BioMarin Pharmaceutical Inc. (BMRN) Earnings Call Transcript & Summary

All right. Welcome, everyone. It's Josh Schimmer from Evercore ISI Biotech team with BioMarin. Yesterday after a really busy day at our conference, I was thinking the last thing I want to do is listen to a 3-hour R&D event from the company. But you can turn it on double speed and turn it into just 1.5 hours, and I thought it was excellent. As I really enjoyed listening to it. You covered a lot of ground and introduced some really compelling new programs. I'm going to vote this BioMarin's best R&D Day yet in terms of shifting my perception of the company.

So I've jotted -- as I was taking notes, I was jotting down questions, there's not going to be like a great order to this. And aside from kind of following the presentations from yesterday, I'm going to start with the steroid prophylaxis commentary about preserving Factor VIII levels with ROCTAVIAN. Are we rethinking the way that steroids work to maintain expression?

I think there's been a constant evolving, I think, in the understanding of what causes vector expression loss and what the role of potentially antecedent or ongoing evidence of liver inflammation is. And it appears there may be more than 1 molecular mechanisms by which that all happens. And I was talking to Jeff, our Senior Adviser and Chief Medical Officer, former about this. And unfortunately, in the next X number of years, all we're going to have is handfuls of collections of experiences as opposed to definitive understanding of the molecular basis of those responses and therefore, the interventions. I think the thing that attracted and compelled us to the regimen that we're using is we have a lot of experience from on-demand use of Valrox in our Phase III trial, which basically taught that playing, if you will, whack them all late in the course didn't really change very much. And also that you can jump on the apparent hepatitis as quickly and as hard as you want, and that's not going to make much of a difference either. So the other end of the spectrum to go to -- it turns out to actually be a lot easier on patients anyway. So you could look at our motivation for doing that in 2 ways, 1 of which would be to say, well, maybe this is biologically going to be the best approach that is prophylactic before the liver really gets ripped up or you could say, well, it's just simplest on patients because always following their ALT didn't really make a big of difference in the end anyway. So for both of those reasons, I think it's really important. Probably, if we can really simplify how we're going about this, I think it really gives us an additional leg up in terms of people's willingness to try us.

Got it. So the next topic you covered was the short stature trial, academic-sponsored trial. It sounds like you're seeing some really good early results there. Are all the 6 indications listed kind of the early targets that you're going to be focused on to think about registration pathways or which ones are the ones we should focus on?

Well, this is kind of a think out loud response because we're just becoming aware of the data. And I do think that Dr. Dauber's enthusiasm propels us to get more timely in terms of our specific plan in this regard. I think there are fundamentally 2 different approaches, and Dr. Dauber, when we're in our initial discussions, sort of undertook both of these, but other leading geneticists have come to us with a similar idea. So the 2 different ideas are mutation-specific or polygenic. Now our groups published some really good data on polygenic risk score predictive ability. So one could imagine taking -- let's say, it's 3 standard deviations, the 0.1% of the staturally impaired population. So they're fairly severely impaired. And those who have a polygenic risk score that says they're going to be less than 3 are potential candidates. That's 1 way to go. The other way to go is mutation-specific. And I think we're in the early days of having gotten VOXZOGO across the finish line in the U.S. and the EU and starting to bring it around the world. The next step for us is -- was outlined by Harold, extending the VOXZOGO indication, demonstrating final adult height and other functional benefits of and starting to think about what other population beyond achondroplasia can be addressed. So a lot on our plate, but I think it illustrates the power of genetics to transform people's lives.

What do you think is going to be bigger, VOXZOGO for achondroplasia or VOXZOGO for non-achondroplasia?

I think we're going to be right by saying achondroplasia for the next 5 years and ask me again in 5 years.

What is the hypochondroplasia phenotype? I guess it's a milder FGFR3 activation? Does that mean patients are [indiscernible]?

Across these different mutations, there's quite a bit of phenotypic heterogeneities. For example, subsets of Noonan have more of a cardiac-dominant presentation. Some of the populations have more and some have less disproportionality. So it's a bit of a hodgepodge in terms of -- interesting hodgepodge too at a molecular level, why mutations and related similar pathways have such different phenotypes. Even within achondroplasia, the mutation is homogeneous, but the phenotype is heterogeneous. You have some children who have terrible foramen magnum narrowing, for example. So I think the next phase is going to be that, again, combination approach, which is validating that we can create a biologically meaning effect for these children who are going to be severely impaired that don't have an intervention as well as documenting the bigger picture health benefits of correcting that phenotype.

Got it. So I think 1 of the next topics that you had covered in the event was the PKU and the preclinical signal and the work you had done to rule out vector integration as an etiology. So if it's -- and I thought you did a good job ruling out vector integration as the cause. But what can you rule in if it's not vector integration, what's the working hypothesis as to what might be causing this? And as you think about, and this goes far beyond BioMarin, obviously, as we think about gene therapy, broadly, AAV targeting the liver. Do we -- should we be cautious about using this type of a therapy in immune-compromised individuals?

Well, I think that's a very relevant question. I -- in our expert interactions, the other hypothesis side is starting to come a little bit more to the fore. As people stare at the fact that, while on the one hand, you can document a certain degree of clonal abundance of certain integrations near certain low side, on the other hand, it's not anywhere close to a dominant finding in the tumors of animals in our hands or in other people's hands. And so the other half of your question is what causes cancer. What causes cancer -- Josh, what causes cancer? And I don't know how much we're going to really dig into what causes cancer. And my -- because our main business, obviously is a little different than that. There's obviously an interaction with the vector. And so a hypothesis could be very small number of animals, but hypothesis could be that something about the transferring product itself, but I think that's the next hypothesis to test. And what I mean by that is you can get in lower species pretty exuberant expression of a protein that is normally under a little tighter control than that, and that can stress the liver out. And so the same way people might talk about, well, a factor VIII is hard to make for hepatocyte. Eventually, that's really hard in the liver and the liver is going to wear out. Well, you can imagine the same thing is true for phenylalanine hydroxylase and liver health. And when they're stressing the liver, especially if it's got an immune system that's not got, especially when the background of it has been mutagenized to create the PKU phenotype, you're sort of sitting duck for the third hit. And how much we should ring fence discrete populations within the PKU community, I think, is going to be an important discussion on the go forward. I think in the first instance, these people are otherwise okay, except for their PKU. And so it's probably odd sub slices of that population who have other conditions as well. But it's a question to keep our eye on, it's years in the unfolding. I think the exciting thing that we talked about yesterday through the PKU is both combination of why we're confident, not certain, but confident and optimistic that we'll be able to negotiate our way through the holders is that we've met the criteria to expand the initial cohort of 6e13 into a larger cohort 6e13, and that could be the basis for picking a registration-enabling dose. So I feel like we're making progress in the PKU gene therapy. We had a little stubbed toe, and that happens in these super mega innovative spaces.

My Internet connection has got a bit wacky, but hopefully we can sort it. Have you looked at like an anti-capsid experiment in the same animal model to rule out the capsid is the cause or other transgenes to rule out, just kind of, again, the nonprotein-specific etiology?

We've done a variety of experiments. I don't know that they're reading exactly objectively like what you just articulated. But I think the group has really gone deep into the area as well as -- we had the benefit of don't forget, an ASGCT roundtable that Kevin chaired, our Head of Research. There's the FDA advisory committee on the topic. And we had a bunch of internal expert groupings as well for the regulators and scientists in the field with the idea of finding where to stick the shovels on this pace. And as I said, I think we're -- you can look at like sort of the Zolgensma footprint in -- on 1 level is obviously maybe -- the other side of that SPINRAZA. But nonetheless, the agencies internationally did not require preclinical, if you will, carcinogenicity types of studies for those. And I think that reflects a -- the fire hasn't left the fence yet. I mean you can see these things in mice, but you don't see anything that harkens to this in humans. And so theories aren't really part of a risk-benefit calculus. And so now where do take these additional findings is there probably will be a step forward in terms of patient communication and education and things like that. But we'll just have to be interacting with the agencies to understand a little bit more about where the common ground is.

The A1AT program is pretty cool. You have the potential to address the heterozygous population. What's the burden of illness for the liver and heterozygous, not as familiar with that?

That's a good question. And I think the question below that question is going to be, can you identify patients in that overall population who are enriched for poor outcomes that are attributable to the alpha-1 phenotype. So if they have -- let's say, you're talking about a population with an odds ratio MZ versus normal population of 2. But then you find a subpopulation of that, which it's 5 and you say, holy cow, this could be a great population in which to demonstrate reversal of polymer. You also have to understand other attributable risk factors as well because maybe they have 10x risk from hepatitis or something. So I think we've got some work to do to sort of discretely identify the subpopulations that are going to meet those criteria. But in the first instance, I think take that same question and apply it to the ZZ phenotype. Because I think the first leg of the race is going to be who can document the biggest impact in liver health by ameliorating this globule formation. The thing that gives us a lot of excitement about that is the polymer reversal data that you saw because -- I mean, it's 1 thing to silence the gene and then whatever is there is just there versus now we can dissolve the polymers because we have found the molecular basis of the covalent interaction that is responsible for polymerization and we can interrupt that. So that's -- that will be an interesting first leg of the race. And I think we like our chances there.

The mechanism itself is really provocative. Is this something that you'd look to find other applications beyond AAT deficiency? I mean so many genetic diseases are caused by sticky protein, sticky RNA that if only you could block [ over-optimization].

Yes. Yes. I think my general sense of the answer to that is that protein-protein interactions disruption using small molecule has been very difficult. It's much simpler to inhibit the protein than it is necessarily to stop it from interacting with another protein. Having said all that, here's a good example of when you can. And so I think that's a bigger picture benefit of BioMarin is we don't come in necessarily saying everything's got to be gene edition. It's like we're not like a hammer in search of a nail all the time here. It's like -- it depends on the biology. We start with how well understood is the biology. And then what is the right intervention that addresses that unique molecular aspect of biology. So sort of I contrast this with like amount of order of a cancer company that I love. And they're incredible at finding pathways, but it still is in humans to determine whether that pathway is as addicting as we think it is. Whereas here, you know you're born with the condition. You know what started the fire. So that, I think, is a big advantage to this -- and that's why I think -- that's why I'm so proud of the team yesterday in terms of bringing together sort of 2 very concrete disciplines. One is genomic sciences and the other is the engineering of drug development to be facile because I think on balance, these molecular insights are going to yield much, much, much more fruit as time goes by.

Probably my favorite program now is 255. It seems still intuitive and obvious almost to the point of like what I feel very surprised that I've not seen other approaches can target this specific enzyme. Why do you think that is just over what -- no one was paying attention? Or is it a difficult target to drug?

I think the Alnylam's credit, they were -- we got this tool, what else -- what can we whack. And they thought carefully about the molecular pathway in PH1 deficiency, and it led them to be able to innovate really quickly. And I think they want to go after some of the same indications that as we pointed out yesterday, it's not crystal clear why they're not at a higher dose, and therefore, you're able to get more effectiveness out of what they're doing. And so maybe there's room on the table there. This is a great example, though, of -- so I think if you have a technology and you're restricted to it, you don't necessarily think about all the rest of the opportunities. So here's where sort of small molecule has the potential to be better by distributed, better controllable. And you don't have to get into that sort of advanced chemistry business. And so this is, again, where the advantage of we have is -- we said, we don't start this by saying, okay, I'm going to make gene therapy work in this case. That's not the approach that we would take. We would ask what's the best approach to take. And we like this. And obviously, other people thought it roughly at the same time and developed similar...

But no other small molecule approaches.

Us, internally?

No, no, no, to inhibit glycolate oxidase -- are you the only small molecule...

BridgeBio announced recently that they had a program, I think, in stealth mode like we did relatively [indiscernible]? Who knows? It's just going to be 1 of these 4, 5 things and we're at [indiscernible] before you know it.

All right. But pretty sizable market opportunities of...

It's crazy. I mean who doesn't know somebody who's had a kidney stone and most of those are oxalate caused, and it's a terrible event when it happens.

I'm going to skip then to the skipper, the exon 51, that was pretty interesting. Like so you're getting -- I'm not entirely sure how this works. You've got a 100-fold more splicing that doesn't mean you get a 100-fold more protein expression, I don't think. But -- well, maybe we can talk about that. Is there a disconnect between the increase in splicing versus protein especially?

Yes, there is. I mean it's not a -- it's not a straight 1 curve. So you do this over a range of doses, like skipping -- you'd like to say, like 1% skipping, 1% protein, 5% skipping 5% protein. It just doesn't -- it's not linear like that. It's more linear in the fat part of the curve, and that's gratifying because we're not looking for a little bit of skipping, which is arguably productive of dystrophin. We're looking for that next big wallet. We were looking for whatever we were was looking for in the first instance.

Right. And so how exactly did you find it? You were just looking for different sites to target that others hadn't explored? Or was there something more to it?

Human power. The [indiscernible] at the -- in the Leiden Group. This gets way back AVI BioPharma, and they were in sort of a competition to exploit this, and the first thing that they came to was using splice sites in the first exon of the splice of this pre-mRNA. And it's interesting there because what the Leiden Group discovered was that it's really -- the better sites are in a specific location. And that's actually the basis of the patent that we have that causes Sarepta owe us a royalty. At the same time, that excitement led GSK to license what became drisapersen. And the rest of the drisapersen story, I'm sure everybody knows. But at the same time, the lab group was saying, well, this is just the first thing we found. What about -- if there's -- if we think that we find a better site than AVI Biopharma found, then it's possible that there are other better sites elsewhere. And they just sort of quietly underground as best as they could in this collaboration context to we'll do what's called tiling. So they basically made antisense oligos for like every frame of every exon. And this site popped up as unbelievably potent, and then they just refined it.

Are you kind of screening other disease indicates -- splice disease indications like to find similar enhancements in other settings?

So this is the whole story of the Deep Genomics.

Absolutely Deep Genomics. Okay.

Because you say splice site, so you imagine a specific molecular mechanism, whereby expression can be controlled. And the cool thing about the Deep Genomics is that they don't care. They're studying every possible alteration in gene expression. As Kevin likes to say, it's a pretty simple thing to say, but a pretty complicated thought to reduce the practice, which is to leverage natural variability in human populations to understand what makes things -- people different from each other. And so it's a massive data undertaking. And so every sort of quarter, we put into the system -- a dollar comes out for us in terms of, okay, that teaches these unique sites if they're this far into an intron are amenable to this blah, blah, blah. And like Kevin said, or I said about, Kevin, his dream is to understand the circuitry. How is this controlled at a deeper level than just how we think of the lac operon.

And last question, just kind of directionally, is your R&D budget going to go up substantially to invest in all this? Or is it going to be reallocation of R&D dollars to advance these programs?

Well, that depends on what you mean by substantially. I think Brian would say it's going to go up, but the percent of R&D spend as a function of our revenue is going to go down. I think I would turn to my team and say, "Wow. Look at all the extra money we're getting." So I think it's going to be a combination of all of those things. I think Brian has got an eagle eye on where the company is spending money. He wants to make sure that the money goes to the right places for the right shareholder value-creating things. And I think my team is -- we've been grinding away stuff for a while, and I think we now have a great opportunity from the pivot of the pipeline to take a look at how we do our work and take our work platform to the next level. So there's a lot going on organizationally as well as scientifically.

Yes. I can jump in there briefly. I know we're over time, but I think quickly, it's loosely sort of yes, yes and yes. We hope to get leverage. We're gaining efficiencies from the current organization as we strive through our operational excellence initiatives, et cetera. We will see a shift in dollars as we see the late-stage programs, ROCTAVIAN, VOXZOGO nearing the end of their development cycle, we're going to continue development of those programs, but we'll be able to reallocate some dollars to the earlier-stage research. And we intend to grow R&D as we grow revenues, but importantly, improve margin.

We're a little over time. Thanks so much. Great, great, R&D event.

Thanks.