Recording in progress. This is only for Jefferies clients. Members of the media and the press are not authorized to attend this event. If you work for the media or the press, you are prohibited from joining the call. The content presented at this event is proprietary to and are subject to the copyrights of Jefferies or third parties. As a matter of legal compliance, we remind you that you must not attempt to elicit from any speaker at this event any material non-public information or other confidential information. Accordingly, the speaker may decline to respond to any question at his or her sole discretion. You may not publish or otherwise publicly disclose the name of, or otherwise identify the speakers unless Jefferies permits it in writing. By attending this event, you agree to all of these restrictions.
With that, I'll hand it over to Eun Yang to begin the webinar.
Thanks, Val. Thanks, everyone, for joining us today. I'm joined by, with BridgeBio Management, Neil Kumar, Founder and CEO, and Cameron Turtle, Chief Strategy Officer at BridgeBio. Thanks for joining us today. As you know, the topic of today is one of the upcoming phase III data, for acoramidis and TTR amyloidosis and cardiomyopathy. Neil and Cameron, maybe we can start with kind of a high-level overview of our phase III trial design and powering assumptions, then we can go into more details.
Yeah, sure. Happy to kick it off. Thanks again, Eun, for the opportunity to chat today about acoramidis, and hopefully we'll get into some other products as well. As you know, this is really going to be the flagship for, hopefully, our ability to build a real leading company in genetic medicine, one that's held by a multibillion-dollar franchise that can help a lot of patients like acoramidis and with several phase II, phase III delivering right behind it. Hopefully, we can build a type of company that's rarely built in the rare disease community. Just in terms of acoramidis, obviously, this is a nested trial design, Part A reading out at the very end of this year, Part B reading out 18 months later. The results that we're looking for in Part A, which will lead to if positive, full approval, would be six-minute walk distance.
In Part B, we'd be looking at CVH and mortality. The trial was designed almost exactly the same as ATTRACT, which most people will be familiar with. It was Pfizer's trial where they were interrogating tafamidis's effect on ATTR-CM patients, with many of the same inclusion/exclusion criteria, and ultimately many of the same powering assumptions. In fact, we were a little conservative, as you and I have discussed in the past, on our powering assumptions, elevating the number of patients so that over enroll in the trial, to ultimately get to the point where if we deliver even the same efficacy as the tafamidis, we would expect a positive trial.
Okay. Obviously the timing of the trial, you said Part A is really very end of this year.
That's calendar.
Do you think there is a chance that it could actually go over to next year?
Do I think there's a chance what? Sorry.
Go over to next year or the next year?
No, probably not. I mean, the timelines are pretty well set right now. We had last patient last visit already. We're in the game just in terms of database close, database cleaning. I suppose if anything's slower, we rely on a lot of third parties, and we're going to be extraordinarily careful throughout the course of this time to make sure that no one is unblinded to anything that could affect Part B. Since this is a nested trial design, that's our number one priority here. We're working with Duke, working with our vendors. I think you should expect data at the very end of this year would be my best guess right now.
This is over 630 patient trial, pretty large. You said last patient exited already. How long does it take to actually lock the database and clean it and then analyze the data?
Couple months.
Couple months.
I mean, typically that's what we found is, it takes a couple of months. I think in this case, it might be a little slower as the data goes through its various, both Duke and our third parties. Usually, it's about two months.
Towards the end of the year, it's a holiday season. Do you think that you could provide more narrow the timing as we get closer to the data?
Yeah, I think we'll probably make an announcement as we get closer, so that people can tune in, as we get more clarity against the timeline. Right now, that would be inappropriate, I'd say.
Part A data coming out toward the very end of this year. This is a 6-minute walk or distance at 12 months time point. Before we get into data expectations, I want to ask you about patient profile that we can compare to the tafamidis's phase III trial. In tafamidis's phase III, about 60% of our patients were categorized as Heart Association Class II, and about 30% was a Class III. When you look at the wild type versus a variant, it was a 75% versus about 25%. Based on that information, do you think your patient profile in the trial will be similar?
Yeah. Basically, I think that, again, the trial design, inclusion/exclusion criteria, everything was basically designed to match up. We've got Class I through III patients as well excluding AL. Very similar approach. I think the only thing that you might think about being slightly different is, my best guess will be that we're going to have more wild-type patients than did Pfizer, just because we're trialing a lot more in Europe. Actually, the V122I mutation is quite common in the U.K., but less common in some of the other countries where we enrolled quite a few patients. We're also finding a lot more wild-type patients now. The levels of diagnosis have increased. I think that will be one difference, would be my best guess, but everything else is going to be pretty similar.
I think what we've publicly disclosed is that if you look at the ambulation amounts for the starting population, they're almost dead on each other. We're at 355 m or so, in that range. Pretty similar. It was designed to be similar.
Okay. Obviously destabilized TTR is the genetic driver for the condition and leading to amyloid accumulation. Better TTR stabilizer would have a better outcome. When we see in vitro data, like last week at the R&D Day, you guys showed the in vitro data looking at the target TTR binding site occupancy, and then your product shows close to 100%, whereas tafamidis at 80 mg showed about 50%-60%. That's the in vitro data. Have you also looked at TTR serum levels in terms of normalization comparing acoramidis versus tafamidis 80 mg?
I think, just to restate what you were just saying, across 4 different fairly well-validated assays, what you can see is near complete and basically complete stabilization with acoramidis versus what's a little less than 50% stabilization at 80 mg tafamidis. Those are obviously in vitro, or we might consider in vitro assays even though you're taking patient sera in many cases. We've also looked at, as you saw from our phase II data, serum TTR levels, which obviously should elevate as you stabilize, and what we saw was about 50% increased serum TTR levels as compared to what tafamidis was able to generate at high dose.
Any measuring one were to look at, I think it's very clear that this molecule is going to stabilize in excess to what tafamidis can, and I think that's consistent with the biochemistry that we know, which is that you're seeing more target, that you've got a superior KD2, so you're binding more target and ultimately doing a better job of gluing the tetramer together, driving that hydrogen bonding at the bottom of the thyroxine binding pocket with a slightly more enthalpic binding mode, as we published in J Med Chem. I think all those dots fit together pretty nicely.
You're right, for some time, Pfizer has indicated that they are seeing a max TTR stabilization at 80 mg, and although your in vitro data looks different. From our perspective, how can you reconcile the difference between what you are showing with your data versus what Pfizer is saying that they are seeing close to max stabilization?
I think that's a product of the way Pfizer has articulated their Y-axes, which has changed in a couple of different FDA submissions. Actually in the appendix of the NEJM paper, in ATTRACT, you can clearly see that they're getting to about 50% stabilization as defined by any of the classic Western blot or FBE assays. I don't think Pfizer's ever come out and said against the assays that are usually used, that they're getting full stabilization. If they have, that's incorrect. Scientifically incorrect. I think the most bold statement that was made was in the amyloid paper that Jeff Kelly recently put out, where he compared 28 micromolar Pfizer, which is their max concentration against 11 micromolar AG10. That was obviously a misnomer.
42 micromolar is our Cmax, and if you compare those, even using his plots, this is the inventor of tafamidis, what you can see is excess stabilization for AG10 that's quite marked. I'm not sure that there's any in vitro assay that says they're fully stabilizing anything. Obviously, if serum TTR levels are elevated with our drug versus their drug, again, in a cross-trial comparison, and I'm not sure that one could claim that either. Look, I think we're close to data. One thing that is for sure is if you were getting full stabilization or close to full stabilization, because remember, there isn't that much of a difference between 20 and 80 in terms of stabilization because of the excess albumin binding that they have.
Four times a dose increase only leads to about 2 and a half times the exposure and a small amount of increased stabilization. Why on earth would diflunisal handily outperform TAF 20 in polyneuropathy? They're not 100% stabilizing anything.
Okay. On six-minute walk distance at 12 months time point, then you said the baseline of a patient is about 350 m, similar to tafamidis. Obviously then, we could expect the placebo with a decline of about 60 m in your trial compared to, or similar to what tafamidis trial has shown. One question, there are a couple of questions. One is that in Pfizer's trial, about 30%, about 1/3 of a patient were on 20 mg dose, which is not optimal. Then they show 25 m decline at 12 months, whereas placebo showed about 60. Do you think that had they had all patients on 80 mg, not 1/3 on 20, do you think that their six-minute walk distance decline would be, I don't know, this is kind of a simplistic way, but about a third less than 25 m?
No.
At least less than 25 m?
That's not how math works.
Yeah, I know.
Come on.
At least less than 25 m.
Yeah, no, I think you could safely say that inasmuch as that six-minute walk distance is connected to mortality, we know that from the Lane paper, and we know that 80 outperformed 20 in terms of mortality. We don't have the exact data for 80 versus 20 in terms of six-minute walk at 12 months, you could assume that 80 outperformed 20 by probably some small margin. Yeah, the patients probably would have done better if they were all, not probably, they would have done better if they were on 80, is my assumption, versus 20.
AG10, acoramidis, shows a near complete TTR stabilization. Isn't it reasonable to expect that it could've hurt? Not just slowing the decline. Could have potentially stopped the progression? That instead of expecting, hey, certain levels, certain m of a decline at 12 months time point, can we actually expect that whatever it shows, whether it's a decline or stabilization, it would be similar to healthy elderly population?
Well, you got to remember ATTR-CM. Yeah, I think it could, and I think it will in some patients. Whether or not a very sick patient with ATTR-CM, it can halt the disease at that point, one doesn't know, because these are sicker patients than our healthy people who are 72 years old, obviously. There's other things going on at that point. I think the theory of the case is that if, 1, we're able to turn down toxic monomer production to the extent that we are, such that you don't get any more further deposition in the heart, that you could come as close to possible as halting the disease. How that manifests in the actual data, if that means that some of the Class III patients continue to decline, but you could catch it earlier, and you may be doing better and better.
Who knows? I don't know. For us, I think a success scenario is if we're able to deliver over 20% relative risk reduction versus where taf was. If it's -25, if we're in the teens, then you start to feel good about, I think, the differentiation, especially if it connects up with, again, 2 x the stabilization, 50% more of serum TTR elevation, outperformance on key biomarkers like BNP, as we saw in our phase II. I think all of those things will go hand in hand.
When we separate data, we are going to see those cardiac biomarker data as well?
We may.
No.
Oh, okay, we're not.
I think the cardiac biomarkers are more likely to come in an academic presentation next year.
Oh, there you go. Academic presentation.
Okay. The clinicaltrials.gov website and cardiac biomarkers were measured. I'm sure that you're measuring periodically, but at 30 months time point and not necessarily at 12 months time point. When the data comes out, Street is going to compare your data to tafamidis data. Obviously, this is a cross-study comparison. As you pointed out, if you show six-minute walks decline in teens versus tafamidis had shown 25 m. Given that it's a cross-study comparison, obviously, we are looking for numerical difference. How much numerical difference really has to be there in order for the financial community to feel more comfortable that, hey, AG10 is indeed is a better drug than tafamidis?
I think the bigger thing we're focused on is what kind of difference we need to show for the physician and patient community to feel that there's a difference. I think that's on a couple different dimensions. 1 is, I think that's why we said 20% or more relative risk reduction, which puts you in the teens, unlike mortality, where the gold standard in cardiovascular disease is like 15% relative risk reduction. I think the second thing is the ability to line up and connect the dots between what we're seeing in terms of better performance on six-minute walk, as well as better performance on serum TTR and some of these other variables that are great univariate predictors of mortality. I think if all that happens, physicians are smart enough, just as they were in hypercholesterolemia and many other diseases associated with cardiovascular disease.
The better you do in terms of stabilization, the better you do for patients. Put it all together, I think people will prescribe this drug.
What kind of a data point did you think you would like to see in order for to have a confidence in that it's going to really translate into better mortality, morbidity benefits? Is there a kind of like correlation between six-minute walk or distance decline versus the cardiovascular outcome benefit?
Yeah. There is. The Lane paper lays all of that out. There's actually a couple different papers, but the best is the Lane et al. paper, which I think was published, I don't know, a year or two ago, that lays out the differences in six-minute walk as they correlate with hazard ratios. Yeah, there's a good predictive element there. There's a good predictive element with serum TTR levels as well. Yeah, I think both of those will be interesting to look at the end.
It's really speculative, but is there a certain level of the six-minute walk distance that you want to see to decline that you want to see in order to have a confidence that, hey, on the mortality, morbidity benefits, our drug would be better than tafamidis?
Oh, yeah. I think if you're talking about something like 8 to 10 m, in terms of a delta, you could expect that First of all, a 12-month difference is going to be exaggerated at 30 months. We know that because it's right. That's obvious. If you're talking about even an 8-meter difference, putting you in the teens obviously, but in the high teens. I think that would ultimately, if you looked at that delta, correlate with a meaningful difference in mortality. I think also, if you continue to see 50% elevation levels or 30%, 40%, 50% elevation levels of serum TTR, that would also correlate meaningfully, with a greater than 15% relative decrease in mortality.
Remember that when we looked at the difference in stabilization between 20 and 80, which is 37% to a little under 50%, you had a 20% relative risk reduction between the 20 and 80 mg groups. Phenomenal risk reduction between those two. I think you could expect to see something pretty meaningful if you were able to drop a six-minute walk by 8 m. Also, that's meaningful for patients. The ability to walk 10 m is a big deal for people who want to dance with your grandkids or want to get your mail or whatever. These are typically healthy people until they get ATTR-CM. It's a big deal in and of itself.
To summarize your expectations, I don't want to put words into your mouth, but it seems like with the acoramidis, assuming that the placebo declined as similarly as the tafamidis phase III trial, which is about 60 m, you would expect with acoramidis at 12 months, around 15, 17 , decline? Is that correct?
Yeah. I think in the teens is appropriate. Yeah. I think starting with the populations we're starting at, I think if we don't hit the teens, I'm not sure that we can say that we're better. Yeah.
Okay. In terms of a trial, patient profile seems to really similar, except for the fact that you may have a little more wild-type patients than the previous.
Yeah, which is natural given the evolution of how we're diagnosing the disease and also the countries where we ran trials.
When you look at your earlier data, the acoramidis is pretty potent against the variant. Do you think, having more little bit more wild-type patient might have some impact on the outcome? Is it too negligible?
We didn't see a significant difference in that in ATTRACT between the mutant and the wild-type populations. We're equipotent, so I feel pretty good about our chances in mutant and wild-type as well. It's not like we're not hitting wild type.
How about on the placebo side?
Well, I think my original point is people are similarly sick. I know that people generally think the hereditary form is sicker. That is not the case as far as we can tell in ATTRACT. I don't know. What were you saying, Cameron?
I was just going to say that the treatment effect size on both mortality and six-minute walk was almost identical in ATTRACT between the variant population and the wild-type population. That's our base case assumption for our study as well. Though in going, we did have some of the expectation you're alluding to, which is that you do see a greater increase in serum TTR levels for individuals on acoramidis and then for variant patients than those with wild type. That's because they start with a less stable form of TTR, and you raise everybody up to near complete stabilization. We think it could or should be possible to see a greater effect size there. That just isn't what was seen in ATTRACT. We don't have that expectation going in.
Okay. The one big difference that I see between your trial and Pfizer trial is really number of patients. If I calculate correctly, so if my math is correct, you have like over 40% greater number of patients in your trial. That's very overpowered. With that kind of a powering assumptions, it's highly unlikely that study is going to miss the endpoint because placebo performed a little bit better than whatever. Given that you have a larger patient, do you think that data points could be tighter so that that could actually help in terms of numerical numbers difference in your trial compared to tafamidis? Or you don't think that's going to impact much at all? You know what I'm saying?
The standard error, given that-
Plus or
-isn't a denominator, will be tighter.
But-
If you're sampling the same distribution but with more N, then yeah, I guess the standard error would be tighter.
Question is, the data points are tighter, so in a way that when you present the data in aggregate, in average, it could actually show better outcome. No?
No, because the standard errors would be tighter to your point, but the mean is the mean of the distribution that you're sampling. Ultimately, it's not going to take the number from 18. It's going to be 18 with a little bit more confidence, which is exactly why you just said the trial is powered better, too. That's all you would see, you would still see the same number. You would see the number that delivers. If tafamidis ran twice as number of patients, the theory of the case would be that they'd still hit the same means, just different standard error, right?
What was your reason behind increasing the number of patients by 40% compared to?
Well, we started with a target of 510 because it was COVID. We were afraid that we would have a higher discontinuity rate than ultimately we ended up having. As the trial enrolls like a hockey stick. At the very end, we brought on a number of new sites. It felt pretty unethical just to say, "Look, we're not going to let these patients on." Remember that most of the geographies that we were running in, by virtue of the fact that tafamidis was approved already, were geographies that didn't have access to tafamidis. Mostly in Europe. It didn't feel right to us to not allow those patients on the trial. We didn't think that there was much downside.
Also, just in terms of commercial seating, I think it's important for us to be running a trial that affects as many sites as possible.
Okay. Another thing that you said was that if Part A is successful, you would apply for approval in the U.S. and Europe by mid-2022, and that could lead to full approval. Question to you is that tafamidis was approved based on the clinical outcomes, like morbidity and mortality. From your discussions with the regulators that they are allowing you to file based on the six-minute walk distance. Can you elaborate on the discussions and why do you think regulators would allow you to file on six-minute walk distance, whereas they approved the tafamidis on clinical outcomes?
Well, the regulatory agencies have always been clear about what constitutes an approvable full endpoint, which is obviously making patients feel better or demonstrating that they die less on therapy, and six-minute walk is a tried-and-true full approvable endpoint. In this case, the agency actually worked with us to identify six-minute walk as a triable endpoint. Not unlike they must have with Alnylam, because Alnylam's obviously running a trial that only has six-minute walk as an endpoint, and they expect full approval off that. We're obviously moving on to all-cause death mortality data. Yeah, I think we basically had that conversation with the Cardio-Renal Division, and we've had that conversation in the Europe, and we're pretty confident that we'll be able to file on that. It's not like a biomarker, like oxalate reduction or something in another one of our trials. It's a true functional endpoint.
Okay. Is there any kind of like a six-minute walk distance data you would need to show? Or just a P value of less than 0.01?
Yeah. P value of less than.
Sufficient enough for-
01. Yeah.
Okay. All right. Do you expect a standard review, 12-month review at the FDA once you apply for approval?
Yep. We would expect that we would use our PRV as well.
Okay. All right. By the time drug gets approved, you would have a Part B data, which is expected.
Yeah. We really hustle it out. It'd be early 2023 where we launch, and we get the Part B data just shortly after that, a few months after. Yep.
In Part B, 12 months, the patients are on acoramidis versus a placebo. After 12 months, physicians can actually utilize the tafamidis.
Yep.
Right? Up to 30 months. Can you kind of comment on what % of patients after 12 months would have gotten tafamidis?
Yeah. That's a product of how quickly tafamidis gains access in the geographies where we're trialing, right? Tafamidis isn't accessible right now in the U.K. and in broad swaths of Europe. My assumption would be as they become accessible, we'd see a lot of drop-on. We obviously have a lot of patients right now that have gone past 12 months. Most of those patients aren't on tafamidis. Because either in the U.S., they accessed our trial because they couldn't afford tafamidis, or in Europe, they accessed our trial because tafamidis wasn't on market, and it's taking a little while for them to get on market. I don't know what the percent will be. Certainly, we designed the trial such that the effect of tafamidis on mortality and CVH would be minimal because you don't see a Kaplan-Meier curve separate up and until 15 months or so.
It's a minimal separation at 18 months. yeah. I can't comment on the %.
Okay. When physicians are allowed to utilize the tafamidis, can they actually add the tafamidis on top of acoramidis, or is it only for placebo patients?
No, this is a binding trial. Yeah, they can-
Oh, they can decide.
They can add it on top. Yeah.
All right. Do you think that addition will make a difference?
No.
From a chemical-.
Because we've shown biochemically that acoramidis displaces tafamidis. It's not like it's two separate binding sites and there would be some additive effect there. I don't think you would expect to see anything additive.
Okay. In Pfizer's efficacy trial, on tafamidis, the 30-month mortality rate was across the 30%. I think the placebo was about over 40%, about 43%. When you look at the risk reduction, which was about 30%. What do you think you would need to show for medical community, patient, or Wall Street to actually look at it, hey, numerically, the product is better in terms of controlling hospitalization, deaths, and things like that?
I think in that 25% range, 15% relative risk reduction, if I'm getting that math right, is where we'd want to be. I think that if we had that, again, connecting the dots to a better six-minute walk, connecting the dots to a better serum TTR, and a superior stabilization, that would be a resounding victory.
We want to get into commercial strategy. Obviously, you guys are developing the product, noting that it's a better drug than tafamidis. When the data comes out, if it's numerically clearly better and there is no question. What if it's kind of like similar, let's say 6-minute walk distance, it's better, but then somehow on the mortality, morbidity benefits, it wasn't really that much different. How would that change your commercial strategy?
Yeah, great question. I think that obviously we've designed the therapy to be a best-in-class therapy. If it's effectively a me-too or viewed by physicians as a me-too, it changes a couple elements of the strategy. I'd say first and foremost, it would change our access strategy, where we'd be very focused on talking to payers about what they'd need to see from us in terms of contracting, rebating, et cetera, to be a one of two, since we're coming later to the market. Probably change elements of the way that we do distribution.
Also it would if we continue to believe that we were superior, like in the case where you just said six-minute walk played out, but somehow it didn't play out in mortality, I think it would completely change phase IV strategy, where right now we're very focused on primary prevention and trying to understand how to move this therapy as early as possible since the mass action disease. I think we'd want to follow up with clinical studies to understand if there are subpopulations or areas where we have superiority, where a better stabilizer ultimately does deliver superior efficacy. Those would be the obvious places that it would change. I don't know, Cameron, if you'd add anything there.
No, I think that's pretty comprehensive.
In the U.S. and Europe, you are planning to launch on your own. Then in Japan, you have it with AstraZeneca. How about the rest of the world? What's your strategy there?
I think Latin America is a very attractive market. We are there in terms of clinical trials, and I think places like Brazil and Colombia offer to a company an interesting place to build. I think we could build there not only for this franchise, but others of our franchises. I would expect that we could do something maybe early with a distributor, but ultimately, hopefully build our own infrastructure there. China is an open question for us still. We continue to evaluate. First of all, it'd be interesting to see how tafamidis, if they get a clinical trial waiver, if they're able to come onto the market, what kind of the uptake is there. We'll be watching that with interest. I think that remains a very interesting marketplace for us. Beyond that, I think we would access distributors or local partners.
How about you? The acoramidis is twice daily whereas the tafamidis once daily. Does that kind of play into pricing strategy?
How so?
Well, obviously once a daily is a lot more convenient than twice daily. No?
Well, first of all, these are 72-year-old pretty sick patients that are used to taking pills morning and night. I don't think convenience once versus twice-daily pills is what's going to drive market share for a disease that's killing 42% of people over 30 months. I think it's ultimately going to be efficacy. If you're talking about the construct under which we're the same as tafamidis, maybe that's one thing that might drive tafamidis market share. Pricing will be a tightwire if we're the same, because as I just mentioned, you're going to have to rebate in the U.S., so you're going to have to keep the price high enough so you can do that. You're going to have to think about, obviously, something competitive in Europe so that you can get on.
tafamidis has been on the market over 2 years, and have you heard what the kind of compliance rate there?
Compliance rates are high from what we've seen, what we've been able to observe. Obviously, there's about 50% of drug being given away for free in the U.S. right now because of mostly the Medicare Part D dynamic. In Europe, it's almost equal selling to the U.S. with about, I don't know, four countries that they're approved in. So much more to come there. The best guess right now, it's at 27,000 diagnosed patients. If you just back out the number that are on paying drug, it is a high compliance rate to generate that $2+ billion revenue. Run rate.
Okay.
+$2 billion.
Yeah.
Yeah.
Diagnosed rate has been going up significantly since the launch of tafamidis, and now it's at about 27%, approaching 30%. If you actually look at the number of patients on the drug, I don't know, accounted for free drug, but it's less than 10%. A couple of questions. Do you think the diagnosis rate, obviously, you just want the continue to go up, but do you think you would reach the point of close to 100%? Once everybody's diagnosed, what's the kind of realistic expectation that 1% of diagnosed patients would actually end up taking the drug?
I don't think it's 10%. I don't think that number works. You may be misquoting that because you don't have the full. The SPs that Pfizer works with don't release their data. You wouldn't have the full number of scripts. It can't be 10%. Just backing into that'd be $550 or so million. It's obviously more than that. Either which way, basically what you're asking about is number of patients diagnosed versus number of people on drug. I think the way that it works is, let's put Europe aside for a moment. In the U.S., you'll have about 50% of patients that still get free drug unless there's Medicare Part D reform in one way or the other, which I think could come, to be honest.
As part of a lot of the bills that are being put forth right now is some sort of co-pay relief or coverage on Part D for the elderly. If it's still today, you got a lot of patients getting free drugs through foundations, and then you've got a lot of patients that are on paid drug. I think that the compliance rates are extraordinarily high. It's not like a PCSK9 type of experience right now because, again, it's a devastating disease. You're losing the ability to ambulate, and ultimately, you're dying fairly quickly. You're feeling it, and people have been filling their scripts. Obviously, there's co-pay relief on the commercial side, which is at least 50% of people who are paying for drug right now as well. In Europe, I think you're going to see a very high compliance rate because, again, devastating disease.
This is like cancer-style compliance rates because it's a cancer-style devastating disease. Yeah. I think that's how I would think about it.
Can you talk about what the kind of IP, intellectual property, position on acoramidis?
Yeah. acoramidis goes out to What's the date again, Cameron?
2035 for composition of matter. Then, of course, a whole bunch of additional filings for additional formulations and method of use, dosing, et cetera, that go beyond there.
Okay. The composition of matter patent goes out to 2035. Is that kind of like Do companies use a crystalline form, or is it a chemical structure? Like a two-dimensional chemical structure. Is that the original composition of matter patent, 2035?
Yep.
Yeah. I think.
That includes patent term extension, but yes.
Yeah. Getting to your point earlier, we're already working on a once-daily formulation. Actually, it's interesting. If we did once daily anyway, even with just AG10, we get superior stabilization to tafamidis. It's just we wanted to keep everyone at the very highest end of stabilization 24/7. Either which way, we're working on a once-daily formulation. That would obviously, to the extent that we're able to switch patients on that would be an extension. Then I think for TAF, the 2029 date, which is, I think, the most conservative date, that one would apply there. I think you could expect that they're going to be looking to extend the runway there just based on the 80 mg or the 61 mg free salt formulation or whatever it is, the meglumine thing.
When do you think we'll see the data for once-daily formulation?
Probably 3 years. 2-3 years. Yeah. I don't think there's any need to do anything right now, but we're certainly working on it.
Because you are coming into the market later, assuming that the data is numerically better than tafamidis, what do you think will be the kind of switching rate on patients on tafamidis to acoramidis? Obviously, you could capture a lot of new patients, but then by then, a lot more patients are going to be on tafamidis. Switching is important to drive sales faster. What do you think would be the kind of switching rate if you have a numerically better drug than you think you should?
It depends on a couple of things. Number 1, most obviously, is the delta there. If we're closer to halting the disease and tafamidis continues the progression, I think from a clinician standpoint, one will want to switch as many patients as quickly as possible. Obviously, though, the second key gating factor is how the access to the medicines work. It's already quite onerous to get onto tafamidis. Once you're successfully on tafamidis, the degree to which our patient hub services and our access backdrop is really easy to navigate will help dictate the number of switches that we get. I think if we have a really compelling difference in terms of clinical efficacy and we have something that's at or better than VYNDAMAX, which is pretty good, honestly, in terms of getting patients on drug, I think you could see significant switching.
Okay. You are also running clinical trial for polyneuropathy trial. Polyneuropathy is a much, much smaller market than cardiomyopathy. From our expert call a couple of weeks ago, what we've heard was that about 10%-15% of cardiomyopathy patients have the symptomatic neuropathy. He has found that tafamidis doesn't really work for the symptoms of a neuropathy. It's now approved in the U.S. for polyneuropathy. Can you talk about your rationale for involving neuropathy, and do you think that your main purpose of running the trial is to actually also reduce the neuropathy symptoms in cardiomyopathy patients, or you're purely looking into polyneuropathy patients opportunity there?
I think the evidence for a full stabilizer being most effective of all the agents in polyneuropathy is even more striking than it is in cardiomyopathy. If you go back just a couple of years, people didn't even really believe. There were a set of believers and non-believers about whether or not 20 versus 80 would separate. As you probably remember, after Eidos went public, when ATTRACT first read out and people didn't see that dose separation, people were like, "Well, I don't know. Every more stabilization is going to be better in cardiomyopathy." Whereas in polyneuropathy, it's very clear that tafamidis 20 wasn't approved in the U.S., as you rightly point out, approved in Europe. diflunisal, which is a 70% stabilizer, does better than tafamidis and actually has a marked clinical effect.
Diflunisal, which is a 70% stabilizer, has an equivalent effect as a 70% knockdown agent, inotersen, and then Alnylam's 84% partial knockdown has the most effective clinical outcome. A 95% stabilizer, by virtue of all of those data points, should be the superior agent in polyneuropathy, and we've got more clinical data to suggest that. It's worth pointing out that the T119M trans-allelic trans-suppressor occurs in the context of the V30M patients, which at least early onset, tend to be polyneuropathic. I think ultimately, there's a great deal of evidence to suggest that a full stabilizer should impact on polyneuropathy symptomatology, and the reason that tafamidis likely doesn't have an effect for those patients is its low amount of stabilization. We already know a good stabilizer has effect on PN symptomatologies. We know that from diflunisal.
The reason taf doesn't have it is taf's not a very effective stabilizer.
In your current Phase III trial for cardiomyopathy, are you also looking at kind of a neuropathy symptom improvement?
We're not getting a lot of them because, again, wild-type patients don't have it, and V122I patients by and large don't have it. We do have some mixed phenotype T60A folks, maybe there'll be a few neuropathy patients. We won't have a robust data set there.
Okay. I want to touch on competitive landscape. There are silencer, TTR silencers and/or stabilizers. Prosi is more on the second generation with vutrisiran. phase III is running, and there they have a combination with the tafamidis and tafamidis alone. Kind of your view of a silencer versus stabilizers. Our expert a couple of weeks ago suggested that when you get to near full stabilization, there is really not much difference between silencer versus stabilizer. When you get to that full stabilization, adding silencer wouldn't make a difference. I want to take your view on those points.
Yeah. I think let's start with, again, the molecular pathology of the disease, which is that 1 is trying to reduce the amount of toxic monomer as effectively as possible. TTR is falling apart to monomer, and that deposition of that and clumping together is what's driving the disease. I would make the point, as I just made earlier, that inotersen and diflunisal delivered about the same in terms of clinical efficacy. The same amount of stabilization is tantamount to the same amount of knockdown. There's not something superior weirdly happening with knockdown. It's just about limiting that downstream. If you have a 95%-100% small molecule stabilizer to meet that similar efficacy, I'd imagine you need a 95%-100% knockdown agent.
The knockdown agents that I know of best, which are Intellia's, about 92% or so, I think, and Alnylam's around 84% or so, or in the low 80s. I would expect, number 1, that a small molecule would outperform a partial knockdown agent like Alnylam would have on efficacy. Secondly, I think there are 2 design criteria that 1 is aiming for in this space. 1 is the precise reduction of toxic monomers. Most precise way to do it is to actually target the disease at its source and eliminate destabilization. It's not a disease of excess TTR, as you pointed out at the beginning. It's a disease of destabilized TTR. The second key design criteria is to keep the protein around, because the protein is doing important stuff, lots of which we don't know about, some of which we do know about.
Someone asked me the other day if the vutrisiran, if it's a once quarterly injection, is going to be more convenient. Remember, what a patient cares about is how many pills they're taking. You're still taking vitamin A pills every day, and if you forget to, you're driven to night blindness. You've also got 20% or so infusion AEs right now with patisiran, and you're eliminating a key core protein that's the 12th most abundant protein in the human body that no one doesn't have, and no one has a half dose of, and no species doesn't have, and no species has a half dose of. If you can be more precise, target the disease at its source, deliver better efficacy in a more safe way, that's how I stack up the competitive landscape.
What do you think of genetic, you mentioned the Intellia. Before it ever gets to the market, how do you think that could actually change how patients as well as physicians are looking at the disease?
Yeah. Look, I thought the data is super exciting. Obviously, the data is super exciting, and I think that the 92% knockdown feels good to me to get into that super potent point. There's a whole lot of stuff we still need to figure out in terms of safety of those approaches, and for how long is the agency going to need to see data, and how long are physicians and patients going to need to see data to really understand that if I could get the same level of efficacy as I can with a small molecule, then I'm going to sit around and do something like that where you've got chromosomal explosions, you've got potentially untoward integration events. How much do we need to see until we get comfortable with that for the same efficacy as a small molecule?
I personally am not convinced that the equipoise favors that, if indeed a small molecule can deliver the same or better efficacy. It all comes down to efficacy here. I think if we're not able to provide efficacy tantamount to what knockdown agents and gene editing can, I think that's another story. I don't believe that to be the case based on what my understanding of the disease mechanism is.
Let's just say acoramidis data comes out the way you expected, better than tafamidis, and then silencers get onto the market as well. Although is pretty far out. Aside from gene editing, when you have those silencers and the stabilizers, how do you think silencers would be utilized? Do you think it's going to be utilized the patients who are not responding well to your drug, or it's going to be just added to tafamidis to have a better TTR, not necessarily stabilization and getting rid of the toxic protein. How do you think its market is going to be positioning for different drugs on the market, aside from gene editing part?
In my view, because the key piece of data that you left out of that hypothetical is the relative efficacy of us versus the knockdown. In my view, if we have the same efficacy as the knockdowns do, and then they come to market, so they've got a four times more expensive agent that's less safe with the same efficacy as ours, I believe that one would likely use our drug as a first-line treatment paradigm, and then ultimately, not every drug works for everyone. If there are people that progress, for instance, or have very severe disease, perhaps they would use a knockdown. Perhaps they would use on top of tafamidis, to your point.
I think then obviously, even though it's 4 times the cost, there are going to be a number of providers that use it because you can buy and bill it, and if you're a 340B hospital on Medicare, you're buying it at a 33% discount and you're buying and billing it at ASP plus 5 or whatever. There'll be an economic rationale to use the more expensive agent in the U.S. I think that those would be the cases under which it would be used if indeed we all meet at the same efficacy.
In your view, to summarize, you think that with acoramidis, like delivering close to for TTR stabilization, you think the efficacy would be at least similar to RNA silencers, or potentially could it be better?
Yep.
Okay. What if the silencers come out better efficacy? Probably at that point, once every quarter injection wouldn't really matter to the medical community, correct?
I think if they turn out to be better on efficacy, as I just mentioned, I think this market is going to be efficacy-driven. I do think that at that price point, there'll still be some market share that a small molecule takes. Depends on how big the gap is on efficacy. In your hypothetical, which I don't think will manifest, but let's just go with it for a second, that silencers are more efficacious than small molecules. I think silencers would take a majority of the market share. That there would still be cardiologists that are more sort of familiar and comfortable using small molecule agents. I think that at the price points, they would probably be used ex-US quite a bit. We'd have to see. Yeah.
It would be a safety rationale, and it comes down to the equipoise in and around the separation in efficacy. If it's a small separation in efficacy, you're eliminating TTR, you've got safety issues, you've got the imbalance of CV effects as you saw with vutrisiran and polyneuropathy, potentially small molecules take quite a bit of market share. If it's a whomping difference in efficacy, I think you start to believe that ONPATTRO will take quite a bit more market share.
But-
If that's true, then I think Intellia would take even more.
Yeah. The silencers, obviously it is expensive, but there is a possibility that once cardiomyopathy indication gets approved, they may lower the price, correct? That is a possibility.
Could be. Yep.
Right.
Sure, of course. Yeah.
Okay. For remaining couple of minutes, I want to talk to you about the pipeline. At our R&D Day, we talked about certain number of pipelines, but you didn't really talk about infigratinib and achondroplasia. Just want to ask you on that. The phase II data is coming out in first half of next year. Can you talk about when you present the data, how many patients, what we are expecting to see from the phase II in the first half of next year?
Yeah. Basically, just as a reminder, what we're doing is we're interrogating our FGFR inhibitor in 4 different cohorts of patients of about 11 patients each, a total of 44 patients or so. What we're looking for is average growth velocities that are hopefully superior to what BioMarin observed with VOXZOGO. We're in the third cohort right now of interrogating it. The reason I guess we didn't get into it was there's not much new to say about it versus last year, and we had a lot of programs on the docket. Ultimately, the presentation will be, 1, safety, which I think is the big unknown for our FGFR inhibitor is, low-dose FGFR inhibition in the pediatric context has never really been tried. That'll be 1, and the second will be efficacy, AGV, how does it compare to VOXZOGO ultimately for these patients?
I think that if we have at or more efficacy, a daily oral is going to be vastly superior to a daily injection. It's got to be super safe.
Yeah. Basically, you are using infigratinib, you are using, if I remember correctly, like a one-third of a dose in cancer. It's pretty low.
If you have to guess about a potential side effect, what kind of AEs would you expect with oral drug?
Well, what we saw in juvenile tox is long bone growth. We saw an exacerbation of the therapeutic effect, let me put it that way. That's why we have a cutoff at a certain AGV, where we don't want to go above it. We don't want to go above the 90, I think it's 90th percentile AGV. We don't want to get kids growing too much. That's the juvy tox that we observed. I think, given the PK profile of the drug, there would be a chance that we hit hyperphosphatemia. In my best estimate, hyperphosphatemia at those lower values, obviously, it's just a lab value and fully reversible.
I think it would be interesting to dose in fine increments all the way up to where we're seeing MTD effectively in children, which would be, again, we won't hurt them with reversible lab values. We'll ultimately be able to assess what the maximal efficacy possible is in this disease, because I think this should drive quite a bit of efficacy. FGFR3 inhibition should be better than the downstream CNPs that we've been using so far. You're hitting both pathways. You know FGFR inhibition or inhibition of at least one of those two pathways leads to efficacy. I don't know why the two pathways would lead to less efficacy. It should lead to more efficacy. It's just a question of how much efficacy can you juice or wring out of inhibiting that receptor. That's my best guess as to how it plays out.
You have been running the trial for a while. Would you say that your confidence levels for your chondrodysplasia program, would you say it's as optimistic as before, better compared to before you started the trial?
Oh, well, it's way better than before I started the trial because I had no clue what FGFR inhibition would do. We had juvy tox in an animal model, but it's not like in the ATTR case, where you're going in with a precedent mechanism to the same patient population, you have a better idea of what the safety profile looks like. You never freaking know what's going to happen when you move into a new population. Thankfully, for the kiddos that we're trying to serve here, through the first 2 cohorts, we had a benign safety profile. I feel better about it for sure.
Okay. Our time is up, Neil, could you actually talk about a couple of key catalysts that we haven't talked about that you are excited about in at least the first half of next year or next year?
Well, I think-
Yeah.
Yeah. First and foremost, I think about that, the wave of 4 important programs, and I think within that, we discussed acoramidis and ATTR, but the one we didn't discuss is ADH1, where we had 100% responder rate now across 2 parts of the clinical trial. I think the key catalyst there is going to be the meeting with the FDA and the kickoff of the phase III, because I think that for that trial, you got to revise up your POS to a pretty high degree. We're the only player in town, and with 100% response rates, I think we've got a really interesting drug to serve patients there. That would be 1 catalyst to be looking for. The second catalyst to look for, I think, would be proof of concept in our dystrophic epidermolysis bullosa trial.
I think you and I discussed a fairly risky approach with a huge protein that you're putting back into people's body. I love the phase I/II data where we showed dose-dependent increases in C7, so we'll see if that correlates into something that's meaningful for patients. That phase II POC should read out sometime in January. I think the final one, I don't know if we'll hit it in the first half or second half, is our LGMD2I trial, where we've basically got an elegant substrate replacement therapy approach. Within that, we'll see whether or not we can actually elevate the amount of glycosylation of that alpha-dystroglycan complex, which I think could be and lead to an accelerated approval. If you look at what happened in DMD, that's 7,000 to 10,000 patients. That's a big unmet need.
Those are 3 things that are on my mind, in addition, obviously, to ATTR and achondroplasia. I don't know, Cameron, would you add 1?
A lot more. We got the PH1 as well. We got SHP2 data. Got a development candidate for KRAS. I think all of those things should be material as well.
Since our time is up, thank you very much. Neil and Cameron, thank you for your time and the discussion today, and best of luck on the phase III data coming out at the end of next year.
Thanks, Eun Yang. Appreciate it. Yeah, we're excited. Thanks.
Thank you.