Okay, we're going to get started. I'm Andrew Tsai, Senior Biotech Analyst at Jefferies. Thank you for joining me today. It's my pleasure to have Travis Murdoch, CEO of Braveheart, with me. Welcome, Travis.
Thanks for having me.
Maybe to begin, can you give us an overview about Braveheart, what you're working on, what drug you have, and what milestones we can expect over the next few months?
Sure. Thank you, Andrew. We are a late-stage, clinically focused company developing what we believe could be a best-in-disease compound for hypertrophic cardiomyopathy, both forms, both obstructive and non-obstructive HCM. We are currently enrolling patients in a global phase III study in obstructive HCM. That study's called LIONHEART. It's designed as a head-to-head study versus metoprolol. We also have an ongoing fully enrolled study being run by our partner Hengrui in China that's on top of standard of care, where we'll have data in the first half of next year in obstructive HCM. Also in the midst of launching a phase III program, NOBLEHEART, in non-obstructive HCM, where we've generated phase II data. We recently had a late breaker at AHA, where we announced one-year follow-up data in obstructive HCM.
Thought the data were tremendously encouraging and also really demonstrated both the need for very little in the way of titration, which has been one of the challenges with this class, as well as some encouraging data both on peak VO2 and disease remodeling. On the back of that, we launched an additional study, which is called TRUEHEART, which is a fixed dose, no echo-based titration study in obstructive HCM that we think will be an important part of the story as well as we continue to build out the clinical development plan.
Okay. You hinted at your differentiation, but maybe let's expand on it more. There's the ease of use part, the convenience, there's efficacy part, there's the safety tolerability part. Maybe can you kind of give us the whole underlying value proposition?
Sure. Thanks for that. We believe that BHB-1893 has unique pharmacology, and we also had a presentation at AHA and an investor day, it'd be a good thing to listen to, that really spoke to that. What that's translated to in the clinic and across now a fairly large phase II data set across two different studies is efficacy that looks potentially best in disease, both on key markers that physicians monitor, which gradient is sort of the thing in obstructive disease that physicians monitor on a day-to-day basis and use to help understand how their patients are doing, but also now on feel and function, KCCQ and NYHA improvement and peak VO2. Efficacy matters, and we think that that's an important part of the story.
In the non-obstructive form of the disease, we've also seen very encouraging efficacy data in a placebo-controlled phase II study where both the echo parameters of diastolic function improved to normal on a number of metrics, but then we saw large changes in key metrics like KCCQ. On safety, we continue to have a very encouraging profile. This is a molecule that was designed with very short half-life, so the reversibility becomes an important feature, but we also know now from the clinical data that the effect on ejection fraction is very shallow. For instance, in obstructive HCM with now 50 patient years of data, we've only had two EF measurements by sites between 45% and 50%. Then importantly, getting to efficacy quickly and with minimal fuss is a key part of the program.
In our initial phase III study, we're going to be allowing a one-step titration at week four. We know from the phase II data that a majority of patients require no titration. We're planning the same paradigm for the non-obstructive study. Given this new data, we felt compelled to also study a fixed dose because we know that patients do consolidate their responses over time. We think this would give a future prescribing paradigm where community docs who may not be wanting to do complicated echoes to get patients onto therapy could start patients at 40, and if patients need to be referred, then we have the opportunity to increase efficacy with the higher dose.
Makes sense. Then as you think about the approved CMIs in the marketplace, there's mavacamten for obstructive, aficamten from Cytokinetics for obstructive and non-obstructive. Or at least for the in obstructive, what do you think is hurting the sales trajectory, or preventing the sales from growing even more? We just discussed the three kind of attributes, major attributes. How would you rank order those?
Sure. I would say in current state, there's not really been a discussion about efficacy, and as I think many people in this room know, that that's often a point of discussion that drives physician brand share dramatically. I would say that one of the major challenges today, and more so with mavacamten than aficamten, is the complex REMS and the rigidity of the REMS. When you think about that REMS, though, it largely pertains to the multi-titration steps, and in mavacamten's case, the need to do multiple echoes before initiating titration. So that, in our market research, if you think about where patients lie, most patients are in the community, but the community economics and the practice patterns are such that many community cardiologists are not comfortable prescribing a product that requires a Valsalva echo maneuver, a complex echo maneuver that they don't do a lot.
That's where we think there will be a big unlock for an agent that's easier to prescribe, but still has the important efficacy that's been seen now with the CMI class and perhaps even beyond.
Makes a lot of sense. Mechanistically, how did you're a CMI too, how did you engineer your molecule to have all these potentially superior properties exactly?
Right. So, there was a fair bit of work done to engineer this molecule to have predictable PK properties. That's important because some of the other molecules have particular CYP liabilities, so the clearance here is mixed. It's renal and hepatic. So we have very well-behaved tight PK. The half-life was engineered to be very short, so the effective half-life is roughly 12 hours. But what we've learned now with further mechanistic work, and we have a team that has been in this space for a very long time and presented some of this data recently, is that BHB-1893 seems to have potency that doesn't change if you have an HCM mutation.
That's an important fact because it allows us, basically with the other agents, you're seeing a right shift to that potency, so you need more drug to have some of the diastolic effects, for instance, that you can see with CMIs. As a result, we have very minimal systolic cost for fairly dramatic diastolic benefit, which is one of the reasons why we think we're seeing such rapid deep gradient responses as well as larger improvements in diastolic parameters like E prime in the non-obstructive population.
I see. Okay. So maybe let's dig into each of these, first with obstructive, because you mentioned Hengrui, your partner, has a phase III data set in first half of next year, and you yourselves have another data set in second half of next year, an interim analysis. So maybe back to the phase II, which I believe was run by Hengrui, what did they see in obstructive? You mentioned AHA, maybe mention what you saw in LVOT, for instance, pVO2. What can we kind of conclude with fair to high certainty at this juncture?
Right. Maybe stepping back, peak pVO2 and CPET, cardiopulmonary exercise testing, was introduced as part of primary endpoint for the mavacamten program, in part because improving gradient enables patients to increase their cardiac output, which ultimately allows them to improve exercise capacity. Gradient relief and depth of gradient relief has been associated with improvements in peak pVO2. What we observed in that phase II study was that within a week, we were getting patients down to normal Valsalva gradient. Valsalva being you have patients bear down and measure the gradient, and that is important because it more reflects what happens during exercise. What we observed in the 52-week data was very much the same.
Patients had a rebound in their gradient after being washed out back to baseline, and then they had a rapid deep resolution of gradient, deeper than with the caveats of a cross-trial comparison than has been seen before. We had 90% of patients at one point or another in the study have what we considered a complete gradient response, which is a normal Valsalva gradient, less than 30. That ultimately, with greater time on therapy, what we saw was initially patients had a baseline cardiopulmonary exercise testing, a 12-week CPET. They were re-baselined after being taken off therapy, and then they had an end of study. We saw 1 point improvement between baseline and 12 weeks, but then we saw 2.7 mL/kg/min improvement at 52 weeks. Keep in mind, these patients were all cycled, which typically leads to a smaller delta.
We found those data tremendously encouraging, given what has been seen before. It also, I think, points to this important piece, which is that gradient relief, ultimately, deeper gradient relief over time could result in greater improvements in peak pVO2, which is the registrational endpoint.
Mm-hmm. Your own phase III has this interim readout in second half 2027. I think it is a 12-week look at LVOT gradient. The question would be, why has Bristol Myers Squibb or Cytokinetics never pursued this unique hemodynamic marker, and why you guys can pursue it? Do you have FDA alignment around that?
Sure. We are in a unique situation where we have a fully enrolled ongoing phase III study with our partner, Hengrui. It is also a unique situation in that as we think about the relevance of that data set, which is China only, to a global market, this is a disease where the patients with HCM are patients with HCM, where no matter where they are, the disease biology is the same, the practice patterns are very similar. The previous response to CMIs has been similar, and we know the PK translates well, given we have an ethnobridging study. We will have CPET data from that study to really help us understand the effect on exercise capacity, but then are using gradient to really understand translation to global population and PK.
We have had very positive, favorable feedback from the agency about the approach of using that combination, that data package, as an initial filing plan, understanding that they also asked us, with very specific feedback, to include Asian patients in our studies so that we could directly compare pharmacokinetics. I will say that the agency has agreed in the past, in some context, to gradient. For instance, the SCOUT-HCM program, which is the pediatric program for mavacamten, the primary endpoint for that study was gradient. But this would not be in absence of also having the prior registrational endpoint.
Right, which comes out later in maybe first half 2029 for your own study.
Yeah, the catalyst for obstructive HCM schedule, I would say, is first half of next year for the ongoing phase III study, which is on top of standard of care. The primary endpoint of that study is a composite of peak VO2 and NYHA improvement, same as mavacamten's approval. Our study, where we will have an interim analysis, 60 patients, 2-to-1 randomized, so 40 patients on drug, in the second half of next year, will also be presenting data, one-year follow-up data in non-obstructive HCM that is going to be valuable to see how time, with the depth of diastolic effects we have had, improves key measures of feel and function. We have not announced publicly the exact timing for the full phase III readout, but with our current strategy, we would expect that would be part of a potential sNDA.
In the best-case scenario for obstructive, when could you get approved?
With this plan in mind, we would expect to be filing either end of 2027 or early 2028.
Very fast, basically. Potentially, yeah. You mentioned, we discussed the phase II data was based in China. One of your studies that you're leading, I think, is global, but you mentioned you did an ethnobridging study. Can you walk us through the conclusions of that? Were there any differences that you saw? Any details on that?
Sure. We did disclose this in our S-1 as well, for context. There was an ethnobridging study done in Australia with Caucasian patients. Typical pharmacokinetic parameters were measured, and overall, it looked very similar to the PK profile in Asian patients. It's probably not surprising. Typically, if you have issues with differences, it relates to CYP liabilities, and this is where there are genetic differences in different geographies, and that's not the case here, given the renal part to the clearance.
Just one more on your own is just how did you power LVOT gradient exactly? What kind of separation are you expecting to see for your own dataset, and how would that compare to Cytokinetics and Bristol?
Sure. I think we would expect to see little in the way of change on the metoprolol arm, which was what was observed in precedent studies. I think we feel good about the powering of that IA, given how deep and rapid we've seen gradient impact with this molecule. The study, as you can imagine, because it's 2:1 randomized, it's a slightly larger study overall than the precedent study, head-to-head study, which was the MAPLE-HCM study. So it's a 210-patient study overall.
Hengrui, like we discussed, has the phase III dataset earlier in first half 2027. Like you said, it's a composite on, is it NYHA and pVO2? Any color, what is stat sig on that endpoint?
We haven't announced that publicly. I will say that there's obviously a precedent here with both EXPLORER-HCM and SEQUOIA-HCM also looked at this composite endpoint as an exploratory endpoint. But yeah, I think we feel good, and we're looking forward to that catalyst upcoming.
And then, in terms of execution of these phase III trials, is there such thing as a placebo response, and how are you controlling for that? How are you maximizing study rigor?
Sure. I would say that with regards to peak VO2 as an endpoint, there's been very little in the way of placebo effects in large enough datasets. Part of that, one of the things that we'll be doing in our study is we have a core lab that'll be helping to over-read as well. In the non-obstructive population, where one of our endpoints is KCCQ, there have been placebo effects. What we observed in the phase II study was that at 12 weeks, there's a very similar placebo effect to what was seen in the ODYSSEY-HCM study, so a little less than 12-point change. But we saw a 5.5-point placebo-adjusted change, which was larger than has been seen in other studies, in particular ACACIA-HCM, at a later time point.
What we've observed in these studies is that KCCQ in the treatment arm tends to consolidate and improve over time.
I should have asked in the beginning is, what is your guys' positioning on the use of echoes as it relates to a potential REMS? Is it your position that base case you are expecting a REMS for your compound? Do you have desire to have no REMS ultimately kind of thing?
Sure. Stepping back, this is the disease where we have a very good treatment target, and it is an objective treatment target in the obstructive form that is gradient. Physicians, in our experience, want to know what their patient's gradient is because we know that obliterating gradient is associated with better outcomes surgically, right? Back when septal myectomies were being done, you would go in and cut a piece out, and if you still had residual gradient, you would go back in and cut another piece out. I think it is important to note this is a very valuable part of the treatment paradigm. The REMS that exist today, the challenge with them largely pertains, in our market research, to the multi-step echo-based titration.
As I noted, one of the key mandates of the program is to get to efficacy that we think could be differentiated, but to get there simply. That is where, really, this combination of a one-step for patients who really need it and a fixed dose for most patients, we think is going to be a really large step forward. To be honest, the question of REMS or no REMS pertains more to the qualities of those REMS, right? We have seen that now with a more flexible REMS for aficamten. I think we are really seeking to move the needle there in a meaningful way. We continue to have great conversations with the agency about how to prospectively generate data to show that these echoes, which largely are normal and done in patients who feel well, are probably not necessary.
Speaking of generating more data, is that one purpose of the TRUEHEART-HCM study, where it is the complete fixed dose, I think? Are there echoes in that study?
Part of the goal of that study, there were two major goals. One is to explore the role and demonstrate using data that a fixed dose over time, there is an impact of time here, as we saw in the 52-week data, where we saw more remodeling in 52 weeks than 12 weeks. We believe that in some or many patients, we may not need to titrate at all if we give them more time on therapy. That is goal number one. We feel confident in that because of the data that we saw from this extension study, including data showing that 67% of patients had a complete response prior to titration, which is higher than has been seen before with other agents at 24 weeks.
The second real mandate of the program is it gives us another opportunity to generate more data in a global population on top of standard of care, because our other study design is a head-to-head study. It gives us the opportunity to enroll those patients who may not want to wash out from beta blockers, which is part of the design of the LIONHEART-HCM study.
Okay. Then in the last couple minutes is shifting gears to non-obstructive. Can you remind us when you are starting the phase III and when data could arrive? First, let us start with that.
Sure. We have indicated publicly that we are, and we would re-indicate today, that we are on track to get that study launched in the first half of next year. We continue to be very encouraged by the feedback we are getting from sites in terms of participation. We are planning that to be a 36-week study. We have not publicly given guidance around the timing of that readout, but we certainly are moving at pace.
Hengrui also ran another phase II non-obstructive study. What exactly did they see, and how does that compare to Cytokinetics?
I think those data, in particular for us, again, with all the caveats of cross-trial comparison, pointed to three different things. One is that we saw very rapid onset of improvement in biomarkers. NT-proBNP, which is a marker of heart failure, was reduced very rapidly. Troponin, which is a marker of a cardiac injury, but also in some ways of diastolic function, because the heart perfuses itself as it relaxes in diastole. If there's problems with that, then you can have a high troponin. The second is that we saw very encouraging improvements on measures on echo. It's not gradient, because there is no gradient, but one on measures of diastolic function, where we saw normalization. It was a larger change in those measures, E prime. There's two different types, septal and lateral. Those are measures that cardiologists care a lot about.
The second is that we saw improvements in measures of back pressure, left atrial volume. We saw large improvements towards normal. When you look at a publication, for instance, that was published at ESC this year, looking at aficamten in particular, you didn't see any improvement in LAVI. The third that was very encouraging is we saw, again, I mentioned this, but large placebo-adjusted improvements in KCCQ. We saw dose-dependent but meaningful improvements in peak VO2 already at 12 weeks. Overall, for us, it gave us conviction that we think we have a molecule here that is acting in a differentiated way, that is a diastolic agent, and that looks to have a potential for really best-in-disease efficacy.
Mm-hmm. The last one is just for that phase III, fair to assume it'll maybe mimic the ACACIA Cytokinetics study, where there's a Is it a dual endpoint, not a co-primary on pVO2-
That's correct.
and KCCQ? Very clear. Thanks for all the updates. I think that's all the time we have. Good luck on the phase III readouts next year. Thanks, everyone, for listening.
Thanks, Andrew. Thanks, everyone.