Arrowhead Pharmaceuticals, Inc. (ARWR) Earnings Call Transcript & Summary

Good afternoon, everyone. I'm Ellie Merle. Thank you so much for joining us at the UBS Global Healthcare Conference in person. Very happy to have Arrowhead Pharmaceuticals here with us for a fireside chat today. Joining us from Arrowhead is Vince Anzalone, VP, Finance and Investor Relations; and Dr. James Hamilton, the Head of Discovery and Translational Research. Thank you very much for joining us.

Maybe to kick it off before we go into some of the specific programs high level, you guys have an exciting RNAi platform with a lot of potential applications. Can you frame for us where do you see the strategy from a prioritization perspective and where you see the company and the platform evolving over the next, say, 3, 5 years?

Sure. And I want to start by saying thanks to Ellie and the whole UBS team for having us today. We appreciate it. So our RNAi platform is called TRiM or targeted RNAi molecules. And we think that we're able to get best-in-class knockdown of target genes in the liver with substantial duration. And so far, knock on my shoe, very favorable safety and tolerability. Where we think we're differentiating is that we are leading the translation of RNAi outside the liver. And so we will -- this is a good timing now. Tomorrow, we're holding a Pulmonary R&D Day here in New York to go over some advances that we're making to the pulmonary platform, talk about 2 of our new candidates that will be going into clinical studies shortly called ARO-MUC5AC and ARO-RAGE and then also talk about some of the learnings we've had from the initial clinical studies and preclinical studies of our first-generation compound, ARO-ENaC, against cystic fibrosis. And then also where we see the platform going broadly. We'll talk about some new programs that we haven't talked about before. And then also with inflammatory pulmonary disease and muco-obstructive pulmonary disease, where we kind of see ourselves slotting in. And so that, again, I think is a big differentiator and certainly can be an engine for growth for us in the next 3 to 5 years. Beyond lung, we also will be entering the clinic in the future with a program against muscle, a muscle target. And beyond that, our goal is to get to a new cell type every 18 to 24 months. We will be talking about what our next 1 and maybe 2 tissue types are at some point shortly, call it, in the next year or so. For how we see the company strategically growing is, if you look back when the TRiM platform first entered clinical studies in 2018, I believe it was, we were very aggressive at expanding the pipeline. And we've gone to 6, 7, 8, 9, I can't recall the number, of liver targets with the TRiM platform that are in clinical studies, either wholly owned or partnered, but we were very aggressive there. And our strategy for target selection has always been the same regardless of tissue type. We're not looking to find therapeutic targets that RNAi can access. We're looking for targets that RNAi confers some type of advantage over other methods or for targets that just have been historically not druggable with small molecule or monoclonal antibodies. That is true also for every cell type that we can get to now and that we plan to get to in the future. Also, a strategic goal of ours is to always maintain a good healthy mix of early stage, mid-stage, later stage, and then ultimately, commercial assets and also a good healthy mix of wholly owned versus partnered programs. We use partnering selectively to do a couple of things. One, to bring in capital that complements our need for external capital. And I think we've done a very good job at that. We haven't had to raise money in the equity market since December of 2019, which I think is saying something. And then also, we bring in partners who can add expertise or resources that we either don't want to build out ourselves or that it doesn't make sense for us to go alone. And so we have existing partnerships with Janssen, with Takeda, with Amgen, with Horizon and with GSK. And over the coming year to 18 to 24 months, those should start to generate significant milestone revenue. Again, that complements our need for outside capital and allows us to expand into areas where we don't plan on building out commercially. And then it leads me to my last point on strategic. We do intend to become traditional vertically integrated pharmaceutical company, which means that we have to build out commercial. And being a platform company, we have a lot more opportunities to discover and do the early development on programs, then we will have the bandwidth to commercialize ourselves. So right now, our thinking is that we would like to build out commercial around cardiometabolic and pulmonary. And I've discussed pulmonary already. But our 2 lead wholly owned assets, ARO-APOC3 and ARO-ANG3, both are our cardiometabolic drugs in Phase -- ARO-ANG3 is in Phase II. ARO-APOC3 is in a Phase III. And so over the coming 3 to 5 years, that time frame you're asking about, we do hope to build out commercial infrastructure and then ultimately sell and market our own drugs.

That's exciting. Maybe first diving into the AAT program, a topic of a lot of conversation. First, how should we think about the regulatory path forward, both in terms of the design of a potential Phase III as well as the potential for an accelerated approval pathway based on the Phase II data?

Sure. So I'll take the former first. The design of a Phase III is what Takeda and we will be discussing with FDA shortly. So we -- the current Phase II study called SEQUOIA has 2 different readouts. First is circulating levels of AAT, and this is 42 patients. We're looking at 3 different dose levels. So circulating levels of AAT, which was a gating item to select a dose for a Phase III. And then the second readout for that study is 12-month biopsy, which we'll be taking the final biopsy shortly in June or July, I believe. And so we hope to have data to report out later this year. So for the regulatory discussions, I don't want to get into too much detail about what Takeda and we are thinking because we have to have those discussions with regulators. But I think that from a commercial standpoint, I think it's important to have fibrosis on -- fibrosis reduction on a label in order to get uptake and reimbursement in certain geographies of the world. And then I guess that leads to your second question about accelerated approval. The discussions about Phase III design and potential for accelerated approval are really 2 different discussions. So that won't be happening at the end of Phase II meeting. Once we have biopsy data from the SEQUOIA study, we'll certainly take a look at those and see if those data warrant a reengagement with regulators to determine if a pathway exists and what that pathway would look like. So I would say more to come on that later this year.

Understood. Maybe to dive into the disease a bit more, maybe to put the conversation around unmet need or potential for accelerated approval in context, can you tell us a bit about AAT and, in particular, the prevalence and unmet need when it comes to the liver manifestations?

Sure. James, do you want to take that?

Yes, sure. So broadly speaking, alpha-1 antitrypsin deficiency, specifically the ZZ variant, it's about 100,000 patients in the U.S. and maybe another 100,000 patients in Europe. In terms of how many of those patients develop liver disease, it depends on which resource you're looking at. There were some recent publications indicate it's probably about 1/3 of those patients will, over time, develop clinically significant. And by that, I mean either F2 or greater liver fibrosis based on a liver biopsy. So if you're looking at it, that's maybe 30,000 or so in each U.S. and then another in -- 30,000 in Europe approximately.

And maybe just in terms of available treatments today versus some of the initial data that we've seen, smaller numbers, but what we've seen so far in terms of reductions in circulating AAT as well as the early biopsy data.

In terms of what?

What we've seen so far from your data.

Sure, yes. So we're seeing, really ever since Phase I, great reductions, greater than 80%, 90% reduction in serum AAT both in healthy volunteers but then in patients also. And then also liver biopsy, we're seeing 80-plus percent reduction in Z protein, so the liver Z protein in the alpha-1 antitrypsin deficient patients. And then that has correlated with not only a reduction in transaminases, so blood markers as liver inflammation, ALT, AST; also reductions in blood GGT and then improvement in some patients in fibrosis, and we'll hopefully be getting some more of those data, as Vince alluded to, in the 2001 study biopsies.

And that was actually a really exciting result for us. So I would say, in our open-label study, which we call the 2002 study. So we have a few different cohorts there that we're looking at baseline biopsy and then at 6 months and then that cohort was eligible to elect to continue to get treatment for another 12 months and then we asked for a third biopsy. We don't require it. And then another cohort was looking at baseline biopsy in 12 months, and then they can get another 12 months of treatment, and then we ask for another biopsy at 24. Again, they're not required to get that. So our initial belief is that at 6 months, we were hoping that we would see a reduction in quantitative monomer of the Z protein in the liver and monomers, essentially the newer copies that haven't polymerized. And so -- and we did see that. We saw a dramatic reduction. What was more of a surprise is that in 6 months, which we thought was may not be enough time, we were seeing a reduction in the polymer of that Z AAT protein. And so we thought that these were very hard for the liver to break down and metabolize and clear, but we were -- then we were also at 6 months, and then again at 12 months, seen a dramatic reduction in polymer. In addition, this is a semi-quantitative measure, but we were seeing a marked reduction in, I think, across all patients in globules, correct me if I'm wrong on that, but many patients completely resolved globules. And all patients had a reduction in their semi-quantitative globule score. And then the last thing that James talked about is that there were a number of patients, and I want to say across the different cohorts, it was somewhere around 40-some-odd percent of patients were getting a 1 point or greater improvement in fibrosis, either at 6 months or 12 months. And again, that's a really dramatic change in a short amount of time. So that gave us a lot of confidence for SEQUOIA and then for ultimately a Phase III study. Now I mentioned earlier, we hope to have a data readout from the 12-month biopsy on SEQUOIA later this year. We're really interested to see what placebo-controlled data look like, right? Because the open-label study, we don't know what the background is. And so that's going to be an interesting thing for us to see this year.

Absolutely. And how should we think about -- I mean, what's your expectation for what we would see on the background in terms of fibrosis improvement on the biopsy study?

So background for placebo, and James knows a lot more about this than I do, so I'll let him talk about it. My assumption is that it's not a fast enough moving or progressing disease to see placebo progress in fibrosis.

Right over that. Depending on the time frame, even a couple of years that unlikely to see definitive progression of fibrosis in a placebo.

And I guess, what are you looking to see in terms of what would be compelling data at the 12-month biopsy data from SEQUOIA, which I guess we'll be getting in a couple of months?

Not too much time. That's right. Yes. I think anything that is largely consistent with what we saw in the open-label study would be exciting. Now there's obviously variability in reading these slides, and there's variability in taking biopsy, so it would be very nice to see in a controlled study, something that was consistent with what we saw in the open-label study.

Absolutely. And forgive me if I'm getting ahead of myself, but the way you're talking about the disease, it seems like decreasing the monomers in the liver is meaningful, decreasing the polymers in the liver is meaningful, decreasing the number of globules also meaningful. And now we're talking about fibrosis improvements over a 12-month period. How are you thinking about what the appropriate end point that would be meaningful for patients should be? I mean I understand that regulatory discussions are ongoing, but outside of the outcome of that, how are you thinking about what should be meaningful for patients from a clinical design perspective?

Sure. So I think that's a tough question to ask. I would say the pathophysiology of this disease is perfectly clear, crystal clear. So progressive liver disease is caused by all of those things that you mentioned. You get monomer that's created that can't get exported out of the liver. It polymerizes. Then it causes these globules, and it causes the cycles of inflammation and repair, and then ultimately, fibrosis and in some patients, cirrhosis or HCC. So that's kind of the progression of it. Now since there have been no drugs before that could reduce the production of that mutant protein, you can't really draw a direct correlation between quantitative measures of protein in the liver and fibrosis. So we hope, ultimately that, that correlation will be able to be made. But I think at this point, you want to look at fibrosis as the meaningful change.

And how should we think about the potential range of outcomes for a Phase III design? I mean, what would be the shortest and smallest realistically looking like? Or what would be the longest and most outcomes-oriented potential Phase III design?

Yes. I think now we're kind of getting into the risky territory. So I prefer to have those discussions after the regulatory discussions.

I understand. And then maybe just in terms of thinking about the cadence of when we will learn from our side, as you guys are engaging with these. How should we think about the time frame for when we can learn more, both on the Phase III design and then after SEQUOIA, potentially if we see some exciting fibrosis data, potentially if there's no need for a Phase III for accelerated?

Yes, we'll have to have those talks with Takeda, our partner. We don't have a plan at this point. We'll play it by year.

Okay. I'm sorry. I've asked you enough about that. I'll move on. Maybe turning to the pulmonary programs. And I know that tomorrow, you have your Pulmonary R&D day, so I'll keep it high level. What are some of the learnings from the initial ENaC program and what are some of the modifications and enhancements to the platform that gives you confidence as you move into the clinic with 2 novel targets?

Sure. And actually, the first question you asked is, there's a whole section tomorrow that James will be talking about of all the learnings we had from the ENaC study. And our study was really the preclinical work that led us to some of these findings that we think give us great confidence in the new programs. And again, I don't want to front-run this, because it's just 24 hours or so from now, so stay tuned on that. But we think that we understand better the mechanism that was causing the preclinical chronic toxicology, and we think we've made the proper adjustments for the new programs to get us over that threshold. Now we'll talk specifically about what that means tomorrow, but we feel very confident about it. And look, biotech is a risky endeavor. And I think with RNAi, especially with liver-directed programs, we have, as an industry, I'm just an RNAi company, and likely you guys as investors, we came to expect over the last handful of years that everything was just going to work, because since we and the other RNAi folks have kind of cracked this liver delivery. The translation from preclinical results to clinical results to early clinical and then early to mid, to late clinical have just been so reliable with respect to safety and tolerability and activity and duration of response, all of that has just -- it's worked wonderfully. The pulmonary platform is a new iteration of the same idea, but it's structurally distinct. These are different structures. And so with the new technology, you tend to get better over time. And I think that's 1 of the key areas of leverage with RNAi that's really attractive to us. That once -- everything you learn from 1 program directly informs the next program. And then the second informs the third, informs the fourth, informs the fifth. And so we feel like we are progressively understanding it better and progressively lowering the risk of unexpected results as we get further along. And one really important part about both of our new targets, but RAGE in particular, is that there is a biomarker for target engagement. For RAGE, we can measure in serum. So we're going to do serial blood draws even in healthy volunteers when we start the clinical study. So we're going to learn a lot about the pharmacodynamic activity of the drug. So we'll know what doses get to what reduction of the protein. We'll know how long it lasts. So that will help with structuring the later-stage clinical development. But maybe equally or more important, it will help us determine the right design for future chronic toxicology studies for the whole platform. So we're getting closer to where we are with the liver. Hopefully, over the coming year or 2 or however long it takes, we will be there.

Absolutely. I mean it's certainly a huge milestone for you guys as well as the field as a whole, if we can expand RNAi on the liver. And one last question on this, and I'll keep it high level on lung as well before we move on to some other programs. But can you talk a bit about some of the unique platform enhancements that you've done at Arrowhead in particular? There's obviously other companies in the RNAi space, but can you speak to what makes some of the platform R&D work that you've been doing over the past couple of years, perhaps unique and proprietary to Arrowhead?

Sure. I'll let James...

Sure, yes. So maybe starting with the targeting ligand looking at different targeting ligands beyond the small molecules to target the liver, the so-called GalNAc like technology. So looking at other ligand receptor pairs in the lung, for example, and beyond the lung, in the muscle and in other tissue types that we're looking at. And then there is a constant effort to enhance and modify to improve the backbone chemistry and chemistries that enhance depth of knockdown, engagement with the risk complex to facilitate better duration of activity, which translates ideally into lower doses and less frequent dose administration. And some of that we'll be talking about tomorrow.

Looking forward to it. Maybe turning to the C3 program. Obviously, a lot of potential indications and applications for the complement pathway. Maybe first, can you tell us a bit about the design, where you're in dose escalation now and how we should think about the time frame for data as well as longer-term thinking about when you move into patients and ultimately think about the opportunity set of different indications?

Sure. Yes. So the study is ongoing now. It starts as a single dose-escalation study in healthy volunteers. We can measure C3 in the serum readily in the healthy, so we can get a good idea of pharmacodynamic dose response. And then once the dose escalation is complete, then the study transforms into more of a basket study, where we enrolled cohorts of different patient populations. And so we'll look at PNH, we will look at some complement-mediated renal disorders. So C3 glomerulopathy, IgA nephropathy and then we can add other baskets of patients as we see fit. The doses for those cohorts will be selected based on the healthy volunteer dose-escalation phase. And then in terms of timing for data, I'll throw that one back to you, Vince.

Sure. So I think for the healthy volunteer portion we are either at or close to the top dose -- the top design dose. So once we finish that, then we trigger the -- we select a dose and then move on to the basket study. So it's likely that we'll have at least healthy volunteer data before the end of this year. The patient data will likely be in '23 at some point.

Exciting. And how should we think about the optimal level of C3 lowering or knockdown?

Yes. So I think you probably need a fairly significant amount of knockdown for C3. And I'm thinking of the peptide inhibitor that is FDA approved, that they describe achieving 80% to 90% inhibition of C3. So that's probably what you are aiming for with an RNAi effect, a reduction in C3 in that range that should achieve. We won't know until we go into patients, but that level of knockdown, you should see some meaningful change in hemoglobin or whatever your other biomarker of interest is.

And from a strategic perspective, both keeping in mind sort of other C3s in development, such as, say, the Apellis program, how are you thinking about potential advantages of using RNAi as well as thinking about strategically, which indications make the most sense to develop?

Yes. I think, again, if you're looking at the peptide inhibitor that's approved, the Phase III data were very compelling. The improvements in hemoglobin that they were seeing in the other biomarker end points, when the patients came off of eculizumab, there was still this nice increase in hemoglobin on the C3 inhibitor. The issue with that approach is that it requires a lot of drug that I believe it's administered as 1 to 2 grams a few times -- 2 to 3 times a week through a subcu infusion. And if we can achieve that same amount of knockdown and get the same amount of activity with a 1-ml injection every 3 to 6 months, that seems compelling to me.

Absolutely exciting programs. Maybe turning to the cardiovascular franchise, especially since you mentioned at the beginning that in 3 to 5 years, you do see yourself as a commercial organization. You have a couple of programs ongoing, both partnered and unpartnered. Maybe can you frame for us across Lp(a) with Amgen as well as APOC3 and Angiopoietin-like 3, where you see the role for each of these programs in terms of the subsets of patients as well as frame for us kind of the milestones from a data perspective we can expect over the next 1 to 2 years?

Sure. So when we were thinking about cardiometabolic broadly and where RNAi could play, we wanted to fall back on genetics. So there's a lot of really good population genetics work going on, and there has been over the last handful of years. And these targets have emerged, APOC3, ANGPTL3 and Lp(a). And so what those do is they modulate lipid parameters for patients in different ways. Where we view cardiometabolic in the future, so 10 years from now. Well, let's look back 20 years. So the only thing or 1 of the only things cardiologists really had to reduce overall risk of cardiovascular disease was LDL lowering, like statins got better and better, and then PCSK9 inhibitors came along. And so they're pushing down this 1 lever. And it reduced cardiovascular risk by 20% to 25% or whatever the number was. But there's still this big risk out there. We view that kind of holistically. So we think that some major contributors to that residual risk cardiovascular disease are triglycerides, residual LDL for patients that still aren't getting to goal, Lp(a) and then to a certain extent, and we'll talk about this in the future, is either low HDL or poorly functioning HDL. And so your question about where do all of these assets fit and how do you segment it. So the good news is we don't have to make that bet today, right? We have to create drugs that can lower triglycerides, lower LDL, lower Lp(a) and they have effects on HDL. And as we do larger studies that look at broader patient populations and look at longer treatment, I think those trends will start to emerge. And our goal 10 years from now or 5 years, whatever the number is, is to give cardiologists and other physicians all of these different options, so that they can personalize treatment to patients individual lipid parameters. Some patients might -- their cardiovascular risk might be mostly triglyceride-driven. And again, these are just hypotheticals here. If that's the case, then we know that our APOC3 program is better than ANGPTL3 at reducing triglycerides. It might be somebody who is more balanced. They need a little more LDL than APOC3 can do. That might be an ANGPTL3 program. Lp(a) is kind of orthogonal to all that. So Lp(a) is something that -- the Lp(a) levels are correlated with increased risk of cardiovascular disease. It's clear. Population data are clear on that. It's also something that's not responsive to changes in diet or lifestyle. It's based on your genes. So if you're born somebody who is a high Lp(a) producer, you will be that for life, and there has not been a good way to reduce it. And now that the Amgen program called olpasiran, which we licensed to them a number of years ago. Now that they can modulate that, it's going to be really interesting to see in larger, longer, larger studies, what happens to these patients, cardiovascular risk. And so I foresee a future where cardiology and cardiovascular disease risk is really treated on a personalized basis because we will just have so many more tools in the future.

Absolutely. And some of these studies, I know, will take time, particularly in running the science around long-term outcomes. But I guess, over the next 1 to 2 years, what are the answers that we can expect to get on some of these programs?

Sure. I'll start with ARO-ANG3. So we have a Phase II study called ARCHES-2. It's looking at patients with mixed dyslipidemia, so elevated trigs and persistently elevated LDL cholesterol. That study was fully enrolled earlier this year. It should complete by the end of this year, and then we'll have a readout in the first half of next year. For APOC3 -- I'm sorry, let me step back. So we also have a study called GATEWAY for ANG, which we just started recently, which is looking at a small number of patients. I think it's 16 patients with HoFH. So we'll take a look and see what those data look like to see if there's -- if it warrants a kind of a fast path to approval strategy in that rare disease population. For APOC3, we have a study called SHASTA-2, which is in severe hypertriglyceridemic patients. So patients whose trigs are above 500. Another study called MUIR, which is also looking at mixed dyslipidemia and then a Phase III study in a rare disease population called FCS, or familial chylomicronemia syndrome. And so the -- I mentioned this briefly with ANG, but with APO, it's really a compelling case that we can have a gradual buildup in our commercial organization. If we can start with these very narrow rare disease populations, and then progressively and in parallel, as we study larger populations and have longer-term treatment, we can potentially get approval on rare and then expand the label into larger populations. So it allows us to get some experience with the drug, get it on the market, get it to patients who are most in need of therapy and then also build out our commercial organization in a stepwise fashion. It would be difficult to imagine launching 30 million patient prevalence disease drug as a small company as we are. But slow walking that up over a number of years, I think is definitely feasible for us.

Absolutely. That's helpful and certainly makes sense on the commercial strategy. Turning to a comment that you made at the beginning, which is that you're hoping to every 18 months or so, if I got it correct, move to a different tissue type or cell type. Turning to muscle cells with the FSHD program. Maybe if you could tell us a bit about the science and how you're able to get RNAi delivery into these cell and tissue types?

Sure. So it's a similar approach to what we use with our other programs. I mean we use a ligand that is not muscle specific. It's actually a peptide ligand, but it targets a receptor on the surface of the myocyte, and that allows us to get the RNAi into the endosome, where it can eventually escape from the endosome and engage with the risk complex and silence the gene. In the case of our first program, we target the gene DUX4 that is expressed -- is actually normally repressed in normal cells and in a condition called FSHD or facioscapulohumeral muscular dystrophy that miss expression occurs. So the repression of the expression that's normal is not occurring. And DUX4 is aberrantly expressed. It's a transcription factor, and so it sets off a cascade of other inflammatory events that injure the muscle and cause a muscular dystrophy. So it's a similar approach. It's a ligand-facilitated approach. The muscle is a large organ relative to the liver or even the lung. So you have to adjust the doses accordingly. But otherwise, the approach is the same. We don't use any LNPs or nanoparticles or liposomes or anything like that. It's still the ligand oligo conjugate.

Interesting.

And all of our trim constructs, regardless of where the tissue is, they're all based on the same idea. James mentioned it. These are siRNA conjugates. So we don't want to have these kludgy large molecules that have to encapsulate RNA. If you can conjugate it directly to targeting ligand and use different linker chemistries that are that optimize uptake, we think simpler, the better. And so far, all of our constructs use that same strategy. We do, in certain cases, add some structures to enhance pharmacokinetics. Because with liver, you don't need it, because so much goes directly -- is taken up directly in the liver. But outside the liver, sometimes you need to increase circulation time or the like to improve pharmacokinetic.

Makes sense. And in terms of FSHD, I know there's been some updates in the space and thinking around trial design. Can you remind us sort of where you are in the preclinical work and how we should think about sort of the update provided at the quarter and when we'll start seeing some patients dose and perhaps longer term, some data?

Sure. So there are not -- there really are no known good biomarkers in FSHD that can be readily measured in an early-stage study. I think some of the data coming out of the small molecule company, Fulcrum, indicate that you may need to rely on clinical end points. And to get to the point where you're seeing improvement in clinical end points, you likely need a longer study like a 1-year Phase II. So our thought was that if you have to get to that point to see any improvement or anything that's really measurable that you probably need chronic tox. You do need chronic tox to get there. And so we're conducting the chronic toxicology studies. And I think we'll make a call and provide updates when we have more insight into what those data show.

Unlike lung with RAGE where we'll get a potential circulating biomarker.

Right. There's unfortunately, nothing like that with DUX4.

Yes, it makes a big difference when it's the first time that a platform goes into man because we don't really have any signposts on dose or duration or anything. And so we think that a slow path to first-in-man studies potentially gives us a faster path to functional readout if we do the chronic tox first.

That makes sense. And with 1 minute left, I want to ask about the potential for RNAi in the oncology setting, particularly in solid tumors. I know you provided an update on the quarter around the HIF2a program in solid tumors. Maybe if you could tell us some of the learnings there and the amount of target engagement you were able to see in solid tumors?

Sure, yes. So I think we saw clear signs of HIF2 knockdown both mRNA knockdown and reductions in protein. And we shared that in abstract form and max knockdown. I think we were hitting 80% to 90% plus knockdown of the protein. And then we had 2 patients that actually had a partial response. So 1 patient had a 68% reduction in tumor volume. And then unfortunately, had to come off of drug due to NSAE. So we feel like we saw proof of target engagement, but then also a proof of concept and a proof of clinical concept in some patients that had knocked down of the target and then also a partial response.

Makes sense. And I see we're at time, so I don't want to ask any more questions about that. But thank you so much for chatting with us, a great learning more about the platform as well as AAT, C3 and various other programs. So thanks so much.

Thank you.

Thank you.