Alnylam Pharmaceuticals, Inc. (ALNY) Earnings Call Transcript & Summary

Great. Thanks very much. It's my pleasure to be moderating this panel with Kevin Fitzgerald, Alnylam's Chief Scientific Officer. And on this panel, we're going to specifically talk about Alnylam's efforts in neuroscience as you guys are expecting some proof of principle, proof of concept data by the end of this year.

So maybe, Kevin, to kind of kick it off, right? I think most folks know about the history of GalNAc in the liver, what did you need to achieve to be able to safely deliver RNAi therapeutics that can penetrate and engage the target in different areas of the brand?

Yes. So thank you for inviting me, Paul. And great to talk about this, and it is timely. I think the beautiful thing about RNAi therapeutics and RNAi, in general, is that it's a natural -- naturally occurring process, right? Every cell has Ago2. Right now, our microRNAs are going up and down. And so really, everything that we've learned in the liver about how to make drugs that are safe and long duration of action, which I think will be especially important in the CNS, all of that remains pretty similar once you move to the central nervous system. Now you have to get in and getting across the blood-brain barrier is not easy, right? So we're doing intrathecal dosing. But moreover, if you're going to go intrathecal dosing, which is a little bit invasive, but it's done a lot these days, but you want to do it as infrequently as possible. So the ability to do biannual or annual type of a drug dosing paradigm, I think, is going to be quite important in this space.

Yes. Okay. And I believe you worked on a lipid conjugate, correct? Can you talk a little bit about that? What are sort of the similarities and differences between this and GalNAc? I guess, with GalNAc, right, it's specific receptor-mediated uptake. How does this work?

Yes. So we've talked about our C16 platform recently. Obviously, we have a lot of other work going on in the area to look at other ligand receptor interactions, but several of the targets that we're looking to go after in the central nervous system are actually fairly ubiquitously expressed, right? So that means they're in most of the cell types in the CNS, and that's really where you want to target them. So when we went about this to try and figure out, okay, what type of a ligand would we use here, we came across specific placements of these lipophilic conjugates. We did large screens to find ones that would distribute throughout the CNS, including into the deep brain regions of the striatum. Now not every target needs that, but that was the characteristics of the platform. So what do we want? We needed it to be safe. We needed it to be able to distribute throughout the central nervous system, and they get into many different cell types. And that's the platform that we've started with.

Okay. Okay. And is it driving receptor-mediated uptake? Or is it getting into...

Yes. I mean we haven't published on that yet, although I will say that there are some receptors that will take this up, and there's always some protein binding that we think is important.

Yes. Okay. Okay. You talked about knockdown across different areas of the brain, and this is, I think, in the kind of library of reasons that people have been searching for why some of the success of SPINRAZA hasn't been replicated in more complicated diseases. That's one of the theories, biodistribution, knockdown and deep brain and things like that. Can you sort of speak to the data that gives you confidence that with IT administration, you can engage your target everywhere?

Yes. So we've done preclinical models in smaller and larger species and been able to show with some of the targets where we want deep brain knockdown, 80% to 90% knockdown in the striatum, it looks very similar to several of our targets like the scan and others in the liver. So to date -- and again, we'll have to do the trials, right, to prove it in people. But if you look at our liver experience, we've translated from the primate species quite well into humans. And so we'll be looking to do that with our ALN-APP shortly.

Okay. Okay. Can you talk a little bit about safety, Kevin? I guess, when you give too much of an RNAi compound in the brain, what is the dose-limiting tox that you ultimately see when you push it? And when you think about therapeutic doses, how wide would you say your margin is?

So I think one of the things with the central nervous system is, it's -- you're limited by the volume you can put in, right? So we -- at the top doses we've gone, we're very happy with where we are, and that allows us to go into a clinical trial and the ultimate proof is in Phase I, and that's why you do ascending dose Phase I trials. So we're happy with the therapeutic index, where we are in the primate models and in the rodent models, and now the proof will be as we move into people.

Yes. Okay. Do you want to kind of set the stage and talk about the initial indications you suggested? And then I can chime in with some follow-ups.

Yes. Yes. So we're going into early onset Alzheimer's disease, and our target is amyloid precursor protein. It's a well genetically validated target in that individuals, their families that have mutations in APP or in 2 other proteins called presenilins that impact APP, and those families get early onset Alzheimer's disease. So that genetic connection, which we've always prided ourselves on picking targets that are genetically validated, that connection is firm. Now others have tried other strategies around APP, but one of the things that I think is highly differentiating about what we're doing is similar to our case of TTR, which the mutant TTR forms these kinds of plaques, we think that much of APP, including intracellular and extracellular portions of that protein are important. And a lot of these other approaches really only we're dealing with ratios of AB40 and AB42, but weren't dealing with both the intracellular ability to form plaque. So this thing, when it's mutant, can kill cells even before it's secreted and then also the extracellular portion. So we get all of that at the same time. And so the idea is to change the balance of deposition and clearance. You've got a natural clearance rate. And in disease -- these disease states which you end up with is too much of deposition and not enough clearance. So you want to shift the balance and that's what we'll be looking to do.

Okay. Yes. So this is -- this gets into a really interesting conversation that I think we can try to maybe index to some of the other Alzheimer's mechanisms that everyone listening might understand well. So you're shutting off amyloid, right, like the source. How does that -- in terms of what you accomplished, in terms of downstream levels of toxic amyloid, how do you think that compares to, say, a BACE inhibitor?

So the BACE inhibitors, first of all, that I worked on BACE inhibitors in the past life. The BACE inhibitors -- BACE is an enzyme that actually has activity on a lot of different proteins, right? It's not just APP, so there's a specificity question there. And then the other portion of BACE is -- and we've done some studies head-to-head about the impact on intracellular APP, and it's not nearly as good. And so I think from that point of -- the point of being very specific to just APP and also having a more profound impact on the overall production of APP, both the stuff that's inside the cell, the intracellular domain and the extracellular domains should be more profound. And our preclinical data suggests that, that's the case when we've looked head-to-head.

Yes. Okay. Okay. Do you want to talk a little bit about the mechanistic rationale in CAA? And I guess, is that -- I know you talked about genetic validation in Alzheimer's, right? But with that, you're referring to the -- am I correct that you're referring to just kind of the general genetic validation for some of the rare familial subtypes with the...

Yes. So we know a couple of things. We know that people with mutations in APP or presenilin that impacts APP, they get an early onset Alzheimer's. They get plaques earlier and they get dementia earlier. And so there's a slower process, but a similar process, we think, in all of Alzheimer's. CAA is very different in that there's a subset. I mean, again, there's a so-called Dutch mutation where there's a small group of families where they have a different mutation in APP. And instead of they -- very specifically, these individuals get deposition of APP. So again, the commonality being you get these plaques, but they get them in a different place. They get them around -- their brain blood vessels, and what that causes is micro bleeds and eventually, fatal strokes. And we think that process is going on actually in probably 20% of seniors, and also, a lot of cases, it may be a mixed phenotype where you've got plaque both in the blood vessels, but also within the neurons. And so we're going to start out in the -- around the genetic, we validated subpopulation, and we're going to move out from there.

Yes. Okay. Okay. So the readout that you're going to have this year, is that going to be predominantly an Alzheimer's readout?

So it's a Phase I trial, right? So you're going to get -- the great news with APP and another reason to start there when you're validating a platform, we've got a ton of biomarkers. You can measure APP and CSF. You can measure all different forms of big data. You should be able to see them going down, and you're going to measure safety, right? And so the ability there to know that you've got target engagement, that you got the right doses, you'll see how long this stuff is lasting. All of that's right into our own playbook, right? It's the same thing when you're going in. First, you want to be able to go in a genetically defined population, you want to be able to look at biomarkers. So you know as you go further, you got target engagement. You know the doses to take forward. You know the safety profile and then you proceed from there.

Yes. Yes. Okay. Okay. As you think about biomarkers, I think one of the -- can't call controversy, but questions of any intrathecally administered drug is, is CSF lowering one-to-one quantitatively representative of what's happening in the brain, right, because you're measuring sort of near where you're injecting the drug, I guess. Maybe I'll just let you react to that. Like how should we interpret CSF biomarker?

I mean CSF turns over every several hours, right? So if you're getting a biomarker, that biomarker is coming from -- it's basically a liquid solution of the biomarker coming in throughout the brain. Now if you're going to get -- if you have an ubiquitous target something like APP, if you're getting 80% knockdown, you must be getting pretty good knockdown. That is not to say there's not a little bit of a gradient for maybe where you injected to deeper parts of the brain, but to get to that level of knockdown, you're doing pretty well across the whole brain.

Okay. Okay. So is your hope that you can get GalNAc like lowering in CNS with these first gen?

We've seen GalNAc lowering, and for some of the targets, you'll know to be very target dependent how far down you want to go.

Yes. Okay. And then for the Alzheimer's disease population you're pursuing, it's an early onset Alzheimer's, right? But it's not necessarily genetic. Can you just maybe talk a little bit about what makes that population distinct from on all comers, MCI, AD type?

Yes. So one of the reasons to focus on sort of the genetic and the early onset is, a, the progression rate, but also they don't have -- if you go in an all-cause Alzheimer's trial, and some people will talk about how some of the issues with those big Alzheimer's trials is that a significant portion of those patients probably don't have Alzheimer's. They have some sort of other Dementia. So even if that's 20% of your trial, the statistical power is gone, right? So by going into this early onset population, a, the diagnosis around Alzheimer's is better. You can actually do imaging, and also, they have fewer comorbidities. And so it's the opportunity to do a proof of concept, find the safety of the platform in a much sort of cleaner segment of the population.

Okay. Okay. Beyond safety and kind of production in APP, what are the other biomarkers that are interesting to you in this trial?

So there, we'll be looking at other markers of potential inflammation as done in the other trials, such as things like NFL. We'll also be looking at other fragments of APP, right? So there's both AB40, AB42. There's alpha. So there's a ton of biomarkers that you can look at that we'll be assessing over time.

Okay. And are there any specific biomarkers that can kind of help you tease out this intracellular/extracellular question? Like if you lowered a certain biomarker, it could kind of really help corroborate that what you're accomplishing is fundamentally different than what's been attained in the past?

I mean we've shown preclinically, right, that we take cells from people that have plaques, right? So the so-called IPF, where they've got these intracellular aggregates of Abeta. And we can lower it there and show that it goes away, right? And so I'm confident that if we can get a nice lowering of APP in the CSF that what's going on in the cell is exactly what we want to happen. It will clear over time.

Yes. Yes. Okay. Okay. Great. Assuming that this study -- I don't want to say assuming, if the study is positive, which would be great -- I don't want to take that for granted, what would be the next steps clinically for these different 2 specific populations, the early onset Alzheimer's disease and then the CAA? And how would you envision -- I mean, I guess, in Alzheimer's, would you be leveraging the same endpoints as everybody else? And then in CAA, how would that regulatory development?

Yes. I mean, I think we'll cross that bridge when we get there into Phase II. Let's show a nice, safe Phase I, show target engagement and then we'll have a lot of different types of designs that we'll -- I'm sure we'll debate for quite some time before we come up with the final design. But this year, we're going to show that the platform hopefully works, and it's safe, and it's doing similar things than what we've shown in the liver.

Yes. Okay. That's very exciting. What do you think, Kevin, in terms of what the dosing interval might look like? Like what work have you done to characterize the elimination half-life from CSF and the durability of...

Yes. So our durability -- and we've shown some of this data at a couple of conferences with APP and other targets. Again, it's very similar to the liver where -- we're anticipating biannual, maybe annual dosing, right? And I think the proof will be in seeing it, but certainly, from our monkey data as we do modeling across that dose interval, I think we'll be probably biannual to annual is what I'm guessing at this point. And so far, when we've translated from nonhuman primates to humans at least from our liver work, we get better in humans and will remain to be seen. And that's what we're looking for, but that's what I'm hoping and anticipating.

Yes. Okay. Okay. Great. Let's see, anything else you want to highlight on brain before we talk about muscle briefly?

I guess the other couple of programs that we have ongoing. Obviously, we're very excited about our SOD1 program that we're working on with Regeneron and was quite sad to see the other SOD1 trial not helping those patients. We had another announcement out today around C-9. So that's a patient population that really needs a new therapy. So we're very excited about moving forward that drug with our partners at Regeneron.

Yes. Okay. Do you -- are you not discouraged by the failure of that program?

In some ways, I'm encouraged that there seem to be some signs that, that drug might have been having some impact, and we're excited about the target. And we believe that we'll be able to knock it down lower and maybe do it in a way that we're hoping is safe, right? And so if you can achieve both of those, then I think you have a good shot at helping those patients with an area that just has huge unmet medical need.

Yes. Okay. Okay. Great. Maybe switching gears. Do you want to just talk a little bit about what you're doing to try to deliver RNAi other tissues like muscle?

Sure. So like I said, the beauty of RNA interference in Ago2 is that it's conserved across all the organs in your body. It works in every organ. So then it becomes a bit of an engineering problem of how do you recapitulate what you've done in liver and maybe now in CNS into other organs. And so we're taking an approach where we know how to make really specific, really efficacious RNAs and now it's about ligands and receptors. And so we're taking a very systematic approach to looking at each individual organ that we're interested in and tissue and the cell types within those tissues to find ligands and receptors that we think are going to be able to carry our material in. And also keeping an eye on all the things that you want for a drug. Is it scalable? How hard is it going to be to make? How expensive is it going to be to make, and all of those things that go into making safe, efficacious and good drugs. And so we have a broad front on that around different ways of making ligands and receptors. We started out with essentially small molecules, GalNAc as a sugar. C16 is a small molecule. We like that approach. It may not be the only approach. Some groups are using antibodies, and we're looking at that and then somewhere in between is peptides.

Right. Right. Okay. I don't know how specific you can get, but do you have a view on all the work that's going on with transferrin receptor in muscle and whether or not that has GalNAc-like potential and could be interesting to Alnylam?

I mean I think transferrin is an interesting receptor from multiple angles, and it's certainly something to stay interested in. I don't -- the proof will be in people being able to do it. There's people using antibody approaches and other approaches, and I hope it works. It gets pretty big and complicated. But I think the proof there will be, I think we'll know somewhat soon around that receptor. It does have some of the properties that you want. On the other hand, it also -- some of the properties it doesn't happen is that it's a pretty important receptor for normal function. I mean I think one of the things with GalNAc is that it's got about 1 million copies per cell in hepatocytes. So it has a lot of excess capacity.

Yes. Yes. Okay. Do you have a view on some of the muscle diseases like myotonic dystrophy, the target is in the nucleus of the cell, something like that inform whether or not you're interested in pursuing that as an indication for RNAi?

So certainly, we look at that, and the question is always whether -- if you look at every transcript starts out in the nucleus, right? And so there's a lot of trafficking back and forth. And so even if there's a little bit of a nuclear depot, sometimes you can just -- sometimes the kinetics is a little bit longer, but you get where you want to go. So I think, obviously, that shows up as we do our target validation activities as to whether it's something of high interest or lower interest for us.

Yes. Okay. Okay. When might we see the first muscle program or IND? I know you did this...

Yes, we're not guiding on that at the moment. Like all of these things, I want to make sure we get it right. And there are a lot of things that go into that in terms of getting the right receptor, getting the right ligand and the type of ligand that you want to ride with for a very long time. So I think those all come into play.

Okay. Okay. Well, maybe since we have a few more minutes, do you want to just briefly talk about vutrisiran and Stargardt? I think that mechanism isn't totally intuitive to people on paper. Like it just -- but it actually sounds like there's a real good rationale. So I'd be sort of curious on your take there and...

I mean it's not really a CNS disease.

I think the eye is probably more similar to brain than muscle, but yes, if you don't want to, that's fine.

Yes. No, I think I'll probably pass on that one. I love the indication, and obviously, the ability to lower vitamin A, right? And some of the properties of that disease is a deposition in the eye, right, over time. And so the ability to slow that down, I think, is an important approach. And there are others out there using small molecules to try and do something similar.

Okay. I wasn't trying to blindside you. I wasn't...

Yes. Yes. No, I understand.

I wasn't going to ask 10, 6-minute walk questions as badly as I would like to, but okay. All right. Well, thank you, Kevin. I appreciate it, and thanks, everybody, for joining. And so we look forward to the first data in the brain by the end of this year.

Awesome. We're looking forward to it. Thanks for having me.

Yes. Thank you. All right. Have a good one.

Bye.