BridgeBio Pharma, Inc. (BBIO) Earnings Call Transcript & Summary

Good day and thank you for standing by. Welcome to today's discussion with BridgeBio, which will cover the company's additional clinical outcomes data from the Phase 3 ATTRibute-CM study in patients with transthyretin amyloid cardiomyopathy. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to introduce your speaker today, Dr. Daniel Judge, Professor of Medicine and Cardiology at Medical University of South Carolina and Co-Chair of the ATTRibute-CM Steering Committee; Dr. Neil Kumar, Chief Executive Officer, BridgeBio Pharma; Dr. Jonathan Fox, Chief Medical Officer, BridgeBio Pharma, Cardio and Renal. I would now like to hand it over to Dr. Kumar.

Thanks so much, operator, and thanks to everyone who's listening in. Happy Diwali. I'll make my introductory remarks brief as I'm excited to listen to and learn from Dr. Judge today. I'll start briefly where I always start, which is a hearty thank you to the patients, families, staff, and physician leaders who make the work we do in ATTR cardiomyopathy and with acoramidis happen. I'm particularly reminded of this today in having the privilege of listening to Dr. Judge. In Dan, ATTR patients have a tireless supporter, a thought leader, and a courageous clinical researcher. In the face of bad or good data alike, I'm always keen to seek Dan's guidance so that we can best understand how to serve the ATTR community. So today, Dan will present the bulk of his presentation, which he delivered just minutes ago at AHA. The data he presents will represent the next important step in our continued interrogation of the ATTRibute clinical trial. One of Bridge's core values is that every minute counts, which is a reminder to us that every gain in time we can achieve in R&D is to the benefit of the patient communities we serve. Likewise, a drug's time course of action is critically important to the patients that we serve. We seek as early an onset as possible, and with as long and sustained an effect as can be had. Coupled with earlier detection, it's exciting to think how we might meaningfully improve outcomes in ATTR cardiomyopathy given the data we'll present today. Before Dan gets to that data, I'd like to briefly remind everyone of the acoramidis program hypothesis to set some context. I'll refer here to slide numbers as we proceed, so you can follow along. Starting on Slide 4, we believe some, now almost 8 years ago, that designing against 2 key principles might allow us to optimize clinical outcomes for patients. First, and most importantly, we believe that stabilization matters and that the more stabilization the better, so we strive to design a near complete stabilizer. 3 lines of evidence suggested that higher levels of stabilization should lead to improved clinical outcomes: first was the well described genotype/phenotype of the disease state where decreasing levels of thermodynamic stability in the context of pathogenic variants reliably led to worsened outcomes. In addition, partial rescue of the outcomes occurs in the presence of a mild restabilizing mutation, and full rescue seems to occur in the context of the fully restabilizing T119M mutation. The second line of evidence was the outperformance of the 80 mg dose in the ATTR-ACT trial. 80 mg achieves approximately 50% stabilization versus the 35% stabilization or so demonstrated with the 20 mg dose of tafamidis at clinically-relevant concentrations. Finally, our experience in the polyneuropathic setting, which showed that high-dose diflunisal at 70% stabilization seems to deliver results in excess of those of TAF 20 mg stabilization and similar numbers to that of inotersen, a 70% knockdown, continued to persuade us that higher levels of stabilization and better control of toxic monomer should lead to better results for patients. So the more stabilization, the better, which is why we wanted to design a 95%-plus stabilizer. Secondly, if through stabilization we were able to minimize the amount of toxic monomer circulating as effectively as any knockdown, which should be possible with a high degree of stabilization, we sought to keep the TTR tetramer present. According to the latest [ NowGene ] analyses using Bayesian or AI techniques. The TTR gene is reliably in the top quartile of cell-essential genes. No wonder, given it is conserved across all species, no species, that we know of, is haploinsufficient or null for the gene, and it's conserved at high energetic costs at a protein level with high concentrations and a less than 2 day turnover. Its removal results in deficiencies with vitamin A transport and other issues in ATTR mouse models. On Slide 5, you can see a cartoon of the path of mechanism of ATTR cardiomyopathy. Briefly, the way we were able to design the highest levels of stabilization that we have measured to date is by phenocopying the aforementioned T119M rescue mutation. We exhibit several published advantages to other stabilizers, including being less albumin bound, so we see significantly more target having a superior binding profile with a lower kd2 as has been published, so that we bind a target more effectively; and employing a more enthalpic binding mode to better glue the tetramer together. The results of these 3 features have, across 4 assays in the hands of several laboratories, suggested superior stabilization than anything we've seen in development or on the market, a result complemented by our post-hoc analyses of serum TTR increases in the presence of ours or other stabilizers in our clinical trial. With that context, I'll turn it now over to Dr. Judge to present more on the data from ATTRibute.

Thanks, Neil, for your kind words. Next slide, this is Slide 7. This is the trial design, the study design for the ATTRibute-CM Phase III trial. The key eligibility criteria are shown here. We required patients to have a diagnosis of ATTR cardiomyopathy made either by a biopsy with mass spectrometry or if light chain amyloidosis had been excluded [Technical Difficulty], we were able to use noninvasive scintigraphy to make the diagnosis. New York Heart Association was class 1 to 3 for eligibility. Randomization was 2 to 1, 800 milligrams of acoramidis twice daily versus placebo. The efficacy assessment was based on prespecified modified intention to treat analysis for those with an eGFR 30 or greater. Tafamidis became available around the world at different times through the course of the study, and the tafamidis use was excluded for the first 12 months. After that, there was an imbalance in the amount of tafamidis used, 14.9% in the acoramidis treatment arm versus 22.8% in the placebo arm. The mean duration of tafamidis treatment was 11 months. All participants who reached the 38-month endpoint were able to go into an open label extension study. Our 30-month primary endpoint was a Finkelstein-Schoenfeld analysis, hierarchical testing of the all-cause mortality, cumulative frequency of CV hospitalization, change in baseline NT-proBNP, and the change in baseline from the 6-minute walk test distance. Next slide. This is Slide 8. These are the baseline demographics for our cohort. You can see that they're well balanced for things like age, sex, and hereditary versus wild-type disease. These demographics reflect modern day populations of patients who are diagnosed with this disease were balanced for things like NT-proBNP, eGFR, serum TTR, KCCQ, and 6-minute walk test distance. We show here for the first time the National Amyloidosis Centre's staging system, stage I being mild disease, 2 is moderate, and 3 is severe disease, and you can see 10.9% versus 10.4% severe disease, acoramidis versus placebo. Next slide. This is Slide 9. These are the primary results that you've seen. The hierarchical analysis, Finkelstein-Schoenfeld method was substantially positive with a p value of 0.0001 using the standard Finkelstein-Schoenfeld analysis, a win ratio of 1.8. It's important to note in this analysis that 58% of ties were broken based on the first 2 components, the hard endpoints: all-cause mortality and frequency of CV hospitalization. Selected secondary endpoints are also shown here. P values again all 0.0001 for frequency of hospitalization, change in baseline, 6 minute walk test distance, KCCQ-OS, the serum TTR change, and this change in NT-proBNP. All-cause mortality in this less severely affected population than in historic trials was 0.057, a relative risk of 25% reduction. For CV-related mortality, it was a 30% relative risk reduction using certain method of analysis, a p value of 0.037. Next slide. This is Slide 10. This is the first time we're showing this the Kaplan-Meier survival curves using a much more traditional endpoint. This is time to first event or death or first hospitalization. The curves diverge at 30 months with acoramidis, of course, being favored, and they continue to diverge through the course of the 30-month study. This translates to a hazard ratio of 0.645 and a p value of 0.0008. This means the number needed to treat to prevent a first hospitalization or death in 2.5 years is 7. If we look at a Finkelstein-Schoenfeld or hierarchical analysis that just includes the 2 hard endpoints, all-cause mortality and CV hospitalization, that p value is also significant, 0.0162, favoring acoramidis over placebo. Next slide. This is Slide 11. This is the frequency of CV hospitalizations, starting with actually the number of subjects who had a hospitalization acoramidis 26.7 versus 42.6 in the placebo-treated cohort. There's a 50% reduction in CV hospitalizations. The frequency per year mean 0.22 in the acoramidis that's modeled and 0.45 in the placebo arm. The relative risk ratio 0.496 and a p value of less than 0.001. That means the number needed to treat to prevent 1 CV hospitalization per year is only 5. Next slide. This is Slide 12. No safety signals or potential concern were identified. We know this is a well-tolerated drug, and the outcomes for treatment-emergent adverse events were quite similar. Next slide. This is Slide 13. This is where we put these results in context. These are contemporary data, again, in a newer era where we're diagnosing this disease earlier. We have new expectations for our patients. Their expected outcomes are better. 30-month mortality at this point with the placebo-treated cohort -- sorry, this is the comparison of what we've done with our current trial versus an older trial. The 30-month mortality rate in the ATTRibute-CM placebo cohort was 25% and in the ATTR-ACT study with tafamidis was 29.5%. Of course, I don't mean to suggest that placebo is better than tafamidis in this. It's just that these are 2 trials that are in very different populations, and it's really difficult to compare things like mortality in a study where our levels were low and our outcomes were better. Outcomes in acoramidis treatment population approach the age-matched general populations, 81% survival on acoramidis approaches the survival in an age-matched U.S. database of 85%. The 0.29 observed mean annual CV hospitalization frequency on acoramidis approaches annual hospitalization rates observed in a broader U.S. Medicare population 0.26. Our Kaplan-Meier Analysis time to separation demonstrated at 3 months represents the most rapid clinical benefit on a composite endpoint of all-cause mortality and CV hospitalization outcomes in an ATTR cardiomyopathy trial to date to our knowledge. Early and profound reductions in CV hospitalization can have a significant impact on public health and reduce overall treatment costs, if we estimate a $20,000 cost for each cardiovascular hospitalization in the U.S., a CV hospitalization, different than a heart failure hospitalization, which would be even more expensive. CV hospitalization is also a predictor of mortality in general heart failure as well as in ATTR cardiomyopathy. Next Slide. This is Slide 14. Our conclusions are that acoramidis improves clinical outcomes in ATR cardiomyopathy. The ATTRibute-CM study results demonstrate that acoramidis improves clinical outcome based on all-cause mortality and CV hospitalization and the hardest 2 endpoints in ATTR cardiomyopathy patients. The primary endpoint, the 4 component FS analysis, showed a significant treatment benefit of acoramidis over placebo, with the majority of ties being broken by the 2 components at the top, all-cause mortality and frequency of CV hospitalization. We had notable early separation at 3 months based on a time to first event Kaplan-Meier analysis with a number needed to treat to per event in event of death or CV hospitalization over 2.5 years of 7. The 2 components, Finkelstein-Schoenfeld analysis shows a significant treatment benefit of acoramidis over placebo. For individual outcome components in this earlier disease, less severely affected, earlier diagnosis population, 25% relative risk reduction for all-cause mortality, a favorable trend, and a 50% relative risk reduction in the cumulative frequency of CV hospitalization, with a number needed to treat to prevent 1 CV hospitalization per year of 5. Next slide. No we'll turn it back over to Neil.

All right, great. Thank you, Dr. Judge. I'll briefly conclude the call before we get into Q&A by reiterating some of what Dr. Judge said within the context of the emerging use case for acoramidis. On Slide 16, you can see the substance of the use case. Above all, we believe that near complete stabilization should lead to profound clinical outcomes for patients, as I mentioned earlier. The first fingerprint of this is the absolute survival and hospitalization rates observed. As Dr. Judge mentioned, hospitalization in turn being a potential predictor of downstream mortality as well. The second fingerprint of near-complete stabilization is improvement, patients actually getting better on treatment over time. And the third fingerprint is the time, course, and early separation that Dr. Judge just went through. Put together, we find patients surviving more, going to the hospital less, improving more, and benefiting earlier than we have seen before to our knowledge. Slide 17 provides further detail on that first fingerprint, the absolute survival and hospitalization rates coming close on treatment to what life might look like without ATTR cardiomyopathy for our patient population. These data speak to the improvement in both medical management as well as to the power of our treatment. On Slide 18, you see recent data from Dr. Masri at HFSA showing again how exciting the improvements in treatment and care are for patients. In his slide here you can see the absolute survival of 81% as being uniquely exciting. And you can see our placebo, as Dr. Judge mentioned, also is resetting the bar for what medical management can do for patients alone. Interestingly, you can see here in the real-world evidence study that tafamidis, in its current clinical context and in a population of healthier patients than ATTRibute as measured by baseline NT-proBNP, delivers longitudinal survival numbers similar to what they saw and ATTR-ACT. Turning to Slide 19 and turning now to data on improvements. As we have previously described, we see a higher-than-expected level of patients actually improving in our trial. Near half of the patients as measured by NT-pro or by 6-minute walk distance, and 13% of patients actually improved in NYHA class. These are new data, these NYHA class data, and these observations are for patients at 30 months. Even if we take all missing data and impute it as negative, the most conservative statistical treatment of these data, you still measure the highest improvement rates that we have seen in this space to the company's knowledge. We will continue to elaborate on these improvement data in the months to come and hopefully at ACC. Slide 20 show the data that Dr. Judge already reviewed, so I won't spend time here, but we'll just remind the audience once again that: 1, these data represent the earliest separation we have seen against this measure; and 2, that the profound impact on CVH is not only important to patients and payers alike, but could also be predictive of downstream effects on mortality, as being published already. Slide 21 then returns us to the roof of the house from Slide 16, our core hypothesis. This data is amongst the most exciting that I have seen from our trial. Might we continue to learn more about how stabilization relates to improved outcomes from our dataset: first, as you can see from our chart, we see a highly statistically significant relationship between increases in serum TTR, an in vivo measure of stabilization, as you know, and outcomes in our trial. This is true against any of the continuous outcomes that we have looked at. Further, we can ask, given the tafamidis drop in our trial, how relative levels of serum TTR increase on TAF or on acoramidis. I'll caveat this by saying obviously these analyses are post hoc and exploratory. Consistent with the extensively published findings that I talked about earlier, we see outperformance of acoramidis versus tafamidis in terms of serum TTR elevation. And so within our single experiment, it appears that thesis holds that ever-improving levels of stabilization are leading to ever-improved outcomes. Skipping forward to Slide 23, I'll wrap up here. You can see our next steps. We plan to submit our NDA by year end, followed closely by submission in the EMA, presently underserved geography writ large. Importantly, and in concert with, our aggressive publication plan, we are kicking off a primary prevention study very soon and plan to, as I mentioned before, continue publishing as much data as possible from the ATTRibute trial at ACC and conferences beyond. So with that, I want to thank Dr. Judge once again, and I'll turn it over to the operator for Q&A.

[Operator Instructions] Our first question comes from the line of Salim Syed with Mizuho.

Happy Diwali, Neil. I wanted to spend a little bit of time on Slide 10, if I can. So I'm trying to tease out the cardiovascular hospitalization from the composite. I apologize for the math in this question, but when you guys separated on all-cause mortality at month 19, in the combined here, you're showing month 3. It seems like from the events from you ESC slides and the slides that the events are roughly equivalent between ACM and CVH, if I'm not mistaken. So just trying to do the weighted average math. Is the right conclusion here that CVH separates at month 0, or what am I missing in the math exactly. I'm just curious if you can provide the CVH-free number at month 30.

Sure. Jonathan, you want to take that? I can elaborate.

Yes, sure. So, actually, when we look at the Kaplan-Meier curve for CVH all by itself, it looks a lot like this one for obvious reasons, because it does separate as early as month 3. But in a way, if you think about it, whether somebody gets hospitalized first or if they die, in a way, it's like competing events. So in other words, to put into blunt language, you can't be hospitalized if you've already died. So in a way, this shows a very early treatment effect on reducing hospitalization, which, as Neil mentioned, is itself a predictor of subsequent mortality. So the fact that the drug is having this treatment effect so early on both this hard end point of hospitalization, as well as what we saw in the early separation in N-Terminal-proBNP, the early separation we saw at KCCQ suggests that even in the absence of any obvious effect on what's already accumulated in the heart, which probably is more of a driver of subsequent mortality in this population, that just halting the deposition of freshly misfolded and aggregated monomers seems to -- it's an inference we're drawing that may be having an important biological effect as reflected in the things that I just mentioned that do separate early. Does that help answer the question?

Yes, I think I got it.

Our next question comes from the line of Mani Foroohar with Leerink Partners.

Apologies for background noise. I'm taking this one from the road. I guess, more for Dr. Judge. When you look at this data, obviously, cross trial comparisons are fraught, but there's an argument for certainly the best mortality that we've ever seen in a population like this, recognizing baseline differences. In real-world practice, either in an academic setting like yours or more of a community setting, how do you think clinicians will think about which patients should be on acoramidis upon diagnosis versus which patients should be switched from tafamidis to acoramidis? And I have a quick follow up.

Yes, certainly a great question. And as a physician, I'd love to say that my opinion is the [indiscernible] as to what drug or what will people get, but a lot of these decisions are made by payers. If you look closely, as you mentioned, cross-trial comparison, it's hard to do and hard to draw valid comparisons. But if you look in vitro, there's clear data with the same drugs at physiologic concentrations that show that acoramidis is a better TTR stabilizer. And the hypothesis that better stabilization leads to better outcomes is well established. So I'm certainly a fan of acoramidis as first-use and as potentially something that would be moved over for those who are certainly doing worse. If they're on tafamidis and getting worse, there needs to be another answer or a better answer. You said a follow-up?

Yes. I think that you were leading into the follow up because you probably know exactly where my logic was going. So a number of physicians, and certainly a number of the competing companies, one in particular, have talked about tafamidis non-responders as potentially being a population that should be treated with other novel therapies. I'm trying to think about in the real world, how does one define a tafamidis non-responder. If they pass away from their disease, that's obviously a non-responder but not helpful from a clinical decision-making perspective. How do you decide what a tafamidis non-responder is? How do you decide from that point? Okay, we move them to acoramidis. Okay, we move them to, a hypothetical, if ever approved, oligo, et cetera.

Yes, I think the whole field has struggled with this question for a long time, because payers often ask that, prove failure of 1 therapy before going to another therapy. And traditionally, my answer is, well, we know tafamidis only lets people get worse more slowly. It doesn't make them better, so it's hard to define. But at the same time, we can come up and colleagues and I are discussing things like that, a 20% rise in NT-proBNP or a 10% rise in troponin or biomarkers that we know do correlate with outcomes in response to therapy, and those progressive rises or worsening of biomarkers or TTR levels, another good marker of TTR stability may be the ones that we'll use to mark a failure of response or insufficient response to established therapies.

I know this is a follow up to a follow up, I'm sorry. I know there's a lot of other analysts on the call. Just wrapping up. So, as you said, this is in part payer driven because of the PDM environment, et cetera. What needs to happen for those datasets that you discussed, those biomarkers? Do they have to go into AHA guidelines? ACC needs to weigh in? What needs to happen for that to be something that physicians can point to and payers to respect?

I think guidelines are important and guidelines in this field need to be updated regularly because the field is changing very quickly. But I think part of what we do as a community, as healthcare community, we'll hopefully help the payers have better answers. In my experience, unfortunately, payers often come up with their own criteria, and we'll need to deal with those.

Our next question comes from the line of Paul Choi with Goldman Sachs.

Dr. Judge, thank you for your talk. It was very well received in the room. My first question is for you with regard to the early separation of the Kaplan-Meier curve that you discussed and presented. When you separate out other factors like COVID, can you maybe characterize what the curve looks like since the trial completed enrollment in 2020, just to help us understand again the impact from that on all-cause mortality? And my second question for the company is, as you think about the filing, for Neil and team, are you going to try and seek time to event and number needed to treat in the label here? Clearly, the numbers are very low and the separation of the curves aren't clearly suggestive of anything. So any color there would be helpful.

Well, I guess I'll take the first question in terms of effect that COVID may have had and how might things be different if we did the study today. We don't think that there was a big effect for COVID in terms of the numbers and the things that we're seeing on this slide that's showing now the Kaplan-Meier survival analysis. So I don't think it would be much different now.

Yes, it's Jonathan Fox. Just to add to what Dr. Judge just said. If anything, just anecdotally speaking with the investigators in various parts of the world, they actually made a lot of effort to help keep their patients out of the hospital. So if anything, it probably depressed our overall hospitalization rates or event rates in this trial. And so despite that, I'll call it a relative handicap, we still saw the results that Dr. Judge presented. There were, in fact, only a handful of serious morbid or mortality events associated with COVID. So, if anything, patients and the physicians taking care of them made efforts to stay away from the hospitals, which as everyone knows, were a pretty bad place to be during the height of the pandemic.

Yes. Just to build on that, Paul, from the standpoint of mortality, as you can see, based on the differences in relative risk reduction between all-cause mortality and cardiovascular mortality was roughly the same. Cardiovascular mortality was slightly improved, actually, as compared to ACM, with a similar absolute risk reduction of 6%. Yes, on your second question, obviously, we can't comment broadly on what will or won't be in the label. We have worked up the USPI and things will be going in that will be an ongoing discussion with the agency. Certainly we feel that these data are important to a prescribing physician, but the form in which they take, be it on the label or through medical publications and further education, that'll be the product of an ongoing discussion here.

Our next question comes from the line of Dane Leone with Raymond James.

Congratulations on all the data. I'll keep it to 1. I was just curious, are you guys going to have additional post hoc analysis of really understanding what qualified for some of the patients to get drop in tafa in the study and how that could influence your go-to-market approach for switches on patients that could be considered either progressing or inadequately controlled on to tafamidis commercially?

I could take that at a high level. Yes, we'll continue to look at that at a patient-by-patient level, Dane. But broadly, the reason that people dropped in on TAF, given the fact that no one knew if they were on placebo or active drug, was simply access to the drug. As you well know, we biased the trial toward European clinical sites. And those sites, by and large, didn't have access for a long duration within the [ ATTRibute ] time course. And within the U.S., as you know, access has been quite challenged to tafamidis. So as people gained access for one reason or the other, then they dropped it. So it wasn't as if some clinical presentation on acoramidis or otherwise on placebo was predictive of drop in, or at least as we can see so far.

Our next question comes from the line of Anupam Rama with JPMorgan.

Maybe one for Dr. Judge. Given the differences you described between ATTRibute-CM and ATTR-ACT, what data points in ATTRibute-CM specifically do you think most underscores differentiation relative to tafamidis? And what resonates the most with the KOL community in your discussions?

So I think the main difference is the earlier diagnosis and so the less severe progression of disease. We know that this disease is something that progresses more quickly at the later stages, and if we get the patients enrolled at an earlier time point in their course of disease, then they're reaching an endpoint is going to be less likely. What resonates in the community? I think there's really a lot of excitement, and this came up in our panel discussion in the session this morning about having more than 1 drug that's available and effective for ATTR cardiomyopathy, particularly wild-type disease, where right now tafamidis is the only drug. And so that's the most important piece of news in my community, which is it's nice to have a choice, it's nice to have some competition in the field.

Our next question comes from the line of David Lebowitz with Citi.

First, on the primary endpoint. Could you run through the impact of urgent care visits on the endpoint that was not included in the tafamidis pivotal?

Jonathan?

Yes, Jonathan Fox here. So we included events of clinical interest, which basically was defined as urgent visits either to the emergency department or more commonly to the heart failure clinic for intravenous diuretic. And that made a reasonable contribution to the total number of events. So as I mentioned earlier, a lot of the patients were encouraged to essentially stay away from the hospital or to avoid hospitalization if they could. So utilization of that treatment pathway, actually, it was not a majority of the events, but it did make an important contribution to the total number of what was grouped together as CV-related hospitalizations. To put it maybe in a different context, most of the CV-related hospitalizations were for 1 aspect or another of worsening heart failure.

And jumping over to patient severity and baseline comparisons. Obviously, a handful of people have asked questions about comparing across trials. Is there a plan at some point to do subgroup analyses based on baseline heart failure class?

Yes, as you can imagine, we have a pretty long laundry list of secondary analyses that we plan to conduct and report on as we can generate them. And certainly outcomes by New York Heart Class and I think maybe even more relevant to the National Amyloidosis Centre staging are certainly in the works. I guess the reason why I said the last comment was the NAC staging basically stratifies people by whether or not they have an eGFR of less than 45 or greater than or equal to 45 ml per minute per 1.73 meters squared, over an N-Terminal-proBNP less than or equal to 3,000 picograms per ml or greater than 3,000. And when you put those 2 things, if you have the renal impairment and you have the higher N-Terminal-proBNP, that's a NAC Stage III. And if you have one or the other of those, that's NAC Stage II. So everybody else falls into Stage I. So I think that if you look at the subgroup analyses on our primary, according to those individual components of the NAC staging, you could probably predict what we're going to find, which is a positive treatment effect at all the stages. That is really a very important contributor to the overall long-term survival that we expect in this disease, in a sense that both low renal function and high N-Terminal-proBNP are independent predictors of mortality in this disease.

Yes, maybe at a higher level. Actually, it's a really important question. It stems back to one of the earlier comments as well, which is, what are the different stages and how do we consider who's not responding to 1 medicine versus another. And so we're trying to work closely with the community to figure out precisely what the analyses are. That can help people to understand, okay, this is the right marker of progression. Is it NYHA class? Is it NAC stage? Is it NT-proBNP? And you heard some of that discussion at the ADCOM for APOLLO-B as well. And then what benefits might our drug be having at those different stages? I think it's not clear cut that NYHA is necessarily the only way to look at it. As Jonathan mentioned, we have nice and positive data across all classes there and improvement data, but we also have nice improvement data with NAC staging as well. So putting that all together in the context of something that's simple and can be communicated to payers and KOLs and physicians is something that's on our mind. But we don't have the full answer yet.

Our next question comes from the line of Ellie Merle with UBS.

This is [ Jasmine ] on for Ellie. Congratulations on the data. We were just wondering if you could give a little bit more color on the impact of tafamidis dropping on this time to event analysis. If you've done that analysis. So what separation and when do you see it in the tafamidis dropping group versus those without?

Jonathan.

Yes, It's Jonathan Fox here. So, as was mentioned, we saw, on a percentage basis, twice as much drop-in on placebo as on acoramidis. It is important to note, however, that as we conducted the trial, we had no way of knowing who was going to drop in, where they were going to drop in, and when they were going to drop in or so forth. Really, the best we can do is look at the results of the analyses over the entire efficacy population of 611 versus excluding those who dropped in from either arm. And essentially, the result is the same, if that answers your question.

And Jasmine, just to add on, this is Ananth Sridhar here. Just to remind you that tafamidis drop in was not enabled in the protocol until after month 12 in the study. So the separation observed as soon as month 3 is independent of any tafamidis drop in.

Well, maybe just as importantly. That's a great point, Ananth. Just as importantly, even after month 12, the slopes of those 2 KM curves don't change.

Our next question comes from the line of Greg Harris with Bank of America.

This is Mary Kate on for Greg. Thank you for the further analysis on clinical outcomes here. I guess, how do you anticipate these impacts from acoramidis treatment to maybe also translate into quality of life for the patient? And then as a follow up, what feedback have you received from patients and physicians regarding interest in acoramidis?

Well, I think our best way to quantify quality of life is the KCCQ, and we do show certainly a difference. And I guess I can look to patients not only who are in the open-label extension for the Phase III study, but I still have a few from the Phase II study who are in the open-label extension, and they're thrilled. So, honestly, I've seen no one who said, I'm done with this and I don't want to be on something experimental, I want to get something else. The drug is very well tolerated, as you saw in the safety slide. And overall, I think the reception in the amyloid community is phenomenal.

So I'll just add to that. Just a couple of things to point out, the KCCQ overall summary score includes components of physical functioning as well as general symptoms of heart failure and overall sense of wellbeing. And all of those things track together in the overall score with a very early separation. Goes back to this idea that I floated earlier that trying to interpret what the underlying biology might be that would lead to an early separation in N-Terminal-proBNP rise with a very big difference at the end at 30 months, with almost half of the people at 30 months actually having a lower N-Terminal-proBNP than they did at entry, walking farther at month 30 than they did at entry. All these things taken together are very consistent with the impact on quality of life. So as we look at the overall results and the totality of the data, it's not just people living longer and staying out of the hospital, but it's living better and feeling better is one way to look at these data.

Thank you. I am currently showing no further questions in the queue at this time. I'd like to turn the call back over to Dr. Kumar for any closing remarks.

Sure. I'd like to thank everyone again for their time and to especially thank Dr. Judge for his time. We look forward to continued interactions with everyone on a go-forward basis. Thank you everyone for your time. See you.

This concludes today's conference call. Thank you for your participation. You may now disconnect. Everyone, have a wonderful day.