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

Hello, everyone. Welcome to the morning session of Day 3 of the Bank of America Virtual Vegas Healthcare Conference. I'm Greg Harrison, U.S. biopharma analyst here at BofA. And I'm very excited that we have BridgeBio here with us today. My colleague, Jason Zemansky, is on the line as well. And here to speak on behalf of BridgeBio is CEO, Neil Kumar. The format is fireside chat. So feel free to send me questions, and I can ask them anonymously. So Neil, let's jump right into the Q&A

Maybe you could just give us kind of an overview of your approach to development? And how your platform can potentially offer a higher probability of success in your programs?

Yes. Sure, happy to. Let me start by thanking you and the entire BAML team for the opportunity to present. Very much appreciated. Yes, this is really the core of how we think about one of our competitive advantages, which is the identification of the right target and the right intervention. As you all know, great teams with bad drugs don't tend to do much, and I think a great drug can bail out a lot of mistakes. So we really take our time in choosing the exact right combination of mechanism of action and mechanism of disease. And as you think about our approach to development, I would say, discovery and development. Remember, most of what we bring in to our company is effectively research that emanates from academic laboratories, often years away from the clinic. And sometimes, it could just be an insight into a disease and we'll start screening against it, or it could be a very early hit-to-lead chemistry, which we'll start working on. So firmly in the discovery camp. The way we tend to try to find these things as we have married what we consider kind of a bottom-up mapping of the 7,000-or-so monogenic disease landscape, see which diseases are truly well described without a lot of things that could throw drug discovery off, things like modifiers or pleiotropy, things of that nature. And then we put that together with kind of a top-down understanding of great academics, great institutions, where a lot of the great literature is being published and staying on top of grants and going to all the right conferences, things of that nature. And fundamentally, what we're trying to find are very well-described diseases, where one can connect the dots from genotype to molecular pathophysiology to symptomatology, and where we have the chance of targeting the disease at its source, typically with some precedented modality like a small molecule or gene therapy, something that has gotten approved in the past through other people's hands so that we understand the regulatory path. And I think in doing all of this, effectively, how we increase the probability of technical success, which I think is, dramatically increased, not just by us but by other people who play in fields like this, is that you're reducing target risk. And you still have idiosyncratic tox risk, you still have CMC risk, things of that nature. But you've run the human clinical experiment, the genetic experiment, if you will, in the human condition, and so one really can understand precisely which target is important, which target quantitatively drives the disease and what one might be able to do in the context of addressing that target at different quantitative levels through symptomatology. So that's where I think we're in good shape when it comes to probability of technical success even at the outset of a very early-stage drug discovery program.

Okay. Great. That's helpful. And then maybe if you could talk a little bit about how you view doing partnerships and you kind of have a centralized portion of the company and then a more decentralized portion for each of your programs. So maybe explain that? And what sort of advantage that gives you.

Yes, sure. So I mean, it's -- the decentralized nature of our company was the only way we could think of to marry what we see as kind of 2 key ingredients for efficiency in this space. Number one is, focus the model of each disease or each asset. So I like teams that are very much expert in a specific space, and the right experts for, let's say, an inherited neurologic disease or inherited dermatologic disease are not going to be the right experts for a germline cancer or something else. And so you want a group of individuals that are very much focused in a disease space, probably even more critical, honestly, in the disease areas we work in because they're so specialized. The person who really understands the assays for pantothenate kinase deficiency are likely to not be portable over to a set of assays that are associated with nerve degenerative syndrome and vice versa. So we like that focus, and we want people really incented to create a drug on the condition or in the condition that they're working in and they love and they're passionate about. But we wanted to marry that focus with a backbone, if you will, or a platform that could make their work a lot more efficient, than if they were to start a company every time they had an idea. And so we take what we think -- the generic R&D activities like Med Chem, CMC, all of the G&A that you could think of, and we put that centrally, so you can draw from it, when you want to, and then give it back, when you're done. And that final piece, what I just referenced is also really key because, in allowing people to take what they need when they need it, and then give it back or to scale up or down over time, we're variabilizing a lot of the fixed costs that would otherwise, be associated with programs like that, if they weren't stand-alone entities. And we're also enabling them to move very quickly. And that is really, really core to our value hypothesis because for many of these diseases that we go after, the peak year sales associated with the drug, maybe $300 million, $400 million or $500 million is not going to be huge drugs. And so you're talking, at the very riskiest points of the drug discovery effort, about trying to be very lean in terms of your cost. We say, we try to get to IND in $10 million or less, and we've done that, I think, with all the programs we put into the clinic to date. So having a backbone like that, one, it allows you to be efficient, quick; and it also allows you to diversify the pipeline in a manner that we think is [ confluent ] with the overall risk profile in biotech. Even though we're less risky, it's still less than 50% end-to-end POTS than -- so it's still a jackpot industry. Just in terms of the partnerships, as you mentioned, we have a wide variety of partnerships, almost 14 institutions that we're partnered with. Now academic institutions, folks like MD Anderson, where we have a SHP2 compound that's about to enter the clinic or folks like, Boston Children's, Cincinnati Children's, Johns Hopkins and others that have really been seminal leaders in the genetic disease space. And we focus on, really trying to be a productive, early-stage discovery partner to turn some of those early-stage innovative ideas or understandings of disease into drugs that could matter. I'd say there's also a number of loose affiliations. Take, for instance, it would be NIH, where we worked with them on our FGFR inhibitor. And someone might have an idea, for instance, the idea around the calcilytic agent that's now in Phase II ADH1 came from an NIH investigator that we were just close to. And in that way, we start to build those co-centric circles of trust amongst early-stage academics and institutions that lead us to new ideas and better ideas as we evolve.

This is Jason from the team. This is very, very helpful and insightful. Hey, going off, what you just talked about, can you discuss a little bit about the go/no-to-go decision? In terms of thinking about the different programs, obviously, there are a lot of different inputs. But there's a wide array of modalities. There's an array of different indications. And obviously, a very different risk/reward profile for each of them. At the end, you have this backbone, different teams coming up with different numbers. What to you says this is a really interesting program with high upside, maybe minimal downside, and we want to go with this, where the inputs might be very, very different between the different teams?

Yes. Great question. There's probably 2 parts to that answer. One, how we discriminate between higher reward versus lower reward programs, if you will; and the second being, how we think about go/no-go or stopping conditions for our programs. So I think, let me take the second and then I'll come back to the first. The go/no-go criteria for most of our programs fall quite naturally out of the mechanism of disease and mechanism of action, kind of intersection that I was talking about, that we require when we bring a drug on. And that's kind of the nice thing about genetic diseases. So for instance, if I have a gain-of-function mutation and I know that it up-regulates signaling by 30%, my go/no-gos are going to be associated with that discovery program in as much as, well, can I inhibit that target? Can I inhibit it 30% of the time? Let's say, it's an AUC-driven effect, can I do it 24/7? Now can I do it biochemically? Can I do it in cellular models? Can I do it in an animal model? Do I have the right of profile, so that I can have the correct target coverage and then all of the drug-like properties and things of that nature that go into go/no-gos? So almost quantitatively from the get-go, given these diseases, if it's a threshold effect, I'm saying, "hey, can I put back, let's say, 20% of wild-type protein, and it's a gene therapy approach?" I'm going to say, look, what is the tropism to the issue of interest based on the gene therapy backbone that I'm using? Can I, in animal models or not, even primate models, actually deliver the required amount of BGC that we think we need based on genotype/phenotype and oftentimes also animal model experiments. And so I think the go/no-gos is kind of quite cleanly fall out of these very clean therapeutic hypotheses, where, going into the whole campaign, one understands, not only what the driver of the disease is, but also what one needs to do to it, to optimally have an effect for patients. So that's on the go/no-go side. On the -- kind of how we prioritize risk/reward or how we think about high-reward versus low-reward programs. I think, we think very carefully about whether or not any program is NPV or ROI positive. And we've actually passed on a lot of stuff that ended up getting called together by other folks and ended up actually, in some cases, getting IPO-ed for reasonable dollars. I think we look at the end markets and try to figure out the cost to get to the end market and ultimately, how much money we think the drug will actually make based on the patient population. And if it's not ROI-positive, we don't take it on, or if something changes, for instance, CMC costs go dramatically up and we find out, we can't manufacture protein X or Y or gene therapy X or Y, or the dose is much higher than we expected, then we can shut it down for those reasons. But otherwise, we do have a diaspora of some high-reward programs like TTR amyloid, achondroplasia, congenital adrenal hyperplasia. These are some of the largest genetic disease markets out there, likely multibillion-dollar end markets. But most markets in the genetic disease landscape are likely to be smaller, because most of these mutations obviously limit reproductive fitness. And so when one thinks about that, one needs to be open to playing a lot of singles and doubles forward as well. And we think as long as those have a high POTS and that they're NPV positive, we can prosecute those alongside of the high-reward programs. And that together, many of those smaller programs might come together to be a couple, interesting kind of groves of high value. And maybe the final thing, I'll just say on that is, unlike maybe oncology where you need to hit a certain scale, really, of revenue to be profitable, I think what's been shown time and time again by many of the seminal companies in this space, BioMarin, Recordati, Sobi, Horizon, et cetera, is that even at low revenue numbers, these franchises can be profitable in and of themselves. And so we try to take that viewpoint around many of the diseases we go after.

Interesting. Very, very helpful. And then, I guess the final question is, when thinking about the overall risk/reward of the entire portfolio, certainly you're exploring a number of more conventional and established modalities with small molecules or topicals or biologics. But then, how do you layer in something like, gene therapy, which could potentially offer significantly more reward, but a lot of questions as to how to progress with that, not only from a scientific standpoint but obviously a regulatory standpoint and then, of course, the payer dynamics where new models are necessarily needed. How does that factor in to kind of the overall approach of kind of, again, balancing risk/reward and go/no-go?

Yes. It's a good question, actually. We think less about modality in that respect than you might otherwise imagine. Actually, the way we think about things is we start with the end market and we say, what would the ideal therapy look like that still allows us to target disease at its source. So for instance, the gain-of-function mutation, I don't see necessarily why one -- I mean, yes, you can address it with a siRNA approach or oft-times address it with a gene therapy approach that's coupled. But if you had a small molecule that was cheaper, safer and just as effective or more effective, and in some cases spares a natural protein, why not take that small molecule approach. And so -- but in other cases, let's say, it's loss of function or it's a dosage disease, disease of haploinsufficiency, if it's extracellular, we might think first about protein replacement therapy or enzyme replacement therapy. But intracellular, we might think first of gene therapy. So we really try to design it from the marketplace and then go backwards. And you're right to say that we -- that the toolkit that we have right now is effectively already precedented modalities. Actually, gene therapy, when we started, was maybe just becoming what you might call precedented. But I think we would all argue that the classic gene therapy approach, the regulatory path is well laid out, the manufacturing is well laid out. We're not trying to be technology pioneers here, what we're trying to do is, take technology that others have shown us how to use and to use those effectively in various disease settings. So the 2 AAV programs that we've announced so far, both use AAV5 and AAV9 -- one AAV5, one AAV9, of course. And so that's not -- we're not trying to reinvent the wheel there. Obviously, there's optimization associated with that, but it's relatively straightforward. So I think, all in all, we try to use the tool that is most practical, and it's going to lead to either a best-in-class or first-in-class solution in the marketplace. And all of these tools that we are using between oligos, gene therapy, protein replacement therapy, antibodies and small molecules, they're all are precedented -- they have a regulatory pathway, so there's not a hell of a lot of technology risk associated with it. And I think that allows us to then say, well, what does the pipeline look like in terms of probability, technical success, more towards the kind of classic attributes of tox risk, target risk and manufacturing risk. So we can quantify it better. Like, how do you quantify the risk associated with CRISPR, it's very difficult to do. When once people start pioneering and getting that stuff on the market, I think we'll have a better idea of that. And our hope would be at that time, we could find applications for which that would be ideal and use it.

Great. That's really, really interesting. Makes a lot of sense. Maybe at this time, we'll switch and start talking about some of your programs specifically. So for AG10, how are you thinking about the market with tafamidis launching off to a fast start? And how that could raise awareness and potentially even be an advantage for you guys? And then where do you see your point of differentiation?

Yes. Good question. So number one, yes, I do think that tafamidis' fast launch is a couple of things. It's good for us. It's good for the patient community. Obviously, when you're taking mortality from 42% over 30 months, down to 30%, you've done something really quite remarkable for patients. So I think, in all respects, we're rooting for that drug and rooting for that market to continue to emerge because although, as we'll probably talk about, there's been an exponential growth in scripts and identified patients, there's still very much more to go based on our understanding of how broad this disease is. It's really not an orphan disease. It's a very, very large population. The way we think about our differentiation comes back to, again, an understanding of the mechanism of disease. This is a mass action disease. You could think about it as almost like Alzheimer's of the heart, where you have this tetramer called transthyretin, it's destabilized. And when it's destabilized, you have monomers that are not being proteostatically cleared, and they're pumping together and depositing in the heart. So the less of that toxic monomer, the more you can turn that faucet off, the better it is for patients. And that's been shown across 4 clinical trials in the context of polyneuropathy. It's also true across the genotype, phenotype of the disease. As you have more and more deleterious mutations, they tend to be more and more destabilizing of the tetramer. In fact, we have a slide on our website that shows thermodynamic stability, it's graft against pathogenicity or penetrance in this disease, and it's remarkably linear. And we also know, I think as we've discussed before or can discuss here, that there exists a rescue mutation or really a transrepressor that shows us that, if you glue back that tetramer at very, very high levels together, one is actually able to completely avoid disease, even when born with a pathogenic variant. And that trans-Oleic transrepressor is fundamentally what our drug is copying. And so the point of differentiation, if you want to think about them are, tafamidis is a loose stabilizer. Stabilizes that tetramer about 45%. You could see it almost as a little bit of an inefficient glue. And what we're doing is, super efficiently pulling that tetramer back together, such that, you shouldn't have much, if any, monomer above the proteostatic clearance rates being created. And that gives us a better chance to have a more profound effect on the disease symptomatology, namely, 6-minute walk, rehospitalization and probably most importantly, mortality. And I would say, molecularly, since that's always my interest, the reason that this molecule is differentiated from tafamidis is that -- actually a couple of things. Number one, it binds with greater affinity, its KD2 is superior to that of tafamidis in the second binding site. Two, upon binding, it does a better job of pulling the tetramer together based on inducing hydrogen bonds at the bottom of the thyroxine-binding pocket, which tafamidis does not do. And thirdly, it's just much more available to see the target, much less bound up [ 30% ] as a superior. And so as you think about all 3 of those things, that's what leads to the superior stabilization, almost 2x better than tafamidis, as shown in the clinic, 50% better -- rises in serum TTR levels, as we showed, and against placebo after 15 months as we showed almost a 40% decrease in OLE data. So not a placebo-controlled trial. But against a placebo number, you can get from a draft, we showed 40% reductions in mortality and 30 plus in rehospitalization. And that was a time point at which tafamidis has shown no separation against placebo. So all of these things start to add up. I think we're just doing a better job so far of stabilizing the tetramer, which should lead to a superior effect in the clinic.

Okay. Yes, very helpful. Now when looking at the commercial opportunity here, I know one of the controversies is diagnosis, how to get that up? So I just wanted to get your thoughts on, what the efforts are there? How high do you think it could potentially go? And then, assuming, you're able to get approved and depending on how the data shakes out, do you think there are room for multiple players in this market? Or is it more likely to be a winner-take-all situation?

Yes. I mean I'll start with the last question. I suspect that this market will be extremely large. It's really like a specialty market, a multibillion-dollar market. And I expect that we'll have a variety of different treatments out there that will beat tafamidis, it will be us. And some -- and at least 1 silencer approach. And I've never seen a marketplace that large, just the one drug taking everything for a wide variety of reasons, from price to contracting, to patient preference, to data. Obviously, there tend to be a variety of different players that can take some share, and even a little bit of share in a large marketplace like that can lead to something interesting. So I suspect that there'll be multiple products, which I think is good for patients and the physicians in that space. I do think that we have a potentially best-in-class profile here. I think, optimally, what physicians will do -- is they want to reach for a small molecule, not an infusion and something that actually spares -- or subcu and something that spares transthyretin. Something that's very, very safe, but at the same time completely turns off the faucet of toxic monomer, as I mentioned earlier, that drives this disease. And I think we're going to be, to your diagnosis question, diagnosing this disease earlier and earlier as more and more physicians appreciate, just how common it can be in terms of a subpopulation of HFpEF patients. Prior to really running these trials, there wasn't a lot of reason to be doing and looking for the disease. And there was also a profound advance in the way we diagnose the disease stemming from -- going from biopsy-based approaches to nonbiopsy, technician-scan-based approaches that I think we've talked about. So I mean, just look at the number of patients that Pfizer themselves have turned up over the course of 12 months. They went from 3,000 identified patients to 13,000 identified patients. There's a variety of papers, one prominently from UCL showing this type of exponential increase in the number of people that they were finding at their site. I think a lot, I feel like -- I can't remember exactly what the numbers were, but it is quite -- maybe 120 patients to above 150 or 200 patients within a 12-month time frame. So the total number of patients, if you just sort of back-end the 3 or 4 reports that look at the HFpEF population alone, one would expect the wild type plus hereditary population in the U.S. could be as great as 400,000 patients. Certainly should be no smaller than 250,000 patients. And so we're nowhere close to finding all of those folks. And I think a lot of that has to do with turning up education in the community as well as the most academic medical center so that we are testing everyone who may have a suspicion of disease.

Okay. And then let's move on to your achondroplasia program. How are you thinking about the hurdle kind of rate for data that would make the program both approvable and successful commercially, given that you could have vosoritide on the market ahead of you? And then following on that, if that's the case, would you expect any switching or would it be more of a new start type of market?

Yes. Good question. So -- and I think this goes back to, I don't think we would pursue anything unless we thought it was, markedly best-in-class or first-in-class and obviously, in both TTR and in achondroplasia, or not first-in-class. So the hope is that we can present dramatic efficacy improvements. In the case of achondroplasia, those derived from the fact that we're uniquely targeting the disease at its source. We just presented a poster, I think, last week that many folks picked up, showing almost 100-fold more potency against the MAPK signaling pathway. And uniquely, we're shutting on JAK/STAT signaling pathway, which is the second of the 2 effector pathways that get turned up with these gain of function FGFR3 mutations in the context of achondroplasia. And so the hope here, obviously, is that with vosoritide, you're getting about 1.6 centimeter per year improvement on AGV. Our hope is that, we could up to double that and really return these children back to a normal height. And I think beyond that, as you've seen from much of the animal work that we published, the hope is, over longer term, we would show effect on things like spinal stenosis, foramen magnum, proportionality, things that we don't believe just targeting the MAPKs signaling pathway downstream with the cAMP or extended-release cAMP, or whatever it is, are going to be able to accomplish, just because they really aren't addressing the disease at its source. And so I think against all of those parameters, we would expect a dramatic advantage or else it wouldn't be of interest. But certainly, AGV is the most obvious one, where we should see a competitive advantage against BioMarin and Ascendis. And I think the other thing to note here is that, this is the only oral product that we know of in the pipeline. That's really important, holding down and injecting a child every day throughout the course of their early childhood and adolescence until the growth age closes is a wearing thing for parents to do. And so our hope is that the route of administration is going to be superior as well as the efficacy.

This is Jason again. If we could switch gears just in the time we have left. I know oncology has been a big focus for a number of investors. Can you talk a little bit about the KRAS inhibitor program? Kind of what's your thinking in terms of overall development and then commercial strategy, especially maybe given some of the preliminary results of some of your competitors?

Yes. So the competitive programs that have been talked about today, and I suppose, you're referring to the G12Cs, focus on one specific mutation and focus on inhibiting them in the GDP-bound state. As you know, our program sought from the get-go to actually approach a multiplicity of mutants, to take a pan-mutant approach, and to do so, in the context of a GDP-bound KRAS. And we're using 3 separate approaches to go after that. One was addressing a bulk-binding region in KRAS, the H95 region. The other was to prevent the processing of KRAS by looking at cysteine 185. And then the third was effectively a PI3K effector breaker, which breaks the interaction between RAS and PI3K, which is essential for much of the transformation that has been published by Julian Downward. And so all 3 of those approaches I think we've given updates in the last 6 months. We have crystal structures, at least against 2 of those approaches. We have less than 100 micromolar IC50s -- well, actually in the nanomolar regimes now, cellular activity, crystal structures. Still too early for us to tell, honestly, when we would nominate a development candidate and way too early to tell, to be quite honest, what the -- how the commercial landscape would shape up. I don't think there's any G12D or G12V programs that are as advanced as ours. But we're not in the clinic yet. So time will tell, who wins that race.

Okay. Great. Well, with that, we're up against our time here. Definitely wish we had more. I could talk to you all day about all the rest of your programs, but we'll have to end it here. But thank you, Neil. Really appreciate. A lot of great conversation. And thank you, everyone, for listening. And take care and stay safe.

Yes. Thanks for the time. Stay safe.