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

Hey, guys. I'm Ellie Merle. I'm one of the biotech analysts here at Cantor. Very happy to have BioMarin here with us today at the Cantor conference. Guys, thanks so much for joining us here. With us from BioMarin is Hank Fuchs, President, Worldwide R&D; as well as Brian Mueller, Executive Vice President, Chief Financial Officer. Thanks so much for joining us today.

Maybe before I dive into questions, maybe, Brian, can you frame for us, not starting with Valrox, where we are with the base business and your confidence in sort of the sustainability and long-run sort of cash generation from this business?

Yes. Thanks, Ellie. Thanks for having us, and thanks for the question. So that's one of the key, I think, considerations, especially after the ROCTAVIAN complete response letter in the U.S. It's to remember the health of our base business. So we're just a few weeks away now from reporting on the third quarter, and we'll give our outlook for the rest of the year. So we'll ask you to stay tuned for more details there. But if you've been tracking the top line story over the course of 2020, we were originally planning to achieve approximately $2 billion in revenue. Now that did include a small contribution from ROCTAVIAN because we were hoping for an approval, but it was mostly the base business. And we will expect that the base business is going to report over $1.8 billion in revenue. And that is continued growth of some of our longest time brands in Naglazyme on the market for over 15 years now but still growing; Vimizim closing in on 10 years in a couple of years and growing, that was 2014; and then launching Palynziq, which is still launching in both the U.S. and Europe. So the health of the base business is intact. We had a modest COVID impact that we adjusted for back in April when we reported Q1 and maintained that same COVID adjustment, if you will, back in early August when we reported Q2. So again, just a few weeks away from Q3, but base business is doing well and long-term prospects remain very strong. We view the ROCTAVIAN complete response letter as a delay, but the commercial prospects haven't changed, just potentially pushing out that revenue curve a bit, but we've got vosoritide hopefully coming next year as well. So commercial business, a change in timing but remained strong.

Absolutely. And thinking about the potential vosoritide launch and also the review process, how should we think about the potential for an AdCom? I know there already was an AdCom around end points, but just in any initial discussions. Do you have any sense for whether this will happen for vosoritide?

I mean I have senses, but they're kind of on both sides of the equation. Like you just mentioned, they've already had an AdCom, but they're a relatively conservative review division. But they have approved things on the basis of AGV and may not need additional advice to interpret an AGV, but they're a relatively conservative review division. And also, the agency has kind of moved away from AdComs and not just because they're virtual. But I could see an AdCom coming up again here because I think they really like to debate these things. They don't -- the -- I think one of the key questions of the review will be on the natural history data comparison because our longest-term data is the 5-year treatment data that we have. And to know how much height gain that you get as a consequence of 5 years of vosoritide, you have to know what an untreated person looks like. And the randomized, double-blind, placebo-controlled trials and p values, the agency is falling off a log on. Natural history comparisons is still a bit less scripted in terms of the agency's playbook. We think we've done a really good job, and there isn't really anything to discuss in the Advisory Committee, but sometimes the agency wants to pressure test things, set standards. So I don't know how to -- I mean I had a lot of facts, Ellie. I don't know how to balance them out and make a prediction.

Yes. No. I know. It's always hard to predict things particularly with the FDA. And I guess just in terms of understanding the long-term implications of annual growth velocity being increased to long-term health outcomes like the FDA might be kind of holding it on, I guess what do we know? And what data do we have? And I guess what are maybe some of the outstanding questions that we don't know from the natural history?

Well, I mean, I think the -- I don't know if it's the thing we know but that if you take an individual who is older than 5 and give them a year of vosoritide, you can document an improvement in AGV, but it's very difficult to document improvements in functionality given a relatively short period of time and given how late you've started within the course of their condition. We have ongoing a study in patients under 5 years of age, and we had to kind of approach that younger age sequentially. So first of all, we couldn't start the under 5 study like right away. Even though everybody recognized the unmet need severity in the under 5 population, agencies wanted to see safety data in children over 5 before you treat children who are under 5. So as a consequence of that, we have much more mature data on children over 5 than children under 5. And it will take those kinds of early interventions to document, I think, the -- in a few more years to document the improvement in functionality or proportionality that I think people are hoping and expecting to see. And further that the under 5 program had to itself be chunked into 3 sequential cohorts. So we started demonstrating the safety in pharmacology in children 2 to 5 years of age and then in children 6 months to 2 years of age. And now we're enrolling the last group of children 0 to 6 months of age into a proper clinical trial 1 year in duration. We finished the enrollment of the first cohort and the second cohort. And the third cohort is imminently going to finish -- we believe is going to finish by the end of the year. So a year from now, roughly, we'll have data on short-term efficacy, 1-year treatment efficacy in children under 5. And I think that's where we're going to really start to see the profound benefits accumulate and be able to see that happen in a shorter amount of time.

Absolutely. What parts of these data will be submitted during the review as you maybe generate the data from the first and second cohort?

So the plan today -- well, plan today, not -- the submission happened with safety data from cohorts 1 and 2, not unblinded efficacy data only. And the FDA's rules about review are the application can only be considered to be complete if it has everything in it at the start to make a review division -- decision. So kind of by definition, the agreement that we reached at the pre-NDA meeting was they don't need efficacy data from the under 5 cohort. Now whether they request that during the review is a whole different matter. But they are at least starting out, believing they can make a yes/no decision at the end of their review process without the efficacy data from the under 5 population. And this was heralded for them, by the way, by the Advisory Committee, who basically said you're going to find yourself in this position just because of the lag of starting treatment in children under 5 relative to starting treatment in children over 5. So their advice was you should be prepared that people are going to extrapolate because the bigger unmet need portion is in the under 5 population and all you'll have is safety data. So the fact that the agency agreed to the pre-NDA meeting to accept the submission absent the full data under 5, I think, bodes well for the review.

Absolutely. And so you will be unblinding each cohort -- you're willing to unblind until you have all cohorts of data?

That was always the plan because the reason for dividing them into cohorts pertain to potentially differences in pharmacology, not necessarily differences in effectiveness as a function of age. And in fact, what we learned was that the younger you are, the higher the administration you have to give. Your heart rate is a function of your age, so the -- and the exposure of the drug is a function of the heart rate. So when you're very young, you have a very high heart rate, you eliminate the drug faster and you need more drug. So we've established equivalent pharmacology of exposure between the 3 different cohorts. And now we -- the original plan was to pool all of those for efficacy analysis.

Got it. Makes sense. And just going back to what you mentioned in terms of like the FDA wanting to understand, really, the natural history comparison for the long-term growth outcomes, I mean what do we know from the data so far -- in the long-term data from the Phase I/II in terms of maybe like long-term outcome benefits outside of just being taller?

Yes. So this is a really key piece of information. What was known at the time in that 2018 Advisory Committee was that there was a gentle downslope in annualized growth velocity in children over 5 years of age, but it was described at the population level. And so that was one known. Another known was the absolute increase in AGV from pretreatment baseline. So in the first 6 months, it was here. The second 6 months, third 6 months, fourth -- and you've seen these data over time. What we did last year is we took an adjustment. Instead of comparing the pretreated baseline, we used the emerging data on the slope of decline to show that the difference in the outer years was not an attenuation of the effect of vosoritide but was actually a constant effect of vosoritide against the background of slowly declining growth velocity. So we had a good inkling about this from prior publications. But what we were able to do between last year's R&D Day and the submission -- and the pre-NDA submission was conduct a much more vigorous degree of analysis at the patient level, matching by age and gender, treated patients to untreated patients to construct a picture of what the expected untreated rate of decline was and compare that to the 10 patients that we've had in the program for 4.5 and 5 years. And we really just moved through that with the agency. I mean that was a lot of good stuff. So that's now known to us, and it's going to be a subject of a review of them -- by them. But it's really the sweet spot of what they were asking for when they asked us for a contemporaneous natural history data. They didn't mean like in a peer-reviewed publication. They meant at the atomic or patient level so they can do matching between our treated cohort and the untreated cohort.

Absolutely. Yes, that seems like a very powerful comparison to make. And outside of looking at height and growth curves in that comparison, are there other maybe more qualitative metrics? I mean, I know it's hard to put up [ numbers ] precisely, but things like surgery or spinal complications, any trends like that emerging as you do this comparison?

Emerging? Yes, I mean there are some emerging things that we're observing in the data overall. So for example, there is a relationship between quality of life and your baseline statural -- degree of statural impairment. But there's more variability there than there is treatment effect in a very short-term period of time. So I think it -- we actually saw a similar kind of a thing in our Vimizim program actually, where the pivotal trial was just a little bit too short to capture the full functional benefit of the effective treatment and you have to go longer. So we see a similar kind of thing happening here, which is, again, that there's a relationship at baseline before any treatment between your height and your functional independence, a small effect of that during the pivotal trial but over time accumulating more dramatic effect on functional outcomes. So the program is sort of playing out as kind of we anticipated it might. That is to say the most sensitive indicator of the effect on stature is going to be in the AGV, which is why there's so much regulatory precedent for approval on AGV and no real requirement for demonstrating these other functional outcomes because the functional outcomes are so tightly linked to the statural impairment that people don't really doubt that if you can improve statural, you'll improve outcome.

Got it. That's certainly an important link for sort of the field to recognize in this submission. Turning to Valrox. Sorry, I know you guys have been dealing with like a lot of questions on this lately. But in terms of thinking about -- now that we're very much focused on this Phase III readout and thinking about what should be looking for in terms of an annual bleed rate reduction, I guess, and it seems like the run-in was a bit higher than you had powered for, I guess, 4.8 versus the 3 to 3.5 assumed in the power rate, I guess what's the smallest decrease in annual bleed rate after 1 year that could still drive a stat sig study? And how should we think about also, outside of just stat sig, what you would view as good data?

So 4.8 to target -- I mean sort of 4.8 is the baseline, 1.5 is the target. Could 2 be significant -- could 2.25 be significant for 2? And 2.75, could that be statistically significant? What's the smallest number? I actually don't know it off the top of my head. It's probably somewhere around 2 or 2.5. But what that means -- I think a different way of looking at this is to really fully understand what power means. So 1 minus the power is the chance of the difference being as large as 4.8 versus 1.5 but having an inability to discern that there is truly a difference statistically. So 1 minus that power is less than 1% in our case. The chances of there being that magnitude of an effect on bleed and us missing it for statistically random reasons is less than 1%. So then what's your confidence that -- so 4.8 is a solid number. That's the number we're going to compare it to. It's not going to vary any longer. It's 4.8. What's the chances that the bleeding is 1.5, 1.8, 1.9, 2.0? Well, in the 3 studies that we've done -- in the sort of 3 cohorts we've looked at, it's always been 1.5 or below. So unless the failure rate is astonishingly higher -- and I mean it's got to be a lot higher, we should be in a very comfortable statistical zone.

And in terms of physicians interpreting these data, especially in the context of Hemlibra, what degree of reduction do you think would really raise maybe physicians' eyebrows or would you see as highly meaningful and impactful from a commercial perspective?

Well, I think you got -- you probably -- it's all like the cancer survival numbers. You got to kind of meet to -- keep up with the Joneses at least numerically. So I mean, I think you've got to be in shouting distance of 1.5 in the first instance. But if you're -- the other key parameter that I think people are going to keep an eye on are the proportion of patients who are having 0 bleeds or 1 bleed, maybe 2 bleeds, call them prophylaxis success stories. The proportion of those patients who are prophylaxis success and free of prophylactic therapy becomes a really important number because the -- as good a drug as Hemlibra is and it's an incredible drug in the inhibitor population, it's a great drug in the non-inhibitor population, but it's still -- because -- compared to where you came from, right? So -- but compared to a fulsome life in hemophilia, emicizumab still does require chronic administration. The longer the interval they use and frequency, the higher the propensity to have breakthrough bleeding. And so what you'd really like to imagine is the state of being 0 bleed, 0 prophylaxis is a much better health state than low bleeding on Hemlibra. So if you start to think about -- imagine a Kaplan-Meier curve that says over time, what's the 0 bleed, 0 prophylaxis rate? So the basis of our Phase III -- Phase II study, that's 86% in the 6e cohort at 4 years and 85% in the 4e cohort at 3 years. It's 95% in the Phase I cohort at 6 months. There's a small amount of attenuation from 6 months to 1 year if you look at the trajectory of Factor VIII curves. But losing as much as 6 units of Factor VIII a year, the proportion of patients who are going to be going back on prophylaxis over time is going to look more like the 4e and 6e group than it's going to look like anything else. And so you could be looking at -- if you -- again, you mentioned the Kaplan-Meier, 0 bleed, 0 prophylaxis proportion of patients, that could be like not hitting 50% median time, median durability. That could look like 7, 8, 9, 10 years. We actually showed this -- Dr. Marks actually, the Head of the CBER, asked us this question during -- on the day before the complete response letter, and we were looking at these low S curves, and I know you've seen black is the low dose, 4 years of exposure from the Phase III study -- from the Phase I study, gold is the 6e dose from the Phase I study. And he said, "What do you project the trajectory after your Phase III results to be?" And we said, "Well, we didn't put a lot of modeling data in the submission because we thought that the set of strength of evidence was so good and it wouldn't even seem necessary." But it looks like it's about -- what he said, he interrupted the -- "It looks like it's going to be about 7 or 8 years before the population is as low as 7% or 8% -- as low as 5% to experience breakthrough bleeding of any meaningful degree." So I think somewhere in the agency is a belief that 0 bleed, 0 prophylaxis is -- to get out to meaningful loss of durability, you're talking about a lot of years.

Absolutely. And it seems like that's highly meaningful to patients. Outside of us all squinting at the Factor VIII expression percentages, it seems like prophy is highly impactful to patients. And just, I guess in terms of the agency, I mean it seems like you had some discussions right before you got the CRL with them and...