All right, great. Thanks very much. Happy to be moderating a chat here with the BioMarin management team. I think Brian, CFO, which I'm sure everybody knows, is going to make some opening remarks, and then I think we're going to use this as an opportunity to spend some time with Greg and talk more about the pipeline that BioMarin's building out. Please take it away, Brian. Thank you.
Thanks, Paul. Appreciate it. Thanks for having us. Good afternoon, everyone. Yes, some opening remarks. We're just a week out of our Q3 report, and lots of strong progress and results this year. We are executing on one of the core elements of our strategy, which is using our financial strength to fund future innovation. Very strong Q3 with revenue growth of 32%. We reached $1 billion of total revenue for the first time in the company history. A lot of that was driven by our Metabolic Conditions business unit with 44% growth. We're also growing earnings per share and profitability. One note there is that we have had two acquisitions in the third quarter, which are driving some level of in-process research and development expense.
A lot of our growth is being driven by the recent acquisition of Amicus Therapeutics that brought two high-growth rare disease assets into our Metabolic Conditions portfolio. That is GALAFOLD for Fabry disease and POMBILITI and OPFOLDA for Pompe disease. These patients, we're growing patients at a double-digit rate. For GALAFOLD, we shared in one specific market, we grew patients at 20% in the third quarter year-over-year. POMBILITI and OPFOLDA is earlier in its growth cycle, and revenues grew almost 60%. Amicus is also more than just a revenue growth story. We think the deal has the potential to be highly accretive to the bottom line. We shared recently that we've identified roughly 50% of operating expense synergies on a non-GAAP basis when looking at the base Amicus operating expenses.
We think those synergies of roughly $220 million will be fully realized for a full year in 2028. The integration will be mostly completed by the middle of next year, and we think that the Amicus business can be run at a roughly 60% operating margin by the end of the decade, which then gets to a healthy inflection on what was already the growing profitability profile of historical BioMarin. I think the last thing I'll mention, and this is where really great to have Greg with us here today, is we are delivering on our objective of rebuilding our pipeline. We've added four new assets to our pipeline this year. The first is BMN 450, most recently announced. This is the first potential subcutaneous injection of a Fabry disease therapy, enzyme replacement therapy. Next is BMN 440.
This is the first potential oral therapy for hypophosphatasia, which is a significant commercial opportunity. Next is BMN 430, which is a preclinical program for a actually recently discovered disease area, disease in ReNU syndrome. Lastly is BMN 820, which came to BioMarin through the Amicus acquisition. This is a partnered asset. We have U.S. rights. It is for FSGS, the rare kidney condition. We are delivering on growth, innovation, and our value commitment objectives, and happy to go where you would like, Paul. Thanks.
Okay, thanks. I will just take the pipeline in a very slightly different order because I wanted to first talk about hypophosphatasia. That is an indication in a market I have been interested in from covering Alexion, and maybe I was wondering what you could talk about the Alesta program, the thesis there, and sort of the parts of the market that it could serve relative to STRENSIQ and the Alexion AZ one.
Yeah. Thanks for the question, Paul. We are really excited about it. We call it BMN 440. It is a small molecule. It targets ENPP1. In that regard, if you think about hypophosphatasia, this is a condition where patients are missing an enzyme. They are missing alkaline phosphatase, and the result of that is they have an increase in their serum, in their bodies, in inorganic pyrophosphate. That increase leads to decreased bone mineralization and a variety of other problems through the patient's life. The available enzyme replacement therapies, we mentioned a couple of them just now, are bone-targeting agents that try to, again, metabolize that pyrophosphate to downstream products. What BMN 440 is a small molecule that is trying to inhibit its production. Think of it as a substrate reduction therapy, not dissimilar to Gaucher.
If you inhibit its production, does that ultimately shut it off and the excess that is around is cleared by the body? Is that the thought?
The thought is that you're actually decreasing the production of it altogether.
Okay.
If you bring it down to a normal level, you can have the same benefits of improving bone mineralization. But unlike the enzyme replacements that are engineered to be bone-seeking, a small molecule has the opportunity as well to reach compartments like the muscle and the joints, which particularly for certain phenotypes in the disease, that's where a majority of their problems come from. It's not necessarily rickets or soft bones that they're complaining of. It's more pain, it's more joint problems, it's sleep deprivation. The hope is here that we can look at each of these sub-segments appropriately, deploy this molecule, and not just deliver an oral, more convenient therapy, but also one that might target these conditions in a way that treats the symptoms that are really bothering many of these patients.
Okay, got it. There's kind of a wide range of phenotypic severity within this. I guess I don't want to say if it's like a family of diseases, it's the same condition, but there's the infantile onset, there's the adult. How do you think about positioning this?
Yeah, I think what we're learning, and some great work done again here by Alexion AZ, is that you need to think of these populations as different opportunities, different needs. Infants, of course, which we're not talking about for BMN 440, have mortality as a major problem.
Why not? Because unseating STRENSIQ is not realistic there?
No, I think it is more that inorganic pyrophosphate is not the issue in infants. There are other sort of vitamin deficiencies that alkaline phosphatase can also address—
I see.
—Where this approach would not be appropriate.
Okay.
But for the children who are diagnosed, again, the phenotype pulls through as children. Those, again, are the ones who have the soft bones. Children that become adults, or adults who are presenting for the first time later in life, those are three different distinct populations that have different needs. In the youngest, it is about the soft bones. In the children who are then adults, we are thinking about things like the six-minute walk test. In the adults, there may be factors involving pain and joint problems that we need to focus on. Our approach will be to develop the molecule in phase II in each of these indications appropriately, think about the endpoints, do endpoint development at the same time that we are doing dose ranging. The data that we have right now is phase I/ II data.
It involves healthy volunteers, also some patient data, and we are presuming that next year we are going to be able to share some of that data. Not only is it PK, but you can think of it as biochemical proof of concept. So we will be looking at blood markers and, again, ability to modulate pyrophosphate.
As you think about these populations going from children to older adults, is there varying probability of success across these populations? Is there a spectrum on the degree to which there are other aspects of the disease in these populations that this could or couldn't address?
Well, the great news here is that pyrophosphate is as close to a validated surrogate as it comes. Now, what I would say is that, again, STRENSIQ and ALXN1850 have shown that many of the benefits that they have shown do track with pyrophosphate reduction. Now we will be decreasing pyrophosphate through a substrate reduction mechanism, but again, as long as we can do that, there is good reason to believe that that is a marker for success. That being said, the established benefits are much clearer in the children when you are looking again at growth plate for things like the rickets severity score or looking at the six-minute walk test. Adults has been a bit of a challenge, where six-minute walk test probably is too heterogeneous of an endpoint.
The work that we're going to do will hopefully not only benefit from a molecule that can reach different compartments than the bone, but also we're going to build on the work that's been done, the great work so far, and make sure that we're measuring the right endpoints.
Okay. Makes sense. Did you say you have some PD data now that you haven't shown?
During the process, we've been able to see both PK and the preliminary PD data.
Okay.
Haven't shown it publicly, but that was data that was an integral part of our thesis to acquire this molecule.
Okay. As we think about seeing the PD data next year, maybe just give people a little bit more context on not just the biomarker, but the thresholds for change that have historically translated into potential clinical benefit.
Yeah, if you triangulate all the data, including genetic models as well as the preclinical models using this particular moiety, BMN 440, we think that somewhere between a 30% and 50% reduction in inorganic pyrophosphate would get us into the levels that we're looking for.
What does STRENSIQ show?
Roughly the same.
Roughly that. Okay, that is reasonably well validated.
Yes.
Okay, got it. How penetrated is STRENSIQ? I know in the infantile population, it is just a no-brainer risk-benefit, but as you get older, what proportion of patients are on therapy?
Yeah, it decreases as you go upwards.
Yes.
In particular adults, not only is there not an approved therapy, but what we have is a mix of willingness to try treatment. Some of that has to do with the profile of the drug. Some of it has to do with being able to identify the patients. Good news here is that the alkaline phosphatase levels in the bloodstream should be a pretty obvious marker to at least begin some of that work.
One of the metrics we have an understanding of is 9,000 diagnosed patients in the U.S., but approximately just 10%-30% on treatment.
10%-30%?
Why is that range so wide? Is that based on age groups?
I just think it is based on available information.
Like epi and just kind of people guessing. I see.
Exactly.
Okay, got it. Makes sense.
There is a lot of untreated patients. The question, of course, is we need to put together a program that proves that, again, the adults can benefit. What we think is we have a good shot at it, given the small molecule moiety that can reach the muscle and joints.
Okay. When we get the data next year, any additional color on how many patients, duration, things like that?
Yeah, not going to share too much more at this point, except to say there will be both single and multiple dose. There will be both patients and healthy volunteers. You can expect, again, that the pyrophosphate levels are going to be an important part.
Are you just unveiling data that you already have, or are you in the process of generating additional data between now and then?
The study is still ongoing.
Yeah.
Again, from a typical phase I/II, as you can imagine, there are dose escalations involved.
I got it. Okay, great. Anything else to add on that one?
No. We're excited to have a molecule again as part of our pipeline that has not only promised to have good biology, but also can reach a fairly large—
Yeah, it's a very—
—Population of unmet need.
Yep. Okay. Let's talk about the deal you recently did with Fabry. I'm a little bit less familiar with this asset as most people are. Maybe just give kind of the background on, one, what got you interested there, and two, how do you see it as complementary to GALAFOLD? Are there different patients that could be served by this? How does it fit into the whole paradigm?
Yeah. The GC Biopharma -Hanmi deal, we call this one BMN 450, so we're moving in a routine fashion upwards. What I would say is that this is an enzyme replacement therapy for Fabry that has an aglycosylated Fc attached to it. The point of adding that moiety onto the molecule is not only to increase half-life, but also is to stabilize the molecule so it can be given as a subcutaneous delivery. As a quick reminder, all enzyme replacement therapies that are available for Fabry right now are given intravenously. Again, that comes with its own treatment burden. Straight up, we think, again, the opportunity for a molecule to try to recapitulate some of that biology, but given as a subcutaneous delivery, is an opportunity, and we hear this from patients and treaters, something that they would value.
For BioMarin, our footprint in this space with GALAFOLD is currently limited to some degree because GALAFOLD, as an enzyme stabilizer, only addresses about 35%-50% of the population. You need to have an amenable mutation, so to speak, and we know that that is the subset that is amenable. The beauty of BMN 450 is it would be mutation agnostic. It could—
Right.
—reach all of the others. Again, we see this as an opportunity not just for more convenience, but we will be studying along the way whether, again, that affords some of the PK/PD profiles that would allow us to look at other end organs, whether it's the heart or other areas, to see if it can bring other things to the table.
Yeah. Okay, interesting. How do you think about the right study and development path for that? As you think about running trials that could register the drug but also be useful from a commercial perspective, are switching studies things that you might think about? I guess there's been different paths, I think, to get drugs approved for Fabry, so how are you thinking about that piece too?
Yeah. The great news here is there are well-established paths. I think there are a variety of subsets of patients, including those with preexisting anti-drug antibodies versus those that are naive, that we'll have to incorporate into our thinking. There's also an opportunity to not only look for naive patients and straight-up kidney biopsy endpoints, but we could look at switching studies. We can look again at other end organs and try to develop the molecule. Another very valuable factor here is that in early studies, you don't have to sit around and wait a year and do kidney biopsies. You can look at blood markers like lyso-Gb3, and again, get a sense of whether or not you're normalizing the biology. Mea culpa, we've only had the drug for two weeks, so we haven't sorted through all of the development plans.
Supposed to know everything.
Yeah. Well, we're not the only people also involved here. We do need to interface with regulators. What you can expect is we're going to move as quickly as we can. There will be a combination of multiple different studies, and we'll be looking at endpoints that go beyond the classic measured endpoints from the regulators.
When could we see some data?
Yeah, we haven't committed to when we're publishing the data, but I will tell you that the study is ongoing. Again, it's a case where this phase I/II is being done in patients, and so it's in its dose escalation phase. As I've mentioned publicly before, we've seen a very low double-digit number of patients. Again, we've seen what we need both on the PK side as well as some of these early markers of biologic activity that we feel confident that this is a very good molecule to be testing in this space. When we have that totality of data, we'll be laying out what you can expect shortly in terms of public disclosures of the data.
Okay, great. All right, let's hit on BMN 333. I think this is one that everyone's fairly familiar with, right? Everyone knows VOXZOGO and the space and the once-weekly thesis here. I want to maybe ask a more specific question, and I know you're not maybe going to get exactly into the numbers, but forgive me because I'm going to give a little bit of context. In achondroplasia, I think we've seen a number of programs go from showing something like a 2 cm benefit in phase I/II studies to closer to 1.5 cm in phase III.
Yeah.
In the context of that, I'm wondering what you think you would need to see in a mid-stage trial to really have conviction that any sort of implied effect size delta will hold up for power and for superiority in a phase III.
Yeah. We are very fortunate here in that we have a wealth of data from VOXZOGO that not only tells us about the variability of response that is inherent in these individuals, also tells us about the factors that go into determining that variability. Things like baseline AGV, age, gender are really important. We, of course, have a PK/PD model, or I should say a PD model, that really tries to get at that. As we dissect our data, we are going to use that, leverage it, and make sure we understand what we have with 333. I think you are also highlighting the fact that when you look at any uncontrolled data source, it is important to look at things like LS means and look at normalized data to the patient population. It is disingenuous to look at raw data and talk about it, again, apples to apples.
Now, looking again at that, we believe we have designed a study with 333 that is testing three doses of 333. It is 30 patients. There is an internal control of 10 patients on VOXZOGO, and that will be supported by a wealth of historic data as well. We have a Bayesian model, again, that will look at dose response and will be targeted for an X% probability of repeating in phase III. We have not said what that posterior probability is yet, but you can assume, again, that it is something that is meaningful. If we see that, we move forward to phase III. Again, we think we have the right molecule with 333.
The hypothesis is that dramatic improvements in AUC, we are talking not just one or twofold, but in our phase II study, as high as eight or more fold of free CNP AUC as compared to a drug like YUVIWEL, that there is more biology to be explored on the dose response curve. It is not a debate about whether there is a dose response curve. I think there is a lot of debate about whether or not that curve has been well characterized and whether there is a shelf. We would argue that the preclinical models and even the clinically available data suggests that this is an unanswered question, and we know that 333 gives us both the PK as well as the early PD.
Cyclic GMP from our healthy volunteer study showed that when you go up higher levels than what YUVIWEL can achieve, that we are seeing two to five-fold increases in cyclic GMP.
The ultimate question will be the card that we can turn over next year. Again, I think in the following quarters, we will give a little bit more clarity on the timing of when we can expect that phase II data to read out. What we have said publicly is it will be in 2027. We are actively recruiting to the study around the globe right now, and again, looking forward to give updates shortly.
Yeah. Okay. That is great. I know I may look young, but I did a deep dive on VOXZOGO in 2013 when it was the animal data and the thesis.
Nice.
Super interesting program. Yeah, it was maybe a little too early. But I remember two concerns back then were, one, if you drive growth velocity too much, you can exacerbate other parts of the disease, spinal stenosis, and two, there was a lot of concern about hypotension, which ended up not really being a big deal with the first gens, but I am wondering with the greater exposure here if that could end up being a problem.
Yeah. Very fair questions, and I think we have had some posters that have shown this data. We have compelling evidence that the hypotension is driven by the Cmax of the molecule. By having a half-life of on the order of 99 hours, I believe it was, instead of one hour with VOXZOGO, you can deliver it, 333, in a way that avoids the Cmaxes by multiple fold that we see not only in humans, but also our animal model, in this case, a non-human primate, has given us confidence. So, the hypotension does not appear to be something that we are worried about. By the way, we did not see it in the phase I as well. With regard to your second question, which is there a level of growth that is too much? In our clinical—
Yeah, too fast.
Too fast. In our clinical studies for achondroplasia and hypochondroplasia for that matter, we haven't seen that to date. Again, being true, we can always ask the question about benefit-risk profile, something that we're going to follow very closely, again, in the achondroplasia and hypochondroplasia market. Haven't seen it in our studies to date.
I think you've also said that a superiority study is somewhat ironically or counterintuitively to people actually smaller than a non-inferiority study. Again, obviously, depending on your assumptions. If it turns out that a non-inferiority path is safer here or you miss that Bayesian probability cutoff, will you still advance this?
Yes. Two answers to that question. When I refer to the study being larger, that's really about a two-sided non-inferiority.
Of course. Yeah.
There are step-throughs that you could think about where you pass—
Do you have to do two-sided? Yeah.
For a formal, in my world, what a non-inferiority study is, yes, you do, but there are ways to actually have a step-through, so you could have a second bite at the apple. That is not the hypothesis that we are testing with BMN 333. We think, again, that the animal models, the clinical data that is out there, and again, the molecule that we have produced is designed to test this hypothesis that dramatically higher AUCs of CNP are going to explore that right side of a dose response curve, and that is what we are focused on at this point. Now, again, do we have contingency plans where we could think about options? We always think about how we could have other approaches, but that is not what we want out of this molecule, and it is not what we think society is valuing.
Yeah. Okay. Fair enough. Let's hit on the FSGS program quickly, and then I'll go back to hitting Brian on a couple quick questions. As it relates to that program, there's already some proteinuria data that's out there. Maybe speak to the data generated so far, and I mean, proteinuria is a well-accepted endpoint. We've seen a number of drugs approved on proteinuria with FSGS. What is ultimately the thesis on how this would be positioned?
Yeah. We're in a very fortunate position where regulators and cooperative groups have worked to move the field. No longer do we need to wait for two years, three years, five years to see eGFR decline or some other harder endpoint. Two-year proteinuria reduction is an approvable endpoint for full approval with the FDA right now in this condition, and some great work done there. That is the endpoint that we're looking at with the molecule we're talking about, BMN 820, excuse me. From that standpoint, the molecule is a CCR2 inhibitor, so it's working on macrophage-driven inflammation. It's an oral therapy. It's designed to be used on top of angiotensin receptor blockade.
The goal here is to see whether or not, again, you can build upon some of the great work that's been done on the vascular side with an orthogonal mechanism of action and improve that proteinuria reduction. There was a small phase II study, again, small in number, though it was mighty in terms of some double crossover there, that showed proteinuria reduction. Again, the phase III study that's fully enrolled had a futility analysis that's been passed through, again, looking to see whether that proteinuria reduction was something that met that threshold or not. So, we're now in a period where the study's fully enrolled, and it will be in 2028 when the card will be turned over and ultimately show whether that proteinuria reduction is achievable. The promise here, again, is that you have an alternative mechanism of action to the purely vascular targeting agents.
You would potentially not have a boxed warning. Of course, the opportunity to target patients that don't have any restrictions on levels of proteinuria, so we're talking about nephrotic syndrome or not. Again, the study is running, it remains blinded, and we're looking forward to seeing those results.
Okay, great. Timing of that data?
2028.
2028. Okay.
It is two full years of required follow-up.
Yeah. Okay, great. All right, back to Brian. How should we think about what the hypochondroplasia launch could look like in year one versus achon?
Yeah. Thanks, Paul. So, just brief recap, hypochondroplasia is the second potential indication for VOXZOGO, with achondroplasia being the first indication approved back in late 2021. I'd frame it up as there's both a tailwind and a headwind for the hypocon launch when comparing it to achon. I'll start with the headwind, which is more complicated diagnosis, which has a lower diagnosis rate overall, and can take more time.
Do you guys have a good estimate on the diagnosed prevalence of that versus achon?
Do we have a rate? Go ahead.
We're following a variety of markers, including—
Yeah.
—time to diagnosis. We again have genetic testing numbers.
If I said to you that our market research suggested it was like one-third as common, diagnosed, does that pass the smell test, you think?
Well, genetically, it should be roughly similar.
Should be equal, but I mean in terms of diagnosed especially.
The number we've put out is, as compared to 24,000 achondroplasia patients around the world in our markets, we think there's 14,000 addressable patients. That assumes that they're being diagnosed.
Right.
That's the work. It's really getting them to diagnosis faster, getting them to the right physician, having the testing, and then identifying them and pulling them through to the treatment option. There is a cut, of course, for severity of disease as well. Our study roughly was -2 standard deviations or less. Again, interestingly, when you looked at the subsets, patients who were between -2 and -3 , as well as -3 or more negative, seemed to do similarly. It was great data, and we think, again, we have a great opportunity with the PDUFA date coming up on May 30th, with the data that we've put in.
Yeah. While there's other factors in that cut from 24,000, which again, for achondroplasia and hypochondroplasia, the incidence is, or prevalence is relatively similar. Part of that cut is the diagnosis, among other things. That's one way to think about how we reduce the addressable population. That's been some of the work thus far. I mentioned the headwind and the tailwind. The tailwind would be all of our years of experience of being in the achondroplasia market and building that over the last few years. Again, diagnosis was already robust from a diagnosis rate standpoint for achondroplasia, but the treatment centers, our objective in the achon launch was to build a pediatric endocrinologist as this treatment home. Since then, there's been this great evolution in single physicians that would see a handful of patients, of building out skeletal dysplasia centers, and now they're seeing hypochondroplasia patients.
Between that, again, diagnosis efforts, once we're approved, digital marketing efforts, we think that's a real tailwind where achon being first in disease, we were really starting from scratch.
Yep. Okay. For next year, it's a May PDUFA. When you guys guide, will you include contribution for that, for hypochondroplasia?
We're working through finalizing the 2027 operating plan as we speak, and that includes talking about how we guide. On one hand, with a May 30th PDUFA date and not having full clarity on—
Yeah.
—label and any other timing elements, that gives me some pause on guiding to it at the beginning of the year. On the other side, it's a mid-year launch, and the ramp with any rare disease launch over the first few quarters, it's largely a second-half launch, as it were.
Right.
You know.
Makes sense.
Will we be able to ballpark it and include it? Possibly. So stay tuned on this front.
Okay. Thanks, Brian. Then maybe just the last question for you is just, you guys have obviously done a nice job of rebuilding the pipeline here. What else should we expect for BioMarin on the M&A side over the next year, and how do you think about the types of deals or deal capacity?
Yeah. So again, starting with Amicus, which was really just a great strategic fit of these already commercial assets, still on a strong growth curve, highly synergistic with the company. So that was a unique transaction. We then prioritized rebuilding the pipeline, and we've talked a lot about today bringing in some of these early-stage assets. That'll continue to be the focus, is clinical-stage assets, ideally within our therapeutic areas, our business units. But if not, there's other areas that are tangential that we're comfortable with.
Could we see BioMarin do a bigger deal, or do you kind of shy away from something like that? By bigger, I mean billion plus.
Yeah. At the moment, we're focused on deleveraging from the Amicus transaction. As we realize the synergies and as we grow our earnings profile and cash flow profile, we'll do that. We've been able to fund these deals thus far this year from our balance sheet with cash on hand and have not changed our deleveraging target of less than 2.5x by the middle of next year. As we delever, yes, we will be able to restore firepower and afford more over time, but at the moment, not committing to specific size of deal.
Fair enough. Okay. Well, thank you guys so much for the chat. We appreciate it.
Thanks for all the earnings questions.
Thank you, Paul.
Appreciate it. Thanks, everyone.