Ionis Pharmaceuticals, Inc. (IONS) Earnings Call Transcript & Summary

Great. Continuing with the theme of Oligos in CNS. It's my pleasure to be moderating a panel with Holly Kordasiewicz, who is the Senior Vice President of the Neurology Group at Ionis, along with Wade Walke, Senior Director, VP of Investor Relations. So good to see both of you this morning. I think instead of like I think most folks are well aware of Ionis is like overarching efforts and the great success you guys have had in CNS. So I thought it would be interesting would be to maybe just go through program by program, I'm super interested in the Angelman's program, super excited in tau, some of the other ALS targets you're looking at. And so if that works for you, maybe we can just get into it.

That's great.

All right. So starting with Angelman's. Do you want to talk a little bit more on the background of ION 542 -- 582 excuse me. Like I know there was a lot of work preclinically that I talked to Eric about really trying to find like the right candidate, one with a wide therapeutic index. So what have you sort of described here preclinically before getting to clinic and at a basic science level, how does the chemistry and mechanism here compare to the Ultragenyx molecule?

Yes. So IN 582 is designed with our mo chemistry with -- and with our rigorous screening and optimization process. Now both molecules, ours as well as the 102 are single-stranded alone oligo on [ pay side. ] But there are some key differences. So the competing models for EV3 are a different class of chemical that we chose not to use for our CNS applications. And as you mentioned, we took the time during the design and optimization of IM 582 to really find the best-in-class molecule. The chemistry using is the chemistry that we know how to work with. We have extensive experience within the CNS. For example, 582 is the same chemistry that's gone through that same rigorous design process and optimization is our MAPT tau program that we released a lot of data on last year, and we'll talk about later. And also the same chemistry as our 2 approved CNS drugs, SPINRAZA and QALSODY. So it's something we have a lot of experience with, and then we've done the work to really find the best molecule. One of the key things that we do is we build those human preclinical animal models that express the full human genes so that we can really find it [ in tune ] in those optimal Oligonucleotides. And then also another thing to highlight for this is the preliminary assessment of our early clinical findings with 582, which we released at [ fast ] last year were really encouraging, and no safety synced issues were identified. So we think that we have a best-in-class molecule and that appears to be supported by the preclinical profile and the clinical profile that's starting to emerge.

Okay. Great. So as it relates to this upcoming readout, it's an open-label study, right? And not dismissing the severity of Angelman's as a disease, but in open label trials in neurology, right? You can always have a big sort of placebo effect depending on the clinical measures. So how are you analyzing these data to understand if there's a real drug effect? What endpoints are you focused on? And like some of the population that you're studying, like how heterogeneous is it? Like what would be the likelihood of spontaneous improvement like -- maybe just kind of walk us through your thought process there.

Yes. So the upcoming read that we have is focused on the MAD portion of the study. We've got 51 patients, so it's a decent size, they do range in age from 2 to 50 years. So there is a wide group of patients. This study includes treatment for 3 months with a 1-month follow-up after the last dose. So that's the MAD portion. This is our first study in this patient population. So we looked at many different endpoints, both established as well as more exploratory endpoints. We selected measures that are both the gold standard for measuring change in key areas of functioning Angelman's syndrome, including both direct assessment and functioning, subjective clinician measures and parent reporting and home assessments to try to capture participants in their typical environment. So we really tried to call -- cover all the different ways that we can look at this disease in these patients and how it's performing. Now for how to interpret the data, as you mentioned, it's open label, so we need to interpret it very cautiously. Fortunately, there's a fair bit of good natural history data in Angelman-syndrome patients, and we're able to compare our treatment data to that. So it's not perfect because it is open label, but there is that nice natural history data that we have. And the Angel and trajectory it's fairly stable. They do have some improvements on some of the measures over time, but those are known from the natural history, and so we can compare it to that. Now for endpoints that we're most interested, this is a safety of PK study. So of course, so many things we're focused on our safety and PK and then evaluating the totality of the data to begin to design subsequent clinical trials. And that's, of course, assuming that it does some -- for it's continued development.

Okay. Because of the really wide age range, just to drilling deeper here, like -- does it make sense to look at this by like a age cohort or demographic and compare it to like subsegments of natural history data? I mean 2 to 50 is like kind of incredible.

Yes. No, it does. And we actually have separate cohorts so they're different age groups. So you've got to look at the data that way. But then there's also -- there's natural history data across all those different age groups as well.

Okay. Okay. Got it. And so you talked about this like EEG biomarker Delta Power and not to dismiss its relevance in Angelman's because I want to hear a little more context about that. But like I feel like Delta Power, Gamma Power, I don't know, just being someone who covers a lot of different stuff in CNS, I've heard people speak to these EEG biomarkers and tell me with conviction that they mean like really, really different things, right? So like why is Delta Power a relevant biomarker and Angelman's and how well established is it that it actually correlates with the severity of phenotype or something.

Yes. And there's pretty down the space. So real quick. So in Angelman syndrome patient, they tend to have an increase in slow wave delta and a decrease in faster wave data. And so to kind of give you a reference. Delta is what's typically found in a healthy [ being ] when someone is asleep. And so that's what the Angelman brain is looking like. So it's giving you a read on brain function in a patient. It's something that you can do longitudinals. So that's the sort of thing that us drug developers get really excited about. So we've assessed EEG and natural history studies in Angelman syndrome patients and the Delta Power predicted the Bayley cognitive scores once you control for age and genotype pretty well. And we shared at [ fast ] last year that the EEG, we saw a reduction in the slow wave delta and an increase in faster frequency data when compared baseline 1 month after the last dose and that MAD portion part 1 of the study. So it is something that we see in Angelman -- it's a readout for brain activity, and it is something that we saw move in the map. So -- it's an interesting marker, and this is an encouraging early biomarker. But to put it into context, it suggests that our ASO is engaging the target and having an effect on the disease biology, but at least for me, that's about all it's telling us. Which is I shouldn't diminish that either, that those are good things.

Yes, yes, yes. No, for sure. Are there was like -- so Angelman's are like the genetic level, right? Like you have -- I guess, like for -- I don't know, tell me if this is the wrong word, right? But like silencing of the paternal gene.

Yes.

I know we use silencing in a different context when we talk about Oligos, right? But like do we understand how to measure the degree to which you're reversing this and the degree to which you're getting this to wild-type levels?

Yes. So that's a great question. So our partners at Biogen, Dave developed an assay to look at UB3 in the CSF. So that's a way to look at pharmacodynamic biomarker. There's also the EEG that has been correlated and correlated in the EEG compared, as I mentioned, to the Bayley, so you can use that. So there are indirect measures to give you an idea of how this is performing. But it's not with anything in the CNS, it's difficult to say, yes, this is an exact one to one, and this is exactly how things are moving with those biomarkers because they all are indirect biomarkers.

Okay. So just to kind of round it out on this upcoming readout, it sounds like we're going to get -- is it fair to say that we're going to get like a number of different endpoints, data cuts, like somewhat comprehensive way to kind of contextualize this, assuming the underlying view here is that you're going to move forward. Is that fair?

Yes. We haven't decided exactly everything, but it's really the MAD portion of the study, a whole bunch of clinical endpoints as well as the biomarkers that we've discussed that we've been assessing.

Okay. And on the endpoint side, so I was talking with Priya from Biogen about this yesterday, too. It would be great to also get your perspective. But I mean there's been like -- there's clearly different views from companies in this space on what the right registrational endpoint is and whether the current end points are well suited and sensitive enough, it sounds like your and Biogen's view is that you can work with them to already established. Is that right? And I guess like why is that?

Yes, so that is correct. And based on our previous experience working to develop disease-modifying therapies and pediatric indications. This is the first time we've done that, and that we've done it with Biogen. So we do expect that we're going to need to have a controlled study with a clinical meaningful clinical outcome measure. Now exactly what that clinical outcome measure is, is going to be determined from the data that we're generating now and then, of course, discussions with regulators and KOLs. And so it's important to note that we have generated a lot of data in that open label in that MAD portion, as I mentioned. We're looking at a lot of different endpoints, including the gold standards, but other more exploratory things that could be interesting. And we also have an ongoing long-term extension. So that was originally 1 year, and we've recently extended that for 2 more years. So that's going to give us a lot more experience with these endpoints and a lot more longitudinal data to help us plan those subsequent studies.

Yes. Great. Alright. That's an exciting one. I wanted to briefly switch gears and talk about the Ulefnersen. Is that right?

Yes you did.

Okay. Your right. Thank you. Program -- I'd just like calling these things by their target for everyone listening the FUS program. So we actually have, I think, it's probably not under the radar at all in your world, but on the Wall Street world, it is a little bit under the radar in terms of clinical evidence for this FUS target. So maybe you can talk about that and talk about this sort of program and where things are progressing.

So FUS ALS is like SOD1 ALS. And SOD1 ALS, which is the target of our recently approved drug, QALSODY, is a dominantly inherited form of ALS. Now in FUS ALS, it's caused by mutations in the FUS gene. So like SOD1, our target is very central to disease in these patients. So the goal is that going after the target central to disease, we'll be able to have a significant disease modifying effect for FUS patients. And this program has an interesting back story and that it was originally an early access program that was treating a oligonucleotide in patients. And in those patients, there was observed benefits. So some patients looked like they were stabilizing and then there has also been recorded a decrease in neurofilament in those patients as well. And so in the ALS world, if you see a decrease in neurofilament like we FUS SOD for a SOD1 program, that's really good evidence that suggests you're modifying the underlying disease.

And sort of how big is -- how substantial of a decrease are we talking on NFL.

It's pretty similar to what we saw in SOD1 with a similar slope in the EAP patients.

Okay. Got it. Okay. Okay. Sorry I didn't mean to...

No, no, no. Of course. And so that's why that's what gives us excitement for this program that we have the target central to disease, and we understand ALS because of our QALSODY experience so then this should be hopefully a successful trial.

Yes. Okay. That's great. And -- yes, so what's the design of that study? When are we getting the readout?

Yes. So the readout for that is in 2026. And this is a Phase III study. It's a single study. So first in human and registrational. So this allows us to get through clinical testing into patients as efficiently as possible. One of the key learnings from tofersen was that you need to wait a bit to see a difference between placebo and treatment if you're using a disease-modifying therapy in ALS -- so the study was originally designed with a 6-month placebo control, and we've extended that out to 18 months. And that's really based on the QALSODY learnings. And so then the ideas that can give us meaningful clinical endpoints looking at the ALSFRS.

Yes. Okay. And maybe a silly question, but how and why did this get carved out in the Biogen agreement?

So the way the Biogen agreement works is they get an option on any of our neurology programs and it's in an early stage of research that they get that option, and there's financial implications for it. So they get to decide at that point if they want the program or if they don't. We make a lot more things than Biogen can advance. We have a very prolific R&D organization. And so a lot of things Biogen chooses not to take forward for business strategic different reasons, and this is one of them.

Okay. Got it. How would you characterize the prevalence of FUS versus SOD1?

Yes. So this is a fair most-common inherited form of ALS. It's about 25% of the incidence of SOD1. So it's ultra rare. A few hundred patients in the U.S. and Europe.

Okay, Got it. Alright great. Exciting one. So we have that in 2026. In the meantime, I think the other wholly owned neurology program that you've talked about a decent amount is in Alexander's disease, can you give us the download there and your confidence in GFAP as a target?

Yes, of course. So again, with GFAP, we're going right to the heart of the disease. Here, the mutations are in GFAP. They cause a dominantly inherited fatal Alexander's disease, and these GFAP mutations lead to a loss of myelin. Those are the sheets that cover in neurons. And then that causes a bunch of neurological symptoms, including motor speech difficulties, cognitive decline seizures. It's an interesting disease and target because the underlying mechanism is that loss of myelin because it's potentially repairable. So in preclinical models, suppression of GFAP with our oligos led to a reversal of those myelin deficits and improvement in symptoms. So the preclinical myelins are strong because they mimic the disease. The big question now is, will this benefit translate to patients? So for our GFAP program, like FUS, we've designed a single clinical study with 50 patients and it's registrational. So we want to move this therapeutic as quickly as and efficiently as we can through clinical development, but that also means that we don't have any clinical data on this yet. And so that's what's going to read out next year.

Okay. And again, from a prevalence perspective, how does this sort of stack up using the ALS story as a benchmark?

More than SOD. So a few moderate.

Okay. Great. Any clinical endpoints there?

So clinical endpoint. So this is because it is a complex disease. The main thing that we're looking at is the walk test.

Yes.

But we also have a whole slew of other endpoints that we're looking at. So it's the 10-meter walk test and the change from baseline, and we're doing that at week 61 is the main one. And then we also have subsequent ones looking at most [ bothersome ] symptom. And then global impression of severity and another different endpoints that we can use to help support depending on what the data looks like.

Got you. Okay. And so from a strategic perspective, maybe this is a good question for Wade. These are 2 wholly owned rare neuro programs. Certainly, a different call point than what we're talking about with say, Olezarsen or Donidalorsen -- are these things that you actually would take to market and launch independently? Or would you try to kind of repartner them or renegotiate with Biogen or consider other sort of strategic alternatives?

This is one of our key pillars. We look at programs in cardiovascular disease and CNS as programs that we want to own more of and commercialize ourselves -- so in the case of CNS diseases, we're looking at several of these rare diseases, which on their own don't look like a lot of revenue generation coming in. But when you bucket them all together using a similar sales force, similar infrastructure, commercial infrastructure, given the number of opportunities that we have and the probability success of the chemistry and the technology that we're using, these actually look attractive bundled together. And when you can do clinical trials that are like the Alexander study and the FUS study where you can do a pretty rapid single study to get to registration. The cost of development is also lower. So overall, bundling these together and commercializing them ourselves looks fairly attractive.

Yes. Okay. Very good. Can we talk about tau briefly?

Absolutely.

So I really like this program, and I definitely appreciate the important difference between trying to access tau with an antibody versus an oligo I guess, do you want to talk about the target engagement data you generated and your level of confidence that you're not just kind of reducing tau proximal to where you're injecting the drug, but you're actually reducing it in the key areas of the brain.

Yes. So for ASO distribution and that's distribution from an intrathecal injection up the neuraxis into the brain, which is where we need to get for many of these diseases. We've been working in the CNS with this type of delivery for a while now. So we have a wealth of preclinical data as well as autopsy data from patients who very generously donated their brains after they were treated with Oligo but ultimately succumb to their disease. And that data demonstrates that our ASO is delivering to deeply target all brain regions. Now that data is great, but for our tau program, we have something even better. And that's because here is tau PET and we can use tau PET to visualize tau burden throughout a patients brain. And you can do this with imaging longitudinal over time. So you can see how tau burden changes the disease progression, and then we can also see how it changes with our treatment. And so we were fortunate enough that this came out, and we were able to include tau PET for some patients in our first in human study with our MAPT [ tau oligo ]. So here, we were able to look at baseline and see where in a patient's brain may have tau pathology and then look again after treatment. We did this after the 100-week long-term extension. And in the patients -- and importantly, also in all brain regions that had tau burden, we actually saw tau reversed. So we cleared tau by stopping the production. And that was true in every region that had tau pathology at baseline. So this, of course, is great for the patient's brain. We're getting rid of tau. But it's also a great piece of data for the technology because it shows that we're able to target tau in all those regions and areas that are a problem in an Alzheimer's disease patient brain.

Awesome. And I'm sure the question you regularly get on this one that we've tried to interrogate ourselves is can we be comfortable that knocking down tau will be safe? And when you know if you're just to read about tau on any sort of basic scientific website, it's always referred to is like an integral cellular structural protein. And look, if that's all real the idea of around keeping that is scary. So what gives you the comfort that this is ultimately going to be okay?

Yes. So there's actually a whole bunch of literature on this. So first of all, today, we have no [ on-target ] concerns from the clinical data, and that's, of course, the most relevant data -- but before we even started the preclinical data is actually very supportive of tau lowing. So you mentioned it has a name that sounds scary and it has some things. But if you actually look at the data, there isn't data that says we shouldn't be doing this. So multiple groups that actually made full tau knockout animal, so a 0 tau. And it's important to note that for an oligo, we never get to 0, but if you have 0, these animals are normal. And in fact, if you cross a full tau knockout or an animal with 50% tau levels, so [indiscernible] removed, you cross that with an a beta model, you actually protect against a beta-driven toxicity. The animals do better. So 0 tau is protective and abated model. You can also take tau knockouts and top heterozygous and they're protected against other excitotoxicity insults like chemically induced or genetically induced seizures or brain injuries. So even though it has a name, at least in the preclinical models, there isn't data that supports that this should be an issue and we haven't seen anything from our clinical work. Though, of course, we'll continue to keep an eye on that.

Yes. Okay. And so in the meantime, that's in a larger Phase IIb now. Is that right?

Yes. That is exactly right.

Okay. And no timing guidance there yet on that datas.

Scheduled to complete in 2027.

So still a ways away, but exciting. Okay.

Yes.

Maybe lastly, where is Ionis with kind of the next-gen delivery stuff, right? And I mean begins on a CNS, but like even specifically in muscle, you guys had some really cool DM1 HSALR data like 3 or 4 R&D days ago, what happened there? And as that space heats up with [ Dine, ] Avidity and others, is that something you see yourselves participating in?

Yes. So DM1 and DMD are absolutely on our radar and remain areas of interest for us. The data from the ongoing clinical programs you mentioned are very supportive for the targets. And as I think most folks know, we have a relationship with Bicycle to use their small peptide-based ligands and we demonstrated that these are excellent at improving targeting of RNA therapeutics to both cardiac and skeletal muscle. And in fact, much of our early validation work that we did with our bicycle technology including data in both mice and nonhuman primates was done using DMPK, which was the original target that we were pursuing for DM1. We are advancing our first bicycle delivered therapeutic into the clinic for our cardiology franchise, targeting cardiac muscle with a bicycle-siRNA. And so I should note that bicycle works great with oligos, and it also works fantastic with siRNAs too and so we think that we have -- the bicycle technology has the potential to be best in class, and we're exploring the application of this technology to not only the cardiac muscle, which is our first foray with it. but also into skeletal cell programs and the obvious targets in the neuromuscular space.

Okay. Why -- like why haven't you moved more aggressively in the muscle space here?

So we're just taking the time to make sure that we have the right molecules. And if we do move forward that we have best-in-class molecules.

Okay. And is this something that Biogen like has the rights to or is already like technically opted into because they did DM1 with you before? Or how does that work?

Well, with DM1, I think they had an option for DM1, and I'm not sure that they decided to move forward with it. So I'm not 100% sure if they're still have an option to take forward. But on other ones, they do have an option for [ muscular ] targets.

Okay. Any other thoughts on just kind of CNS delivery stuff like transparent and if this is something you could kind of work on too?

Yes. So I think the next frontier for CNS delivery strategies is crossing the BBB. So we're pursuing crossing the BBB with multiple, and that's the blood-brain barrier. So when you deliver [indiscernible] systemically, they don't get into the brain. We have multiple different receptor targeting strategies to do this. We focused in on TFR 1 as our first pursuit in this space. And I already mentioned bicycle, and that's with TFRs delivery. And the muscle data is looking great, but now we're applying that to the BBB. And one of the benefits of bicycles that they're really small peptide ligands and their small size give them advantages over other modalities because it reduces the total amount of drug required, so which is more convenient for patients and also beneficial for manufacturing. And we and Bicycle has demonstrated that TFR 1 targeting bicycles can cross the BBB. There's a lot of work still to do, it's really early, but it's encouraging. And we've also expanded our ligand toolbox with the recent relationship with Vect-Horus. So Vect-Horus has their VECTrans system. And both Ionis and Vect-Horus has generated compelling preclinical data with that modality for crossing the BBB, and we'll be sharing that shortly at an upcoming meeting. So it's a really exciting new area in my opinion, and delivery of our RNA therapeutics for central indications via systemic deliveries is really that next delivery frontier.

Alright. Excellent. Well, we covered a lot. So -- next up, looking forward to Angelman's data. That sounds pretty cool.

Yes. That's [ true ].

Okay. All right. Well, thank you both. I appreciate it. And thanks for putting up the Stifel [ mug way ]. That's awesome.

Yes. I thought it would be nice.

Thank you. Yeah it's a great touch. Alright. Have a good rest of your day.

You too.

You too.