Good morning, everyone. Thanks for joining us for the fireside chat for BridgeBio. I'm joined here by Neil Kumar, who is the CEO and Founder of the company. Just before we get started, a disclosure. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/research/disclosures. If you have any questions, please contact your Morgan Stanley sales representative. Why don't we get started? Neil, maybe just for people that don't know the company as well, maybe give a you know few minute overview of the company, what you're up to, maybe the philosophy and maybe we'll start there.
Yeah, sure. Well, 1st of all, thanks so much for the opportunity to sign here, and thanks to the entire Morgan Stanley team. BridgeBio, for those of you that don't know, is a company that was founded about six and a half years ago to primarily target well-described genetic diseases at their source. That means both Mendelian or monogenic diseases, as well as somatic cancers with clear genetic drivers. Over the course of those six and a half years, we've developed a late stage pipeline that has five programs that'll read out its phase II or phase III over the course of the next 18 months. Then really a barbell shaped early discovery stage research program that's highlighted by our RAS franchise that'll head into the clinic sometime in the next 12 months or so.
Over the course of the last 12 months, for a variety of different reasons that I'm sure we'll get into, we've done some pipeline pruning, we've done some cost pruning, and we've really focused in on what we think are large market drivers. When I say large market drivers, you can think about our late stage pipeline as delivering something like $9 billion in peak year sales in non-risk adjusted peak year sales, by 2030, and about $4 billion on a risk adjusted basis. That's a pretty late stage pipeline in some pretty large markets. It's really driven by phase IIIs that we'll get into in three large markets that'll read out in the next 12 months or so.
One is an ATTR cardiomyopathy, the other is an ADH1, which is a subset of hyperparathyroidism, and the other is in limb-girdle muscular dystrophy type 2I. Those are three phase IIIs that'll read out by the end of next year, early the year after, and then two big phase IIs in congenital adrenal hyperplasia and achondroplasia. Yeah, that's a lot of word salad for those of you that aren't in monogenic disease, but I think the important thing as we'll go through is those are all best in class or potentially 1st in class, molecules that target really well-described diseases at their source. The therapeutic hypotheses are simple, and we think the probability of technical success is reasonable. Yeah, that's the company at the-
Great.
At the high level. I'm happy to dive in, to any of the specific programs or...
Yeah, let's go into some of these specific programs. Maybe we'll start with ATTR, obviously. Do you wanna just talk about why you think this could be a best in class molecule, as well as stabilizers versus silencers and how that kind of gets wound into this? Then maybe we'll go through some of the data and what we're expecting from it after that. Why don't we start with the 1st part of that?
Yeah, sure. This is an ever evolving field. ATTR cardiomyopathy, for those of you that don't know, is one of the largest genetic-oriented marketplaces out there. Some 13% of heart failure with preserved ejection fraction patients actually have ATTR cardiomyopathy. There's a commercial drug in the space called tafamidis, which is a multi-billion dollar franchise. There are really two design principles associated with our efforts within this space. Design principle number one is that every time you do better in managing what's toxic here, which is a monomer that deposits in your heart, you do better for patients, and that's been shown across several different clinical trials, including some recent data that came out.
I don't think we've seen anything over the course of the last couple years that changes our understanding of the molecular pathophysiology of this disease. Then Secondly, that we wanted to, using a small molecule approach, preserve an important protein called transthyretin, which is really the driver of the disease. I think, again, there's nothing in the datasets that thus far suggest anything other than the fact that this could be a long-term and interesting protein to keep around. We have a program in phase III that's a 95% stabilizer as compared to the competing products, which is around 50% stabilizer or knockdowns, which you could think about knockdown efficiency and small molecule stabilization as tantamount to each other.
Knockdown efficiency is around 86%-87% for the next nearest competitor. What we're reading out is against the hard outcomes of mortality and CVH sometime early next year. There's been some interim reads in the context of six minute walk that I think have been confusing in the context of how they pertain to mortality. I think there's been some big wins for patients, which we can get into. The best in class principle here is that we believe that with the highest degree of monomer elimination or stabilization, we should be able to provide the best mortality data, which in turn we think is gonna win the day commercially.
Right. Let's spend a 2nd on the data because.
Yeah.
I think there is a lot of, you know, people are asking questions about it. Maybe describe what we saw in December, what that means, and then why we, what we think the mortality, next year, mid next year, you know, could still be, could be a significant benefit for patients. Yeah.
Yeah. Just as a reminder, our phase III in ATTR cardiomyopathy is effectively a nested trial design where we had 12 months and then a readout against a six minute walk distance, and then another 18 months for a total of 30 months where we read out the final and most important endpoint of mortality in a hierarchical analysis with CVH as well. What happened in that interim read at the end of December, as most of you know, is unfortunately the drug did not show a statistically significant difference as compared to placebo. It was interesting because we saw no placebo decline whatsoever. On treatment, the drug did about approximately what a healthy volunteer would do.
That is also what happened with the placebo, and we can talk about reasons for why that might have been. Against everything else, Kansas City Questionnaire, which is the quality of life questionnaire. Against BNP, which is perhaps the most profound univariate predictor of mortality.
Mm-hmm.
Everywhere else we could see any sort of decline, the six minute walk, for instance, in the variant population, we saw a statistically significant impact of the drug. Biochemically, we saw high levels of stabilization and increases in serum TTR that were beyond our expectations. The drug seemed to be safe, biochemically active, doing what it could do, whenever there was a deteriorating baseline. Unfortunately, against the six minute walk, we saw no deterioration in that baseline. Our whole goal over the subsequent couple of months was to understand whether or not there was actually a baseline of deterioration on mortality against which the drug could show something. Just as a reminder, the mean age in these trials tends to be late 70s. For us, it's 77. The death per annum is something like 1.5%.
Right.
You know, the blinded death rates as of January, February in our trial were 13%, 14%.
Oh, wow.
They've crept up to 18% now with some six months to go left in the trial. Long story short, we believe there is a baseline of deterioration, and if our drug is active, we should have enough mortality events in the trial to show a statistically significant impact. We feel good but confident in that endpoint. We'll just have to see in 1st half next year.
Great. Maybe just back to that placebo response for a 2nd. When you unpack that, or is there a reason why you think that might have happened or?
Yeah. The plot thickens a bit.
Yeah
with, you know, obviously Alnylam's wonderful patient's data that was delivered, I guess, you know, a couple weeks ago now. I guess the theory of the case is that we had a healthier population, and that's true as compared to ATTR-ACT. I think every trial that's gonna run now has a healthier population. We didn't have a healthier population as compared to Alnylam's clinical trial population. We had more Class Three patients, for instance. Our mean and median BNPs were higher. Our death rates on placebo were higher, and our quality of life questionnaire declined.
Right
more meaningfully over 12 months. It's not immediately obvious that we're capturing the right descriptors of severity at baseline. The thing they had much more of was double the number of variants that we had.
Right.
They had a category of so called Tafamidis progressors, where they were on top of Tafamidis, and that might have made some sort of a difference. Again, until we see those subpopulation analyses, it's hard for me to exactly understand why their placebo went down more meaningfully...
Right
...than ours. Again, I just, when I think about mortality, I think about the fact that Class Three tends to do worse than Class Two, that the higher the BNP and the more significant the change from baseline of BNP over the 1st 12 months, the greater the rates of mortality.
Right.
This has been well-published, I think, since our study and in natural history studies prior to our study. The six minute walk, there could have been a training bias, a context bias. There could have been something around the way we baselined, that was a little different than Alnylam, but hard to tell right now still.
Great. Okay. Thank you. Why don't we move to onto achondroplasia, and maybe talk just to orient people on the size of the market, and, you know, who can get treated and maybe some of those types of things.
Yeah. Achondroplasia, for those of you that don't know, is the most common form of dwarfism. It's a very large unmet need in the area of skeletal dysplasias, and it's probably the simplest of the diseases we go after in the sense that there's one mutation in a receptor called FGFR3, gain-of-function mutation that drives the disease. Our therapeutic approach is simple. Again, we're inhibiting that FGFR3 receptor with a small molecule. The marketplace itself is described as some 55,000 people affected-
Mm-hmm
By the condition. There's maybe, you know, 10,000 or slightly fewer kiddos that would actually be available for treatment because you probably can't treat after the growth plate closes.
Okay.
Still altogether, that's a multi-billion dollar market. I think what you've seen from our commercial competitor, BioMarin, is a pretty healthy launch. That drug is tracking to be, you know, well over a billion dollar franchise for BioMarin. I think there has been some dialogue around the importance of treatment and how early and how parents might think about it. Remember that 80% of these mutations are de novo mutations.
Right.
Most parents of normal stature are faced with this condition and thinking, "Well, what's the best thing for my child?" Certainly it's not just stature that's a big issue here. It's proportionality. It's spinal stenosis. It's infection. It's all of those things. I think people are keen for solutions, and that's why you've seen a healthy launch for Voxzogo.
Great. When you're giving you know therapeutics to kids, can you talk for in this condition?
Yeah
oral versus, you know, other forms of delivery and what that means for a parent and compliance, et cetera?
Yeah. Look, our belief, 1st and foremost is that efficacy and safety are gonna rule the day here. But that ROA, especially for any condition like this, you know, route of administration is gonna be a very large deal, you know, for any of you who are parents, injecting your child every day from the moment of diagnosis to the time the growth plate closes is quite onerous. We have a sachet formulation that allows one to mix the drug in with food, and we demonstrated in the interim readout in our phase II that it's also a very safe drug.
Right.
I think that's gonna be very, very appealing to physicians. I think already actually, if you go out and survey physicians, some 70%-80% would prefer an oral of similar efficacy to even what you see with BioMarin. On the efficacy side, it stands to reason that directly targeting FGFR3 should lead to better efficacy than what we've seen with the CNPs, both the Ascendis approach and BioMarin, 'cause they're getting only one of two of the effector pathways. The number that we delivered was 1.52 in terms of average growth velocity or annualized growth velocity. That was slightly higher than what BioMarin has shown against baseline, which is 1.35. Look, I think AGV, like six minute walk distance, is a pretty variable metric, so I'm not out there saying we're definitively better than.
Right.
Certainly the efficacy is at least the same as Voxzogo and with a safe daily oral versus a daily injectable. We've got a cohort reading out in late June, that's one of those big phase II studies, or late January, sorry, of next year, that is at double the dose of our 4th cohort. We continue to expect that we should have best-in-class efficacy, benign safety profile, and with the once daily oral, we'd imagine that this would take the majority of the market.
Right. Let's go to that data that you released just a few weeks ago.
Yeah.
You had a few different dose cohorts, so maybe describe.
Yeah
how you started, you know, where you ended up seeing efficacy, and then why you added the fifth cohort?
Yeah. You know, this drug, the FGFR3 inhibitor, it's actually an FGFR1, FGFR 2, and FGFR 3 inhibitor that we have is used at very high dose in oncology. Whenever you have a drug that's used at high dose in oncology, regulators, patients, physicians are concerned 1st and foremost about safety. You can see that in the neurofibromatosis story with MEK inhibitors as well. We started at extraordinarily low doses, about 150th of our NOAEL, the juvenile tox model that was wild type. Effectively homeopathic dosing. We doubled the dose just to make sure, since we had never inhibited FGFR in the context of kiddos before, that this would be safe. The obvious safety signal that we were looking out for was hyperphosphatemia, which is driven by FGFR1 inhibition.
We were also looking for things that, you know, the sort of unknowns as well. The 1st three doses were well under what you would expect in terms of the AUC that drives efficacy.
Right
... in mouse models, which have been fairly faithful in this condition to date. The 4th actually got us right in the range, maybe about 15% lower than where we would like to have had equivalent efficacy to BioMarin and in the mouse model. Again, you know, doubling the dose from there should put us firmly in where the dose looked like we were getting, you know, three to fourfold increases against BioMarin and Ascendis in terms of long bone growth and some effect on things like L4, L6.
Right
spinal stenosis in the mouse.
Great. Okay. Why don't we switch for a 2nd to oncology, and perhaps just describe for the audience your precision oncology strategy and your team and, you know, what the things you have in your portfolios, and then we can dive into some of the specific programs.
Yeah. We've been active in precision oncology from the get-go. One of our founders is Frank McCormick, who founded Onyx and was a close collaborator of ours at Third Rock when we were all there. Primarily our efforts have been focused on the MAPK signaling pathway and primarily within that, within RAS. Our leaders there are Eli Wallace, who many of you know was the CSO at Peloton, prior to that at Array for a long time, and then a wonderful group he's assembled, including Pedro Beltran, who did a lot of important work at Amgen in the PI3K space and others.
It's a wonderful group of scientists that have been knocking away at various approaches to RAS and SHP2 over the course of the last six years. Our SHP2 compound, which we believe has the potential to be best in class, mostly from a safety profile side, is partnered with BMS-
Right
is entering a series of phase IIs in combination with importantly the KRAS G12C GTP inhibitors, but a few others. I think, you know, something that people don't really pay a lot of attention to, but I think is important, is we believe we have a best in class direct inhibitor of G12C GTP-bound KRAS, which we can talk about in a moment. That should head into the clinic in less than 12 months. There's some important reasons to believe that what we've seen thus far with the GTP-bound class, which includes Amgen's drug and includes Mirati's drug, is wonderful, but can be further improved on, again, with a much more potent inhibitor and one that takes out both the inactive and active forms of that. Right behind that, we have a pan-KRAS approach.
Mm-hmm
a PI3K breaker approach as well, both of which should be in the clinic in the next 24 months.
Right. Actually, let's just go to the G12C, because obviously a lot of people are paying attention to that. What kind of evidence do we have right now for some of the potency and potential differentiation versus the agents that, you know, a lot of people are focused on right now, given where it's at?
Yeah. Maybe I'll focus on.
Yeah
Three different areas there.
Yeah.
I mean, one is the hypothesis, the 2nd is the preclinical data, and the 3rd is the clinical data that we have seen to date. The hypothesis is that in general, when one is inhibiting KRAS, there's a couple of different drivers that would make it more potent to inhibit the GTP-bound form. Number one, there's a wide variety of resistance mutations that occur upstream associated with RTKs or upstream MAPK signaling that are more potently shut down in the context of a GTP-bound inhibitor. In fact, if there is a GTP-bound population and you inhibit the GDP-bound population, there's still gonna be leakage to that GTP population, and you can see that both clinically as well as preclinically.
You can see that GTP inhibitors, ours and Revolution Medicines, completely abrogate that signaling and shut it down and drive cell death where Amgen and Mirati Therapeutics' compounds would not. That's number one. Number two is there are several important research papers published in the last 12 months that suggest that GTP-bound synthesis is one of the ways a cell reacts to GDP inhibition, and that's just basically targeting one of the resistance pathways at its source. There is also resistance mutations that occur in the binding pocket. This is a different binding pocket that the drug is active against, and we've published data on that. The 3rd and final piece is that GTP is obviously at much higher concentrations than GDP, and inside is also millimolar concentration.
You'll have active re-engagement as GDP comes unbound and GTP becomes rebound of KRAS, and you would have labeling only in the context of a drug like ours or RevMed's. I think going to the preclinical data, what's most profound is you can see that the responses are deeper and more potent with a drug like ours. Just to give you some sense, it's about tenfold more potent than Amgen's drug, close to 500 times more potent when you look at the KI over Kinact. If you look at the IC50 around phospho work, you're getting down to almost full inhibition within 20 minutes or so with the drug. It's a quicker deeper response in the cellular model.
In the PDX models, I think importantly what we've seen is in areas where the Amgen compound is only able to ameliorate but not fully regress tumors, we're getting 100% tumor regression. That owes in a bit to again that deeper more potent challenge, and I think we've put out some of that data with our Q, and we'll have a fulsome review of that data in the October KRAS conference upcoming October 17th or so. That's the preclinical. Then on the clinical side, I think every time we do a little bit better in terms of potency, we seem to be doing better for patients. And-
Right.
The recent Genentech data I think showed us that, you know, more potent GDP inhibitor alone can get you better and better ORRs. We'd expect, given the fact that RAS is a clear driver in many of the conditions we're going after, that the ORRs that we've seen to date can be surpassed, and you've seen that in other targeted areas. I'd sort of think about this like a, you know, like an osimertinib-like play, where we're really trying to be best in class, obviously not 1st in class. We'll just have to see how that plays out in the clinic here.
Yeah. Maybe just walk us through the timelines for, you mentioned October 17th for some of the preclinical, but just maybe how the rest of the program will unfold there.
Yeah. The program right now is in IND-enabling studies. I think October will be the right time for us to kind of unwrap the development candidate package. I'd expect it'd be in the clinic somewhere nine months from now, or so. You're probably looking at a trial with, you know, sort of a 50/50 split of naive patients and refractory patients. You know, another 18 months or so before you get real data out of that in terms of whether or not you have a differentiated ORR-
Right.
Against enough patients that you can hang your hat on it.
Great. Okay. Maybe just, for oncology, anything else in the precision oncology portfolio that you wanna just mention?
Yeah. Behind the G12C GTP-bound program, there is, as I mentioned, a pan-KRAS program that's about six months from development candidate. I think importantly, we have a very novel approach that stems from our research in RAS, which is a PI3K breaker approach. As you probably know, RAS and PI3Kα come together as a part of the effector pathway in many cancers. Instead of directly inhibiting either on a mutant-selective basis or on a just a pan basis PI3Kα, which has its issues, what we've demonstrated is the ability to actually inhibit that interface between RAS and PI3Kα. That compound is pretty close to development candidate. There's a wide variety of uses there, both in RAS-driven cancers as well as outside of that in PI3K-driven cancers. That's a very exciting compound that we hope to show more on sometime late this year in terms of the development candidate package.
Great. Okay. Let's move to ADH1, and maybe, as we did for [guess], maybe an overview of the condition and the total addressable market, I think, would be helpful for people to hear about.
Yeah, sure. This disease I think is one of the more overlooked diseases in our portfolio because there really is no competition here. That's a good thing I think, commercially, but perhaps a harder thing in terms of education. This is a disease that's quite prevalent for our portfolio, at least. It's 10,000-12,000 patients affected. Some 3,500 patients alone that are diagnosed in the U.S. today. The unmet need stems from the pathophysiology of the disease, which is effectively hyperactivating mutations in what's called a calcium-sensing receptor, which you can think of as a thermostat for calcium.
Mm-hmm.
In your body. What these hyperactivating mutations result in are low serum calcium levels and high urine calcium levels, all of which directly result in the diverse array of symptomatology that one sees here, both in terms of muscle cramps, tetany, seizures in case of low serum calcium, and nephrolithiasis and downstream kidney disease in terms of high urine calcium. It's a bad disease. It's a well-described disease. Some 70-odd hyperactivating mutations have been described to date in the CASR. Once again, a very simple therapeutic approach. We're taking small molecule that's an allosteric inhibitor.
Mm-hmm
Of the hyperactivating mutations. What we demonstrated in our phase II results that we released at ENDO, now four months ago.
Mm-hmm
is an extraordinarily high responder rate, above 80%, in terms of urine and serum calcium levels. You have about a 2% responder rate on standard of care, which is vitamin D and calcium. Obviously, calcium is gonna exacerbate urine calcium levels, so there's nothing obvious that allows one to go after this disease, right now. A pretty poor standard of care, very high responder rate. I think the really good news on this one is that the endpoints are basically response rate against serum and urine calcium.
Mm-hmm.
That phase III reads out towards the end of next year. If I put down all that together, it's a, I think, a high probability of technical success, 'cause the endpoints are understood, that the mean-
Right
... is greater than the standard deviation. It's not measuring growth or six minute walk or things that can have a higher standard deviation. I think it's a large market with no other players.
Right
that we can see so far. That's the kind of an overview.
Great. That's great. Why don't we go to the muscular dystrophy program.
Yeah
just maybe give an overview of that program. We can go there.
Yeah, sure. This is the final, well, I guess we haven't done CAH yet, but this is the 4th of our main value drivers that should be reading out its phase III sometime 1st half of 2024. Limb-girdle muscular dystrophy type 2I is one of the largest, it is the largest of the limb-girdle muscular dystrophies. LGMDs, I think many people are familiar with based on Sarepta's great work in 2B, 2D, and 2E. This one's uniformly driven by loss of function in an enzyme called FKRP, which is responsible for the glycosylation of a complex called the alpha-dystroglycan complex. We're taking a very, very simple approach once again, which is we're providing substrate back to the enzyme that's not actually fully loss of function, but has a lower productivity.
With a substrate replacement therapy, we should be able to get back to normal levels of glycosylation of that alpha-dystroglycan complex. The disease presents like muscular dystrophy, typically diagnosed sometimes in adolescence, and leads to progressive loss of ambulation, as well as in about a quarter of patients, cardiovascular manifestations as well. It's a devastating disease. What we showed in our phase II data that we read out, I think in March of this year, was that we were able to increase the glycosylation of the alpha-dystroglycan complex some 40%. The goal was about 10% based on genotype, phenotype, and mouse models, so we're well within the range of where we want it to be in terms of therapeutic efficacy. We were able to reduce creatine kinase levels markedly.
Mm-hmm
...over 70%. The good, I mean, that's a noisy endpoint, but the good news is if you put increases in alpha-dystroglycan glycosylation against decreases in CK levels, it was very highly correlated.
Right.
It seems like something real was happening there. Then, you know, with all the appropriate caveats here, we had a natural history run-in on 10 m walk time, as well as modified North Star. You can see against both of those, you would expect to see a decline for patients, and you actually saw an improvement, a slight improvement.
Right
on drug. That's really where we're at now, is talking to the agency about trying to do an accelerated approach, looking at just alpha-DG levels over the course of 12 to 12 months or so, and then going out for a longer period of time against the modified North Star test.
I think we're actually out of time. Neil, thanks for joining us today, and thanks for being here at the conference.
Yeah, thanks for having me. Appreciate it.
Great. Thanks.