Cytokinetics, Incorporated (CYTK) Earnings Call Transcript & Summary

All right. Good day, everybody. My name is Ash Verma. Welcome to UBS Biopharma Conference in Miami. Our next company on the stage is Cytokinetics. And with me, we have Andrew Callos, who will be Chief Commercial Officer; and Fady Malik, who is the EVP of R&D. How are you guys doing today?

Doing well.

Great. Thanks for joining us. So since Andrew, you're sitting next to me, so maybe I can start off with some commercial questions first. And I guess maybe just to preface that, like it will be helpful to get a little bit of a background about the story. So if you can just give us a high-level outline for people who might not be familiar about Cytokinetics, just a 2-minute pitch on like what is the platform and what are you trying to do here?

So our later-stage programs are really focused on muscle biology and specifically in cardiovascular. Obviously, our late-stage program is aficamten, focused on hypertrophic cardiomyopathy. So that's our key platform. I don't know if Fady, if you want to expand on that, given your long history of -- with Cytokinetics.

Well, I mean you said it well, I think for 20-plus years, we've pioneered research in the sarcomere, which is the fundamental unit of muscle contractility. And we've developed activators of the sarcomere, inhibitors of the sarcomere. And as Andrew mentioned, right now, I think our most important program is aficamten, which is a cardiac myosin inhibitor, and its focus is to treat hypertrophic cardiomyopathy.

Great. Thanks for that introduction. Maybe just talking about the hypertrophic cardiomyopathy or HCM market in general. I mean, we've seen one drug launch in the space, and the launch hasn't been that exciting, based on like pretty much everybody how most of the investors look at it. So like, Camzyos, I guess the question that I get from investors a lot is like, is this like a potential shortcoming of the drug? Or do you think that this market in and of itself might not be realizable? So just wanted to understand from your perspective, like HCM, like is it something that more cardiologists are just waiting to get a better drug in their hands? Or is there like any kind of like a structural issue in the market that might be limiting the uptake of the competitor that you had before you?

Sure. So the HCM market overall there is about -- so there's 300,000 patients that are diagnosed with HCM in the U.S., 2/3 of those are obstructive. That would be the first indication we're seeking with our clinical trial SEQUOIA. So the obstructive HCM market of 200,000 patients, the ones that are, say, New York Heart Class II/III, meaning they're symptomatic/ And therefore, there are -- most of them are treated over 90% on beta blockers, calcium channel blockers, yet still symptomatic. So there's a high unmet need. Those symptoms look like not being able to do normal functional things, sometimes walking up a flight of stairs. There's also a high correlation of later disease, heart failure and AFib. So there's -- it speaks to a longer-term morbidity as well. So there's unmet need, none of the drugs that are treated before mavacamten came to market were specifically focused on the underlying disease. So we have CMIs that focus on the underlying disease. We're hearing anecdotally that the patients are feeling better when they take CMIs. And therefore, when you have 130,000 patients, again, or the New York Heart Class II/III of that 200,000, there is a market with a high unmet need, there certainly will be there. Sometimes markets take longer to form. It could be an element of a new mechanism, new drug, how -- physicians learning how to use it if you have a REMS program, there's workflow associated with that and safety elements associated with that. The FDA basically puts the REMS programs in place because they say the label itself is not enough to ensure safe use or what other elements of safety. So I think it's a matter of time. Sometimes it just may take a little longer to form. But I have every bit of confidence that the market size is large that there's a number of patients who need this therapy and that the number of physicians that will move outside the centers of excellence over time and into more general cardiology because right now, you see the first end market, obviously, BMS' Camzyos product is pretty linear to launch. And we haven't seen a trend change in that linear revenue and patient projection. But once you start to get into the general cardiology, that's when it will change. It could be a second agent helping do that as well, expanding the market and accelerating, but it will happen.

Got it. And this thing, when you say expanding the market and trying to go to like the community practice, like is there where the diagnosis expansion element of the market is like you think that over time, like you can potentially increase the diagnosis of this disease?

Sure. There -- as projected, there's been studies that based on general population that shows that the true prevalence is probably over 1 million patients. So the true prevalence is probably 3 to 4x what the number is now. The diagnosis rate is increasing with population about 1% in oHCM but there's good reason to believe with genetic testing now as being part of the guidelines with 2 agents if we get approved, if aficamten would get approved, because now you're at medical congresses, there's an underlying drug that can change the underlying course of disease. These things typically will accelerate diagnosis. So we're expecting the diagnosis rate to increase as well beyond that 130.

Yes. Got it. Okay. So I wanted to just talk about the upcoming Phase III readout, which we know is happening in late December, and there's a lot of debate on -- around the setup of the study, how -- what are the different endpoints that you're using, even the time line of the study, results disclosure is the topic in itself. So maybe to start off. So just in terms of like how you have the study design when you're looking at exercise capacity, right, peak VO2 as the primary endpoint. To my understanding, this is something that we did not study in a previous study -- in a previous trial rather, and this is the first time that we are looking at it. Like what gives you confidence that aficamten can show benefit on this endpoint?

Well, first of all, as you know, aficamten is not the first cardiac myosin inhibitor that's been developed. And so we have evidence from mavacamten that cardiac myosin inhibition can improve our primary endpoint, which is peak VO2. The -- we have evidence from outside of our trials where the goal of therapy in obstructive HCM has been to remove the obstruction either through surgery or interventional techniques as a means to improve exercise capacity. And so -- and we've shown with aficamten that the drug results in a very substantial drop in the pressure gradient a relief of the obstruction. And so I think we have pretty strong conviction that the main bad actor, which is the obstruction is relieved by aficamten, and that should improve exercise capacity and we have precedent for it as well. So all of those things, I think, give us fairly good confidence. We've also seen data with regards to other metrics of patient symptoms such as the KCCQ, Kansas City Cardiomyopathy Questionnaire or NYHA functional class that all show benefits in the right direction, both in our open-label extension studies as well as in the placebo-controlled Phase II study, biomarkers that head in the right direction. And so I think all of those things give us confidence in the ability in the aficamten to improve the primary endpoint.

Yes. Yes. I mean for this study, right, like when you think about guided to a specific endpoint or the clinical profile, I know like watching on the sell side and buy side seems to be hyper-fixated on like whether the efficacy measure is going to be something similar or better than Camzyos. And so the physicians that we have talked to and realizing that there is a fair bit of variability in PVO2 and it's a very fine measure. So is there something that you think that in the study design, there are ways to structurally like limit any kind of variability when you are taking these measures so that you are getting to the clinical profile that you want to be at?

Well, we do very careful training of the sites at our investigator meetings. We actually demo a CPET and show them how we do it, how they should do it. They have to qualify. Each site has to qualify and send in a qualifying study. Studies are all read centrally. We monitor the variability over time. And so as we said on the last earnings call, the variability is well within our expectations. So I think all of those things go to optimize the fidelity of the test and variability is inherent to the disease. It's not just the test, but the disease itself is variable from day today. So there is some variability that just the patients themselves. And -- but again, having the benefit of another study conducted in obstructive HCM, it gives you a sense of how to power that endpoint and what the intrinsic variability is in the population.

And so you recently highlighted at your Investor Day, how like using specific baseline characteristics for a patient population increases the chances of teasing out the treatment effect? If you could elaborate on that a little bit, that would be helpful.

Well, I think the drug itself is what -- and the way it's dosed, the way the dose can be optimized is all important to maximize [indiscernible]. The patient population that we enrolled into SEQUOIA represents, I think, a real-world population, people that have medical therapies that are deployed in the real world, beta blockers, calcium channel blockers, disopyramide or even combinations of those drugs. We work to ensure that beta blockers were not overrepresented because medical therapy is not overwhelmingly beta blockers in the real world. And we also know beta blockers tend to blunt the increase in heart rate, which comes with exercise, which can also have a blunting effect on other mechanisms that might improve exercise. So we think that's important. Peak VO2, we targeted patients that had objectively reduced exercise capacity. So their peak VO2 at baseline needed to be 80% or less predicted for their age and sex. And there again, we see a peak VO2 that indicates a significant exercise capacity deficit, which gives us headroom for improvement. It's the main reason for their exercise capacity, is the fact that they have this obstruction is gradient and it's all cardiac related, then we have a more headroom by which to improve.

In terms of like the expectations for the study on the efficacy front, I mean, do you think that you have to surpass the efficacy of your competitor in any way? Like is that even or do you think that this has to be like independently assessed how the things can...

Yes. I don't like to -- I mean we're not going head-to-head studies, and we shouldn't be making comparative statements. But the focus on the degree cardiac myosin inhibition, as we've seen -- as we've seen mavacamten is incredibly effective at improving patient symptoms and function. And it's not really, I think, a question of which one is more. They're both going to be, I think, mavacamten has the data, we expect and hope to see substantial improvements with aficamten and sort of like you're taking a patient who at baseline, has trouble walking up one flight of stairs. And after treatment, they might be able to walk up 4 or 5 flights of stairs whether that's 4 flights or 5 flights or 6 flights. I don't know if it really matters, right? I mean the -- they've gotten over to where they can function in a more normal way because they can get over that first flight of stairs. So I think the focus on how much efficacy you can squeeze out is maybe a little misguided. We think we'll have very strong efficacy, but also there are other aspects that are important to patients and to physicians, data, safety, ease of use, convenience, all those sorts of things. And we think aficamten has some benefits in those areas as well that given its reversibility, given what we've seen in the open-label extension in terms of the rarity of low EF events and things like that, that speak to having some intrinsic advantages.

And what we've done very robust research, 450 physicians, half of them have used mava, half of them are from centers of excellence, and we show them our profile across a range. Safety is the most important thing, preservation of ejection fraction, followed by efficacy things like gradient reduction, New York Heart Class change. So we show these physicians a range of where we could wind up on our primary endpoint and their utilization maybe goes down a little bit, but still we would get more patients on aficamten and not if the safety elements wind up being what we're expecting.

Yes. Yes. I mean -- and to that end, I mean, on the safety front, the LVEF less than 40 and 50 data that you have highlighted from the FOREST study, I mean that has been pretty remarkable. Is there a reason to believe that in the SEQUOIA study, this profile could be any different compared to what we saw in the FOREST study?

Well, we know -- as we also disclosed in the earnings call, we know that we haven't had any low ejection fraction events below 40%. We're blinded to those between 50% and 40%, but the patients in FOREST are the patients that came from SEQUOIA, right? So we think the data we have in FOREST should be pretty representative of what we're seeing in SEQUOIA because these patients have undergo, the same dose type they're using they undergo, they're the same patients. They're more of them actually because the placebo patients are put on drug in the open-label extension, where they were on placebo and SEQUOIA. So we think the FOREST open-label extension data are fairly compelling. And I don't think you can ask for a better insight into a blinded study than that to repeat the blinded study in an open-label fashion.

Yes. Yes. Absolutely. And then again, on this piece as well, just like in terms of the variability between the measurement of LVEF less than 50 or 40 or 45, like I mean like how much of variability is there between like different sites and different investigators making a measure or something like that?

Ejection fraction can vary anywhere from 5% to 7% even with the same person in the same site day-to-day, either that intrinsic ejection fraction is not a fixed number. It's not like measuring somebody's height, changes dynamically depending on loading conditions and hydration and things. So there's intrinsic variability of ejection fraction. But generally, that's the range of variability you see. So in measuring it, if someone has on one measurement of ejection fraction of 52, it's quite possible variability will have one measurement below 50. They may also have one that's closer to 60. And so the way our dosing is deployed kind of account for some of that and that we kind of ignore ejection fractions between 50 and 55. We don't act on them. There's no dosing changes that are made. So you only -- you increase the dose of the EF is less -- is greater than 50, and you decrease the dose of the EF is less than 50 -- I'm sorry, the increase the dose if it's greater than 55 and decrease the dose if it's less than 50, you don't do anything for whatever in between.

But it sounds like on every data card like so far on this metric, I mean aficamten seems to be doing well. So my question really on this is that like is this -- does this type of a profile would give the conviction to the physicians that you need to do the echo monitoring on a less frequent basis. And that's meaning like one of the hot button issues with Camzyos, do you think that this type of data and if you're able to replicate that in Phase III, there would be less of onerous monitoring.

Well, we would hope that the data would support a convenient dosing and maintenance schedule of echocardiogram. So I can't really comment on where we'll end up because, ultimately, that ends up being a discussion between us what we propose and what FDA wants to see. But I think from just a resource utilization standpoint, echoes are burdensome, and they cost the system money. And as we showed in FOREST of the 579 echoes that we've gathered in the maintenance phase, the only 3 of them had a treatment impact. And so what's the right -- there's got to be a more efficient way of ensuring patient safety but not necessarily getting lots of echos that contribute no information. We have a few ideas about how to get there, I think.

Got it. Okay. And so in this case, I mean, we are like near term, we are looking for the clinical trial data, but eventually like what the FDA wants to do in terms of the label will be a key determinant of the value right here? I mean, yes, from your perspective, like if you think that this is the type of profile that is emerging for the drug, do you feel confident that you would not require REMS or even if there is view like a much less onerous REMS as you have the harden that [indiscernible]?

Why don't you, Andrew, take that one.

Sure. From a REMS point of view. So I think that as Fady outlined, we will suggest the REMS to the FDA and then it's a negotiation based on what the FDA want to [indiscernible] sort of safe use. Once we see the data, then we'll put a REMS program together. I don't want to use less or more because then we're comparing to something. And I think it's more around what do patients who would be considered for aficamten, what does the FDA and what we feel, given the mechanism needs to be in a REMS program. We do think we're going to suggest a REMS program. It's likely going to be any [indiscernible] REMS and we'll -- maybe we have monitoring a dose adjustment, but with a several week window. Maybe we have once or twice a year once on stable dose or maybe it's risk-adjusted based on where their ejection fraction where a patient element is. So I think we're going to try to put something together that's reasonable for the health care system but it's sort of safe to use as informed by our data. So our data will stand on its own. And if we see data like REDWOOD, then I think we have good confidence that we should have a reasonable negotiation with the FDA that gives us the REMS program that could be preferential for treatment. But there's a variability. And at the end of the day, the FDA can decide that they do or don't agree with us.

Right. So -- and then just in terms of the timing of the study, top line disclosure and what data we might expect? I mean, yes, this is kind of like a topic that has taken a life of its own. I mean, so anything that you can comment to clarify, like when -- [ as much ] as you can, like when are we going to get the top line? And what would it have versus what can we expect at the conference next year?

I mean I think we can be pretty simple. The answer is late December, that's as precise as we'll get. The top line release is still to be determined. We don't expect to be able to give lots of quantitative data in the top line release because we would like to see the study presented at the American College of Cardiology, and they have rather strict rules about that. But we certainly will hopefully get some qualitative directionality in terms of both efficacy and safety. Pointing towards 2024, with the American College of Cardiology one will see, I hope, because it hasn't been submitted nor accepted yet but we would hope to be presented there as a late breaker and have the full data set really laid out for consumption.

Good, good. Okay. So just one quick pipeline question before we wrap up. So for 586 program, I just wanted to understand like what's the value proposition for HFpEF? A bunch of different options available here. I just wanted to see how you think this would differentiate versus existing options?

I'll start, and I'll turn it over to Andrew. But thanks for asking the question. 586 is another cardiac myosin inhibitor. It is different from aficamten in the way it works. It's now in Phase I, and we have positioned it towards a subset of HFpEF that we think is a mimic of the non-obstructive HCM patients. So kind of extending the population into an adjacent group of heart failure patients, they have thickening of their heart, they have heart failure symptoms, they have elevated biomarkers. In a lot of ways, they look like in HCM patients, maybe not quite as severe in terms of hypertrophy. So we think that there are hundreds of thousands of those patients, and they similarly could benefit from a treatment that improves their function and their disease status. And maybe Andrew can comment a little bit on the research we've done there.

Yes. I mean the subset is probably going to be the ejection fraction that's greater than 65. SGLT2s are used there generally and indicated. So by the time, if this gets to market, would be on top of SGLT2 likely. Entresto, from a data point of view, has HFpEF in the label but from a publication point of view, I think the number starts to in terms of where benefit range is when you start to hit the higher 50s like 57, 58. So that's probably not a competitor per se Entresto in that market. But that subset is pretty large. It's 1 million patients, high unmet need. So this is an area where we feel like we could provide benefit on top of what the standard of care would be at that point.

Great. All right. With that, we are just a little bit over time. So thank you so much for this, and good luck, and looking forward to hearing more about the top line.

Thank you, Ash.

Thank you very much.