be really important here.

Correct.

I guess historically, the way PROTAC has been talked about is that maybe they're larger than a standard small molecule. I guess how should we think about PROTAC's ability to cross the blood-brain barrier and penetrate the CNS?

Yes, we get asked this all the time. Are we using some formulation trick to get it in across the blood-brain barrier? And the answer is no. We've learned a lot about the chemistry of PROTAC, pharmacokinetics, the drug metabolism properties. I think also the way to think about PROTAC is they're chameleonic in nature, which means that they actually can be physically smaller than they appear if you just draw them out. They can collapse on themselves. It's been seen for other very large small molecules or larger-than-average small molecules that if you design them properly, they can actually behave smaller. But I think that's part of the activity in terms of crossing the blood-brain barrier. People think, oh, it's 1,000 kilograms significant across. But it's -- they can -- it folds on itself. It can actually behave smaller. And again, that catalytic activity don't need much to get in to really be able to have an effect in terms of lowering the protein levels, the target protein levels. And I think the selectivity is key there as well. We've talked about it with Huntington. We're just degrading the mutant Huntington, not the wild type. So -- and we've shown that over other PROTACs, very selective molecules, and that helps when dealing with some of these neurodegenerative diseases in particular.

And how should we think about the preclinical data around the delivery and distribution? I guess what drives your confidence that clinically, we will see PROTAC cross the blood-brain barrier?

Yes. I can start. We have a lot of data, particularly in synus as well, which we think are a really good surrogate for the human brain, the human situation. Obviously, we have data in rodents as well. But I think the synu has been proving to be pretty predictive overall for human. And since our data is so strong and Angela can talk about it with respect to LRRK2, we have pretty high confidence that we'll see the same in humans. And of course, we have the readouts that we'll look at early on, part of our trial to see that the compound is getting across the blood-brain barrier into the CSF, for example. So we've got a relatively early read on that. So we're pretty confident.

I love the confidence. And I guess I want to ask once we get to the clinical study and sort of how you will measure this distribution. But maybe first, Angela, could you tell us a bit more about LRRK2 and the connection between LRRK2 and Parkinson's disease biology and kind of your confidence in this as a target.

Yes. Happy to do it. So LRRK2 has been implicated in terms of human genetics. There are both familial and genetic variants that implicate LRRK2 activity scaffolding function as well as increased expression in LRRK2 that drive pathology in disease and in preclinical models. So all of those combined data, but most importantly, the human data compel us in this direction. But there are also really strong nonclinical data that support the role of LRRK2 in pathology and mainly driving neuronal death. And so the scaffolding activity has been implicated in driving neuronal death. Really very nice anti-sense studies have been conducted to show that reducing LRRK2 in the brain of synucleinopathy model reduces new brain pathology, but also protects from dopaminergic cell death in the disease process in those nonclinical models. And so that's compelling. And then there's really nice in vitro evidence that reducing LRRK2 versus inhibiting LRRK2 also is beneficial in rescuing neuronal death. So many lines of evidence in that direction. So we've been really encouraged by LRRK2 as a target? And certainly, the data that we've been generating in our studies also further to drive us to be pretty excited about the differential potential of reducing LRRK2 in terms of engaging the lysosomal system and improving lysosomal function, which we think would improve clearance of pathologic proteins in the disease as well.

Absolutely. And maybe we can dive into this a little bit more. Can you sort of -- just given that there are these LRRK2 inhibitors in development, can you maybe compare and contrast inhibiting versus degrading LRRK2?

Yes, yes. So we've been spending some time on this question since the inhibitors are certainly ahead of us in the clinic. And we're certainly rooting for those inhibitors to have a meaningful impact to patients. But the data that we've been generating pharmacologically to differentiate from the kinase inhibitors has shown that the PROTAC increases lysosomal number, increases the degraded capacity of the lysosome. And we've also been able to show, as Ian mentioned, this iterative catalytic function where the PROTAC has superior target engagement. So we see a 100-fold improvement in potency in terms of degradation of LRRK2. And this translates into a 100-fold improved potency and pathway engagement in terms of lysosome pathway engagement. And this is measured by a direct substrate of LRRK2 software can in the brain of G2019S animals that we've been studying. So we're pretty encouraged by those data, and we've also compared directly for -- along phenotype. So reducing LRRK2 in knockout animals and also kinase inhibition has been shown to induce type II pneumocyte enlargement. So one of the key areas that we wanted to compare right away was looking at the effect of a PROTAC in terms of equivalent target engagement. Do we see type II pneumocyte enlargement? We do see it but to a much lower extent than we did with the Merck kinase inhibitor scaffold. So that gave us further confidence that the PROTAC also has potential safety and tolerability profile that might be beneficial. And so we don't know that for a fact. We're testing and moving forward, but we're continuing to evaluate in safety studies.

Yes, maybe if you could tell us like a little bit more about the, I guess, questions around potential lung toxicity, I guess what the theoretical concerns are there and how degraders can get around this?

Yes. So there's some evidence that the kinase inhibitors redirect LRRK2 to the microtubule. And of course, reducing LRRK2 would not do that. So that's one hypothesis that we might differentiate there. But one of the things that Michael J. Fox Foundation has funded is this investigation of the reported organ systems that might be impacted by LRRK2 kinase inhibition or reduction. And so looking at what might happen if you do reduce the kinase, we've been studying that. And so we've been looking at a reported finding for type II pneumocyte enlargement, which might impact [indiscernible] levels and lung effects. But in some of the studies that the Michael J. Fox Foundation supported, they did not see that. More recently, Merck has published on their scaffold that they do see this type II pneumocyte enlargement with their kinase inhibitor scaffold. And it did induce remodeling of the lung and also collagen deposition. So to date, we have not observed collagen deposition. We do exactly what the Merck team has been looking at, which is doing some histopathology in our studies, in our safety studies to evaluate this. And we have to date not observed that. And so we're being very cautious, and we're monitoring in the way that the Merck Group has monitored. So, so far, we're not seeing that type of effect. And everything that we have observed in terms of the type II -- the modest or minimal type II pneumocyte enlargement has reversed and washed out. So -- but more to come. More -- we're certainly taking it very seriously.

Absolutely. And just from a theoretical perspective, what's the ideal level of degradation? Do you want to fully degrade LRRK2? Is there a sweet spot where you degrade most, but not all? How are you thinking about that?

So we've been thinking about this in a few different ways. Mark Cookson's group has been evaluating LRRK2 levels in post-mortem brain. and what he found in microglia is that there's a twofold increase in expression of LRRK2 in microglia in post-mortem brain. And so that's one line of evidence that 50% reduction may be beneficial from human data. And then in preclinical studies, Laura Volpicelli-Daley at the University of Alabama at Birmingham also showed that 50% reduction appeared to impact pathology in her Parkinson's disease model and also rescue dopaminergic death. And so we feel that 50% is a reasonable target based on the human genetic data, human omics data as well as the preclinical data.

Absolutely. And when we talked about the differences versus like from degraders versus inhibitors, I mean a couple of the components you highlighted were the scaffolding function and lysosomal protein clearance, maybe just help us understand specifically in the context of like Parkinson's, like the importance of this and particularly with LRRK2.

Right, right. So there's many lines of evidence for LRRK2 playing an important role in lysosomal protein clearance of alpha-synuclein in Parkinson's disease. And then there's also data to suggest that LRRK2 may play a role in microglial function and neuroinflammation actually, and there are many neuroinflammatory markers that we've been studying preclinically, including biomarkers that we think are important biomarkers that would give us an indication that we could alter markers of disease progression. So [ GNMPV ] is one of them. We're also looking at -- we'll look at neurofilament light chain as we progress for this impact to neuron guest and seeing if we can move the needle in our clinical trials in patients. And so that's really important. Those are really important disease markers that we'll be tracking when we move to disease.

Absolutely. And maybe, I guess, just starting with the phase I in healthy volunteers. I guess can you give an overview of this trial design and the initial biomarkers of focus there and healthy?

Yes. So our goal there is to run a single ascending dose study to really evaluate safety and tolerability as our primary endpoint that we'll be looking at. And then, of course, we'll be looking at exposure in plasma and in CSF. As Ian mentioned, it's important for us to ensure that we've achieved the exposures that we expect to see based on our nonhuman primate studies preclinically. And then ultimately, we want to show proof of mechanism. And so this would be looking at LRRK2 degradation and also pathway engagement. So we'll be looking at BMP, which is a lipid that is a lysosomal lipid. So we'll also be looking at impact to the pathway of like some functions. So all of those markers will be examined in whole blood, and then we'll also look in CSF as well. We're very excited to see the translation from our preclinical studies.

Absolutely. And maybe just how do you measure sort of the distribution of the effect? I mean, I guess, first, just high level being able to show that you cross the blood-brain barrier in humans, I think, will be important. But then, I guess, what does that look like from measuring the CSF? And then second, I guess, how do you measure the distribution of the degradation across the different brain regions?

Yes. So one of the ways that we've been looking at this is, obviously, in our preclinical studies. We've been looking at distribution of reduction in all of our preclinical species, including nonhuman primate. One of the advantages of nonhuman primates is that they most closely represent what would happen in a human system. And so the transporters are very similar. In rodents, there are many more transporters. So it's actually, believe it or not, more difficult to degrade in a rodent frame than it is to actually in the primate. So I was kind of surprised by that. So you asked me earlier what was one of the surprises. That was one of the surprises. We had potency that was actually superior in the primate brain versus our rodent models. And what we've been doing is looking anatomically at the regions of the brain that we need to impact to treat Parkinson's disease and progressive supranuclear palsy. So as you know, neurons fire within circuit. And so it's really important that we get to the anatomic regions that are important for neuronal function. And so we get to deep brain region in primate. So we're able to see very uniform reduction in our target pharmacologically across the primate frame. So striatum for Parkinson's disease, caudate, we get to the caudate for Huntington's disease. That's an important region for that disease. We also see really very uniform degradation in the cerebellum. And so the dentate nucleus of the cerebellum is a really difficult area for genomic therapies to get to in a primate brain after intrathecal invasive administration. We're seeing very uniform degradation across the cerebellum as well. And so we expect in progressive supranuclear palsy that, that region may actually allow us to impact potentially cerebellar gaze functionally. So we're hopeful that we'll be able to see these types of things in our clinical studies. So that's what we'll be attempting to show. And so I should say also that in those studies, we've run [important to know ] studies so that we can look at what happens to cerebral spinal fluid over time. And what we've been able to show in our studies is that we see a very uniform pharmacokinetic and pharmacodynamic effects in brain versus cerebrospinal fluid. And so that will give us a good -- we believe that the CSF is a very good surrogate tissue for us to evaluate to look at the type of reduction that we'll see in the brain.

Interesting. That will definitely be an important measure. Maybe on that note, I know there was probably only limited that you could say, but I guess how should we think about when we could potentially see initial data from this trial? And I guess what we should expect from the initial update?

Well, we haven't guided to when we'll be completing the study. But I would say that we're dosing healthy volunteers now. So that's ongoing. And hopefully, you'll hear more from us probably early next year.

Exciting. And then beyond sort of looking at the measures in the CSF, are there other biomarkers that you're focused on, both in healthy as well as when you move into patients?

Yes. So we're focused in healthy volunteers looking at whole blood. So we'll be looking at whole blood for LRRK2 reduction. And I mentioned the lysosomal marker phospho Rab, which is the direct downstream phosphorylation substrate of LRRK2. We'll be looking at that. We'll also be looking at urine BNP, which is the lysosomal marker. And then, of course, cerebral spinal fluid, we'll be looking at LRRK2 reduction. We'll also be looking at BNP and other downstream factors in cerebrospinal fluid as we move forward. And that will be more meaningful as we move into patients. And so that's where we'll expect to see more of the disease pathology markers that we think will be relevant.

Absolutely. And I guess if you were sitting in [our street ], how would you compare and we want to like compare inhibitors versus degraders, what are sort of the key measures that maybe that you're looking at to compare when you either get initial data in healthy or inpatients?

Yes, it's really difficult to compare the inhibitor and the PROTAC, I think, in that way. But we've been comparing in our preclinical studies. And again, looking at lysosome function, we've been seeing significant differences between the PROTAC and the inhibitors. So we'll be looking at lysosomal markers as we move forward. And one of -- anecdotally, we've heard that inhibitors have not moved lysosomal markers in CSF. And at least based on our preclinical studies in PROTAC, we have observed changes in some of the lysosomal markers and some disease markers that we think will be really meaningful. So we're really encouraged, about the data about the pharmacology that we observed in terms of potency that looks superior to the inhibitors. So I mentioned hundredfold more effective at engaging the target and the pathway in the brain. So that iterative catalytic activity is really phenomenal at least in our nonclinical studies. So we're looking forward. We're dosing at fairly low doses, and we're very selective. So more to come.

And when do you think the right time course is to intervene with LRRK2 in terms of both Parkinson's as well as PSP? Like is there a time point in the disease course that would be too late? And how should we think about that in terms of longer-term clinical endpoints?

Right. It's a really good question, Ellie. I would say that some of the data that's been really compelling at least in progressive supranuclear palsy is work coming out of ED Jabbari's labs at UCL, where they've been able to show that in a decent-sized PSP cohort that you can actually impact disease progression with the SNP that they've identified in progressive supranuclear palsy. So they've found that the SNP that is related to expression of LRRK2 increases or accelerates time to death by a year, which is very meaningful in that disease population. They were able to replicate it in a separate cohort. So they've looked at over 1,200 PSP patients, and it's meaningful. So I think those data are very compelling. So we're looking to see if we can impact the disease in that way. And then certainly, the fact that LRRK2 is a late-onset Parkinson's disease genetic indicator also, I think, supports that modulation. Even latent disease may actually move the needle for those patients. So we're looking forward to seeing what could happen. And certainly, the impact of overexpression in microglia and suggesting that it's actually the pathologic neuroinflammation process that might be at play as well would give us an indicator that LRRK2 may take the brakes off the lysosomal system, allow for clearance, but also impact immune function.

Absolutely. And last question for you, guys. Moving to your preclinical programs for other targets, you've disclosed programs for Tau, alpha-synuclein and mutant Huntington. Can you tell us a bit more about your progress for each of these programs? And I guess what about these targets do you think makes them attractive for the PROTAC modality?

Yes, we're really excited about the work and the progress that the teams have made at Arvinas in terms of targeting pathologic Huntington. So -- and sparing wild-type Huntington. So we think that's going to be a huge advantage. We are now seeing that we're able to reduce mutant Huntington in the brain of preclinical models. So we're encouraged by that. We also have been exploring oral bioavailability and merging the features of oral bioavailability with targeted protein degradation and selectivity from mutant Huntington for the treatment of Huntington's disease, all of which would be a game changer. So as you know, the anti-sense do reduce Huntington, but they are reducing all of Huntington. And so we feel that by sparing wild type, we may have an advantage there in terms of maintaining cognitive function. It remains to be seen whether that's really the challenge there. But we certainly can intervene early. We're certainly following the work of Roche in terms of how they're designing their trials to go into early Huntington's disease. And certainly, the earlier you go, the more benefit you may see. So that will be really interesting. And for Huntington's disease, since it is a dominant disease familial, there will be an opportunity to potentially treat prophylactically in the future, which I think is also very exciting with the selective degrader. And so then Tau, we're very excited about the progress that we're making degrading Tau. We've stepped back from some of our initial hypotheses around P301L Tau, which is frontotemporal lobar dementia to a more general approach to target Alzheimer's disease Tau. So we've been focusing on pathologic species and modeling using Alzheimer's disease, Tau and seeing really very nice reductions in forms of Tau. And then we're also looking at targeting more of the monomeric species of Tau and then moving to synuclein, that's exactly what we're attempting to do there as well. So as we move forward, we're encouraged by the work that we've been doing. We've been keeping track of biomarkers in the space and making sure that we can, at least in the case of Tau, can we look at target engagement using CSF, but also potentially using PET imaging reagents as well to look at occupancy in a noninvasive type of way. So these are all opportunities for us. So we're very encouraged about the progress that we've been making in all of our programs.

Great. Well, lots of exciting things to come. Angela, Ian, thank you both so much for joining us today. And we look forward to seeing some of this data.

Thank you, Ellie.