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

Welcome, everybody. I'm Hank Fuchs, and this is BioMarin's R&D Day 2021. Since we last saw you in 2019, it's been a busy year. We've been noses down, and there have been a lot of deep technical and infrastructural developments that have occurred at BioMarin to support our pipeline pivot that I want to tell you about. We're thrilled to share some of the early outputs of that today. The agenda is here, but part of my introduction will walk you through the key elements of each of these presentations to highlight some of the most important things you're going to hear. Let's dive in. First, I want to remind you all of our scientific strategy often referred to as the core 4. Then I want to brief you on the outputs and next steps of our pipeline pivot. You've heard us talk a lot about it, and today, you'll get a look under the hood. I want to provide a little bit of a spoiler on some of the important things you'll hear today from each of our individual speakers. And finally, I'll then hand it off to the speakers for a more complete picture of what's new. Our scientific strategy can be summarized by what we refer to as the 4 core attributes. And I want to revisit these because they are an important reminder of what BioMarin has been doing all along, and they are a framework for our early-stage programs that we'll be revealing today as representative outputs of our pipeline pivot. So to remind you, the core 4 attributes are: first, we work in disease spaces in which the molecular basis of the condition is reasonably exactly defined as is often the case in genetic diseases. We develop interventions targeted at these fundamental molecular defects in order to demonstrate benefits on either biomarkers or early signals of clinical effectiveness that can tell us that we're on the right track quickly, and which will themselves lead to transformative impacts into important medicines. We've applied these attributes in the rare and orphan spaces, and now we're applying these principles to larger targets. How we translate genetic discoveries into transforming medicines is mapped over time. And the resounding point is that while our capabilities and expertise and targets and impacts are changing, the core of what we do is not. The foundation of the company, and therefore, our reputation has been in rare and orphan conditions by focusing on the underlying expertise that enabled these therapeutic commercial advances. And this provides proof points that the core 4 works and demonstrated why we're well poised to take advantage of the genomic revolution. We've developed multimodal expertise in discovery, development, commercialization, manufacture, supply and launch of small molecules, peptides and biologics. We've developed a global fully integrated organization. We have a durable commercial portfolio and an increasingly global revenue stream to support the pipeline pivot. And the organization is made up of and led by leading scientists in their field. The present BioMarin, the pipeline has pivoted. It's pivoted around that strong foundation, enabling us to apply our core scientific principles to larger targets. We create toeholds in therapeutic areas. For example, you've heard us talk about vosoritide in achondroplasia and its potential to expand in other conditions. And through a series of toeholds, I'll demonstrate in a couple of minutes, we've expanded the footprint of application of BioMarin even more broadly. We're developing deeper and broader talent, and we're developing more and more development efficiencies, ultimately designed to enable long-term growth and near-term profitability. And then that leads us to what's next, and what's next has got me really excited. It's really about the continued demonstration of the success of this engine applied to the genomic revolution and recognition for the value we create for patients and shareholders maximally leveraging the core 4 to enable top value as a biotech organization and to enable transformational medicines to come out of our shop. The future of BioMarin is described by a virtuous cycle of invention and innovation. By creating value for patients, we created an enormous amount of shareholder value and an enormous amount of employee fulfillment, which enables more advances on behalf of patients and the cycle can continue. This simplified diagram captures the essence of why core 4 applied to genetic advances enables a sustainable flow of pipeline opportunities, which we can turn into new medicines. Many common diseases are groups of genetically distinct relatively rare genotypes. This is shown on the left and the middle parts of the diagram. However, using genomic sequencing and analysis relatively genetically uniform subsets illustrated in the middle part of the diagram can reveal vast numbers of promising therapeutic targets. And those therapeutic targets will reveal the opportunity to create transformative medicines that address potentially larger populations and enable us to break heterogeneous populations into tractable populations for investigation. Thus, this reclassification of rare genetic subsets to lead to more therapeutic candidates, each of which has the potential to have broader and broader applications. We've published this methodology earlier this year using vosoritide for the heterogeneous population of individuals with short stature as an example. To illustrate further how leveraging the foundational expertise we just discussed on the left, core 4 attributes, which compromise our scientific leverage, and on the right, where that scientific leverage meets therapeutic and investment leverage. Investment leverage comes from indication expansion, franchise building and the development and application of our deepening and technical expertise. For example, lysosomal storage diseases, which, in many cases, are skeletal dysplasias, have led us to achondroplasia, another skeletal dysplasia, which has further opened opportunities for expansion into other skeletal dysplasias, not caused by the same mutation that causes achondroplasia. Brineura is a treatment for CLN2 deficiency or Batten's disease, and it's been identified as a genetic epilepsy. And through genetic sequencing of children who have seizures, we can identify other genetic epilepsies that can be targeted without having to start all over from scratch. With expertise in AAV gene therapy, we're able to leverage learnings from ROCTAVIAN and apply them to BMN 307 for phenylketonuria and to BMN 331 for patients with hereditary angioedema. And finally, building out therapeutic franchise areas, we're developing progressively better therapeutic options for patients with, for example, phenylketonuria. To exemplify the progress that we've made over the past couple of years, the pipeline demonstrates a rich abundance of [indiscernible] under therapies. Many of these remain unnamed, and several of which you'll be hearing a bit more about today. However, there are many rows of undisclosed assets, which represent discrete candidates. And these numbers are accurate as of today. In other words, in addition to our disclosed programs, we now have in addition 6 undisclosed cardiovascular programs, 10 undisclosed CNS programs, 3 undisclosed hematologic and metabolic disease programs and 5 undisclosed musculoskeletal programs. Now I'm going to take you through some of the most important things you're going to hear today before handing it over to the speakers themselves. Geoff Nichol is going to talk about valoctocogene roxaparvovec and an important prophylactic steroid study, which is designed to improve and simplify the patient experience receiving ROCTAVIAN. In regard to VOXZOGO, we're excited about the positive actions that have been taken by health authorities around the world, but we still have work to do on VOXZOGO for achondroplasia, particularly in the United States. And we also have opportunity to expand its application. And in particular, a few years ago, Dr. Dauber spoke to us about the potential on a scientific basis of vosoritide to help children with genetic forms of short stature. Today, he has an active IND and a final protocol. And in fact, he's begun recruiting children with statural impairments due to genetic mutations other than those underlying achondroplasia, and he's generated some data today that you might find of interest. I'm thrilled and pleased to introduce Harold Bernstein as our incoming Chief Medical Officer. Harold has the privilege of building on the accomplishments of Geoff's team, handling the life cycle programs and the later-stage programs that you're going to hear about and helping the pipeline pivot in the early development spacing. You'll see from Harold's background how perfectly suited he is to join BioMarin at this particular time. Kevin Eggan is going to give you an update on our emerging asset in our pipeline, namely BMN 307 for phenylketonuria. And to talk about our current understanding of the potential for AAV vectors to be responsible for insertional oncogenesis. Kevin is also going to talk about the progress that we've made identifying toeholds and therapeutic foci beyond where we've already discussed, so that BioMarin can continue to scale our success through mechanisms of leverage that I just discussed. David Lomas is going to take you through the biology and disease mechanisms underlying alpha-1 antitrypsin deficiency and tell you why targeting the liver component of this disease presents an important therapeutic opportunity acknowledging that others are also pursuing this. Our approach, however, has the potential to be a best-in-class approach. As exemplified by Dr. Dave Jacoby, who's going to follow Professor Lomas and talk about the relationship between alpha-1 antitrypsin deficiency and our molecular approach through BMN 349. Dave is also going to talk about 2 additional products, which have moved into or will be moving shortly into the clinic, BMN 255 for hyperoxaluria and BMN 331 for hereditary angioedema. Kevin returns with news about our next IND, BMN 351, for Duchenne Muscular Dystrophy and to highlight partnerships that broaden our undisclosed portfolio and illustrate how genomic advancements are creating these toeholds as launch pads for sustainable innovation. Now it's my privilege and pleasure to hand it over to our speakers. So with that, now over to you, Geoff.

Thanks, Hank. As we engage with health authorities for the review and approval of Valrox, we're also actively broadening our evidence package with these 3 exciting expansion opportunities. First is our Study 203 in those with preexisting antibodies to AAV5. Second, Study 205 in those comprising 20% to 30% of the hemophilia A population with active or prior inhibitors to Factor VIII. And third, and the study I'd like to concentrate on today, Study 303 that uses a prophylactic corticosteroid regimen. As you know, we saw gratifying results in the limited number of subjects we treated with prophylactic steroids in our Phase I/II study. In contrast, for the Phase III study, we used a reactive steroid regimen that commenced with the first ALT elevation, which spared about 20% of our study population from the need to use steroids at all, but with somewhat lower overall Factor VIII expression than in the Phase I/II study. As you can see here, with Factor VIII levels in the first year after treatment shown in green for the Phase I/II study at the 6e13 dose, in red for the same dose in the Phase III study and in blue for the 4e13 dose. We did learn additionally that more intensive immunosuppressive regimens aimed at maximal reductions in ALT made no material difference to Factor VIII expression at 1 year compared with a less intensive regimen. Study 303, using a prophylactic regimen, builds off both the Phase I/II experience as well as some preclinical data showing greater expression with a prophylactic versus a delayed regimen, as you can see on this slide. On the left, steroid use in the mice started the day before or the day of the AAV infusion and led to increased expression of the transgene, in this case, plasma human alpha 1 antitrypsin. While on the right-hand panel, steroids started a week after AAV infusion did not increase transgene. In this case, plasma Factor VIII expression, suggesting benefit from starting steroids early, but not later. Study 303 rounds out our exploration of steroid use in pursuit of the simplest and least demanding immunosuppressive regimen for Valrox. Here's the scheme of the study, enrolling up to 20 subjects over 18 with severe hemophilia A, who received a prophylactic corticosteroid regimen starting immediately after the Valrox dose, with a subsequent taper and adjustments as needed for ALT and Factor VIII changes. Each subject will be followed for a year with the usual assessments for efficacy and safety and then enter a long-term follow-up. While we anticipate results from this study late next year, what I would like to highlight now is what we see right now, which is just how briskly this study is enrolling, exemplifying the appeal of Valrox to people with hemophilia A. It's easy to overlook in the many discussions of approvability, integration and other very important scientific and technical aspects of AAV gene therapy, the striking impacts of this therapeutic option as seen by our investigators and trial subjects. While the details of these observations remain private and confidential to the subjects themselves, our investigators described their subjects having a life transformation, a liberation from constant treatment and the burden of needing to think of access to Factor VIII everywhere and always and an ability to push their bodies without the constant fear of bleeding that transcends what can easily be measured in currently available quality of life instruments. This potential for deep therapeutic transformation is what we live for in BioMarin R&D, and we can't wait for this to become more fully realized in the months and years to come. Speaking of therapeutic transformation, I'd now like to discuss vosoritide. No program better exemplifies the 4 core attributes driving R&D at BioMarin, shown again on this slide. More specifically, first, the pathway stimulated by VOXZOGO is constitutively inhibited in achondroplasia by a genetic gain of function mutation of FGFR3 with widespread negative effects on bone growth. Second, the pathway itself shown here is, as Andrew Dauber will shortly describe, now seen as a canonical regulator of bone growth and is exquisitely pharmacologically upregulated by VOXZOGO. Third, the baseline constancy of mammalian growth facilitated early proof of concept with near normalization of growth in achondroplasia in a short time in a small number of subjects. And fourth, because the underactive endochondral growth plate results in essentially all the negative manifestations of achondroplasia, treatment with VOXZOGO throughout the growth cycle has the potential to have a transformational impact on many of the manifestations of this disorder. I emphasize throughout the growth cycle because this was uppermost in the mind of the EMA CHMP when they asked BioMarin for data that might allow them to extrapolate the striking results seen in our pivotal study of 5 to 18 years olds to the 2- to 4-year-old age range, with the opportunity to add 3 more years of impactful treatment to the fleeting lifetime opportunity to maximize growth outcomes for children with achondroplasia in Europe. The data we provided covered pharmacokinetics, pharmacodynamics and evidence of actual enhanced growth in sentinel subjects in the 2 to under 5 cohort in our ongoing randomized, placebo-controlled study in infants, toddlers and young children with achondroplasia. First, pharmacokinetics. What you see in this slide is area under the curve in the left panel, and maximal plasma concentrations in the right panel, comparing the 2 to under 5s in the red bars with the 5 and overs in the blue bars, and clearly showing virtually identical pharmacokinetics between the 2 age groups. Second, pharmacodynamics. We were so fortunate that the scientific community delivered a new validated biomarker of our core pharmacodynamic effect during our development program. That biomarker is collagen X or collagen 10, a bone matrix protein expressed by chondrocytes in the growing epiphysis, and CXM is a breakdown product of collagen X that can be measured in the circulation. Here, you see very recent data showing a very strong correlation of CXM with AGV measured across a range of AGVs in children with average growth. In our Phase II and III studies in older children, we saw a clear cut increase in CXM along with AGV with VOXZOGO that clearly overlies the correlation seen in this slide. As shown here, when we aggregated in a de-identified fashion, the CXM changes in the younger children in our ongoing blinded study, shown in red in the exposed and blue in the unexposed subjects in the 2 left-hand bars. We reproduced the CXM increases seen in active versus placebo in the older age groups in the 3 pairs of bars to the right, providing very strong evidence, predicting an increase in AGV in the younger children. Finally, the 2 to under 5 cohort in the under 5 study included 4 actively treated sentinel subjects to assess initial safety and PK. This slide shows the height of these 4 children followed for over a year, shown in the red dots, compared with box and whisker plots of untreated children at the exact same age and sex from our natural history database. As you can see, over time, all 4 children progressively and significantly increased their growth compared with their matched untreated peers. EMA accepted this data, emphasizing the importance to EMA of starting VOXZOGO treatment as early as possible to maximize benefit. Building on the strength of our science, extrapolating benefit risk to the 2 to under 5s and resulting in a full marketing acceptance in the EU from the age of 2 up, a result of which I and the entire R&D team at BioMarin are hugely proud.

Hello. I'm dropping in post-FDA approval of VOXZOGO, November 19. Obviously, all of us at BioMarin are thrilled for families who want to choose VOXZOGO to treat their children with achondroplasia. We are very grateful to our colleagues, investigators, patient advocates, and of course, the children who enrolled in our studies and their families for their support over the last 10 years through the development of VOXZOGO. And now I'd like to welcome back Dr. Andrew Dauber, Chief of Endocrinology at Children's National Hospital. Andrew presented at our last R&D day on the science underlying genetic short stature disorders, and has kindly agreed to provide an update on this science and his ongoing investigator-sponsored evaluation of vosoritide in this setting. Andrew?

Hello, everyone. My name is Dr. Andrew Dauber, and I'm the Chief of Endocrinology at Children's National Hospital in Washington, D.C. I thought I'd start by just giving you a short background of myself. I went to medical school at Harvard in Boston, and then I did both my pediatric residency as well as my fellowship in pediatric endocrinology at Boston Children's Hospital. And it was during my fellowship that I first started to do research into the genetics of short stature. I did some advanced training. I got a masters in clinical research, also from Harvard, and did advanced training in genetics at the Broad Institute and at Boston Children's. And my first research study was focused on identifying genetic causes of short stature. And during that study, we came across a gene called the NPR2 gene and found that quite a number of patients with idiopathic short stature had mutations in NPR2. And you may remember that NPR2 is the receptor for C-type natriuretic peptide that is targeted by vosoritide. So I was very interested and excited when I first learned that a number of companies, including BioMarin, were developing drugs to target this receptor. So today, I'm going to tell you about a clinical trial that I've been leading to try and bring vosoritide to a larger selection of children with genetic forms of short stature. You may recall that around 2 years ago, I came and presented at this meeting and discussed potential uses of vosoritide for dominantly inherited short stature. And my basic theory is that when you look at the height distribution of individuals in the general population, there are some individuals who at the extremes actually harbor underlying genetic mutations causing their short stature. And when you move to the endocrine clinic, where we see many patients referred to us for what's turned idiopathic short stature, more of those patients may actually have what we call monogenic causes or single gene causing their short stature. And furthermore, when you look at individual families where there's a clear pattern of inheritance from one side of the family, those families are either further enriched for genetic causes of short stature. While in the past 2 years, we've been very busy and made a lot of progress on this study. And in designing the study and after a lot of consideration, we decided to not just base the inclusion on these general clinical criteria, but to really focus on genetic disorders that we thought there was sufficient evidence to show that vosoritide could benefit these patients. So we've settled on a study that looks at 6 selected genetic classes of disorders, which lead to short stature. So pictured here is a chondrocyte. That's the cell in the growth plate that's responsible for making the growth plate elongate and making an individual ultimately grow taller. And this is going to show the schematic for these 6 classes. So the first one is hypochondroplasia. So you've all heard a lot about the use of vosoritide in achondroplasia. Hypochondroplasia is a similar rationale. It is due to a milder activating mutation in the FGFR3 gene, which will then signal through this Raf MEK MAP kinase pathway to inhibit growth of the chondrocyte. Another class would be individuals who have CNP deficiency. They're missing the CNP protein itself. Now this is very rare. We haven't found any of these patients yet, but they are natural candidates to benefit from vosoritide therapy. The next category are individuals who are heterozygous for mutations in the NPR2 receptor, meaning they only have one functional copy of the receptor that's targeted by CNP and vosoritide. And we and others have found that approximately 2% of individuals with idiopathic short stature are heterozygous NPR2 mutation carriers. And they don't have a complete loss of signaling through this pathway, but we think we can augment that with vosoritide therapy. The next includes a group of disorders called RASopathies who all have mutations in the central signaling pathway that is affected both by CNP and by those FGFR3 mutations. The most common RASopathy is something called Noonan syndrome, but there are others, including Castello syndrome, cardiofaciocutaneous syndrome and neurofibromatosis Type 1, amongst others. SHOX gene mutations. So SHOX is a transcription factor that's known to decrease FGFR3 expression and activity. And therefore, if you have a mutation in SHOX, you'll have more signaling through this pathway, again, inhibiting growth. And SHOX mutation carriers are also estimated to account for approximately 2% of idiopathic short stature. And the last category are individuals with aggrecan gene mutations. Aggrecan is a proteoglycan found in the extracellular matrix of the cartilage in growth plates. And again, aggrecan gene mutations have been shown to affect the signaling pathway with increased levels of this ERK signaling. So we thought there was very good evidence that all 6 of these categories may benefit from a precision medicine that could target this pathway. So the study design is a pretty simple study design. We're enrolling patients anywhere between ages 3 and 11 for males and 10 for females. They have to be prepubertal and have a height below minus 2.25 standard deviations, which is the approved cutoff for growth hormone for idiopathic short stature. They must fall into -- have a documented mutation in 1 of those 6 categories, and not have growth hormone deficiency or any other significant past medical history. They can't be treated with growth hormone during the course of the study, but prior treatment is fine. And we follow the patients initially for a 6-month period of observation so that we can get a baseline growth velocity. Since these kids all have different conditions, we really need to know how that individual patient is growing before we start therapy. And then we start them on a year of vosoritide therapy, the same daily injection dose that was used in the achondroplasia studies. Our primary study endpoints are, of course, first, the safety, looking at the incidence of adverse events in these individuals and then looking at efficacy by measuring the change in their growth velocity and height standard deviation scores. We have some secondary endpoints looking at body proportions and their bone age as well as we are studying some pharmacokinetic and pharmacodynamic markers as well as bone density and quality-of-life surveys. So as I said, we've made great progress in the last 2 years. We enrolled our first subject in August of 2020 and currently have 20 subjects enrolled in the study out of a goal of 35 subjects. Of those, 14 have hypochondroplasia, 2 have mutations in the NPR2 gene, 3 with Noonan syndrome and 1 with the mutation in aggrecan. They cover the whole range of the inclusion criteria between 3 and 11, with height standard deviation scores ranging between minus 2.25 and minus 4.38. Of the 20 subjects, 12 have finished that first 6-month observation period and have now started on vosoritide therapy. And of those 9 have already completed 6 months, so we are able to look at their growth velocity on treatments compared to the baseline. So I'm now going to discuss all of the details of the results of the study yet, although I hope to present this data at some of the academic endocrine meetings in the spring. But as a first pass, the medicine has been very well tolerated with a similar safety profile to previous reports. And our preliminary results suggest a very positive response in all of the subgroups. So for the patients with hypochondroplasia, which is the majority of patients, most of those patients have had a response that is as good, if not, significantly better than that seen in patients with achondroplasia. And for the 3 patients, the 2 with NPR2 mutations and 1 so far with Noonan syndrome, they have all had remarkable responses to date to vosoritide, suggesting that a precision medicine targeting that halfway that can be very beneficial for these patients. Of course, these are all very preliminary results. They only reflect small numbers and 6 months of data, but I'm very excited to continue to follow these patients and await data from the additional subjects in the 12-month outcomes. So in conclusion, it is very clear to me that the age has come for precision medicine to reach individuals with genetic forms of short stature. And I think that vosoritide or other medications that target the CNP pathway could be official for any condition in which there's a pathological increase in that ERK signaling which is found in, as I showed you, all of these subgroups and potentially other conditions leading to short stature as well. Thank you for your attention.

Thank you, Dr. Dauber. And now on a note of partial farewell, let me say that when I joined BioMarin 5 years ago for the final chapter of my management career, I committed to Hank to help get 3 to 4 of BioMarin's late-stage assets across the finish line, all of which I saw as innovative, transformational and each a master class in clinical drug development. With Brineura, Palynziq, and now VOXZOGO, all approved therapeutics and Valrox home straight, it's time to declare that commitment achieved. As BioMarin moves on with its pipeline build-out, I could not be succeeded by a finer or better experienced CMO than my colleague, Harold Bernstein. Thanks for the opportunity, Hank and J.J. And now please let me welcome Harold.

Thanks, Geoff. I'm Harold Bernstein, Chief Medical Officer and Head of Global Clinical Development at BioMarin. I joined BioMarin last month from a Boston-based biotech, where I was the Head of Translational Medicine and Vice President of Global Medicines Development and Medical Affairs. I've been asked why I chose to join BioMarin. It was a pretty clear choice for me. As a physician scientist and drug developer, with strong academic [indiscernible], there are multiple touch points that link my own experience to BioMarin's mission. First and foremost, I was drawn to BioMarin's approach to drug discovery and development, with its focus on the genetic causes of disease combined with a breadth of tools and technologies to advance the most innovative medicines for patients who have little or no therapy. BioMarin has always been able to do the hard work needed to transform the lives of patients, many of whom have been minor patients. BioMarin has a proven track record of developing drugs that have a real impact with regulatory and commercial success. That's what compelled me to bring my 30-plus years experience in translational and clinical development to BioMarin. And I'm especially grateful to my predecessor, Geoff Nichol, with whom I've been working to ensure a seamless transition during an exciting and busy time for BioMarin. Geoff will continue to provide significant institutional knowledge as we complete the ROCTAVIAN mission. In my role at BioMarin, I will have the opportunity to work closely with world-class scientists and colleagues to bring the pipeline into the clinic and through full development. Some of our paths have crossed in the past, and I'm excited that we'll be able to collaborate now. We'll be moving forward those assets that have the highest probability of success. We are also broadening our knowledge to expand potential indications through life cycle management. With Brineura for CLN2 disease, our clinical data is opening the way for us to tackle other genetic epilepsies. With Palynziq for the treatment of phenylketonuria, we are pleased to start the 12- to 17-year-old study in the first quarter of next year, with the goal of making this highly efficacious treatment available to the broadest possible age range. And as you heard today from Dr. Andrew Dauber, about his experience with VOXZOGO in other genetic statural conditions, we have the opportunity to leverage all the possible avenues to extract value from this program. For ROCTAVIAN, we have not seen loss of expression grade enough to warrant redosing, but we are aware of the potential future need for a readministration strategy. We're prepared, however, to be the first to determine if that need becomes relevant. And we're pacing our research investment accordingly. It is this portfolio, one that is rich with late-stage assets as well as the compelling pipeline of candidates that will pave the way for a new level of sustainability at BioMarin. The R&D engine is our business, driven by innovation to advance a pipeline of medicines matching the right modality to the right disease. This is another reason I was drawn to BioMarin. Other companies talk about using multiple modalities, BioMarin has already done it. We're at a unique time in the history of life science and drug development. By leveraging genetic discoveries and tools, one of BioMarin's 4 core attributes, we'll continue to develop targeted medicines that address the underlying cause of serious disease. Today, you will see the breadth of BioMarin's development pipeline, assets that span a number of exciting new therapeutic areas and over 20 candidates at various stages of research and early development. I am thrilled to join BioMarin at this important inflection for the company. I joined my colleague, Kevin Eggan, who came to BioMarin 1 year ago and who's the Head of Research and Early Development. Kevin and I share a deep appreciation for how BioMarin has and will leverage genetics and biology to continue building out one of the most innovative pipelines in the industry. I look forward to meeting and speaking with you in the future. I will now turn the presentation over to Kevin. Kevin?

Thanks so much, Harold. I'm sincerely looking forward to working with you as we move our pipeline forward in the coming years. Because patient safety is our highest priority at BioMarin, we take any potential safety signal from our nonclinical studies very seriously. We've, therefore, endeavored to fully understand the findings we recently reported from a rodent durability study of BMN 307 and carefully consider their potential translatability to patients. Now that we've had an opportunity to review these findings with the highly expert group of advisers made up of scientists and former regulators, I would like to take you through them in more detail. First, I'd like to update you on what we have found thus far. We conducted a study to understand the durability of BMN 307 expression over a range of doses in an immunocompromised double mutant mouse model of PKU. In the highest dose group of 2 to the 14th vector genomes per kilogram, we found evidence of 5 nonmalignant and 1 malignant tumor. To better understand the potential contribution of BMN 307 to this observation, we performed vector integration site analysis. These studies identified vector integration near several proto-oncogenes and some evidence for clonal replication of cells carrying these insertions. After obtaining these data, we immediately communicate that the potential safety signal to the FDA, which then place the BMN 307 clinical program on hold and requested more detailed information concerning our findings. Since our initial public communication, we've completed more quantitative integration site analysis, which have revealed that the expansion of these vector containing clones is not readily explained by vector integration serving as the primary driver for these tumors. In each tumor, we found far too little expanded vector DNA for each tumor cell to carry an insertion site. In fact, several of the tumors show evidence for vector insertion in only a very few percent of cells. These findings have led us to the initial conclusion that vector integration is unlikely to have been the initiating event for these tumors, and that instead vector integration or clonal expansion likely occurred following tumor formation. We have also recently completed whole genome sequencing as an orthogonal approach for examining vector integration. Our analysis of these genome sequencing data are in line with our quantitative integration analysis and confirm that only a small percentage of cells show evidence for integration. To pressor test our findings, we conducted a review of our data with an external expert advisory council, which endorsed our conclusions. They also suggested additional experiments to confirm the limited clonal expansion of vector integrates we observed, which are currently underway. Our conclusions as of today are that our findings are consistent with the broader literature that vector integration can occur at highly expressed genes following AAV administration, which was established and published well before the approval of gene therapies like Zolgensma. However, our results are not consistent with the notion that vector integration was the event that initiated tumor information in this nonclinical study of BMN 307. The broader takeaway around this issue remains that cancer signals like those in highly sensitive mouse models have not been observed following lifelong treatment of hemophilia dogs or in patients many years after treatment with investigational or approved AAV gene therapies. In the meantime, and shown here are the next steps for the BMN 307 program. We are concluding experiments along the lines I've described to interrogate the causes of the signal we observed and are continuing with our already robust safety monitoring procedures, which are a core component of all BioMarin clinical studies. Finally, we're continuing our conversations with FDA as well as global health authorities to provide them the information they've requested. We hope these data will convince health authorities to allow our PKU gene therapy clinical study to safely resume. In closing, insertional immunogenesis is, of course, an important topic for all gene therapies. We have been and continue to be transparent with the regulatory agencies regarding the emerging preclinical data from the BMN 307 program. We have seen EMA adopt positive opinions for several other gene therapies when cognizant of the theoretical risk of insertional oncogenicity. And we have no reason to believe that this issue would preclude approvability of ROCTAVIAN in Europe at this time. As I said, the emerging BMN 307 nonclinical data do not change the body of evidence that was available when the BMN 270 clinical program was initiated, and therefore, do not change the way we're looking at the importance of either our gene therapy platform programs and other gene therapy efforts in the field, including, of course, our own valoctocogene roxaparvovec. We believe that our experience with and investment in AAV gene therapy can continue to provide a sustainable competitive advantage as we attempt to bring transformational treatments to patients living with genetic disorders. That being said, part of identifying the best therapeutic candidate for a given condition is based on a comprehensive understanding of the current therapeutic landscape. This includes ensuring that patients can experience the full magnitude of potential benefit with an optimal route of administration. Also, for many of the conditions and targets we're pursuing, titratability and the ability to withdraw therapy may be essential for speed in clinical development and achieving an optimal outcome. These are important considerations as we seek treatments that not only halt disease progression, but also reverse existing maladies caused by a given condition. Over the course of developing 7 approved products, we have honed our expertise across multiple therapeutic modalities, including small molecules, oligonucleotides, biologics and gene therapy. As this slide describing the modality distribution, our current programs clearly shows, while we are committed to development of AAV therapies and will continue to be so, we intend to bring therapeutic candidates across multiple modalities to the clinic in the coming years. To do this, we plan to leverage the capabilities and expertise of our partners to fill gaps, navigate known headwinds and fortify existing capabilities for the efficient identification and derisking of promising candidates. In short, our proven expertise in discovery and development across multiple therapeutic modalities allows us to continue to exercise nimbleness in selecting the best approach for each indication, each disease mechanism and each patient population. To amplify the impact of the investments in our discovery and development efforts, we are increasing our focus in a few key therapeutic areas. It is our aim to maximally leverage our in-house capabilities and to focus the impact of those provided by our external partnerships. As you can see in this slide, the majority of our programmatic efforts are currently focused in 4 key therapeutic areas: heme and metabolic disease as exemplified by ROCTAVIAN and BMN 331 for HAE; musculoskeletal disorders as exemplified by vosoritide and BMN 351 for Duchenne's Muscular Dystrophy; cardiovascular disease is exemplified by our collaboration with DiNAQOR on HCM. We've also launched new efforts in nervous system disorders, really motivated by our experience in developing therapies for PKU and CLN Batten's disease. Looking forward in this space, we plan to build on those initial successes through an emerging effort to develop targeted therapies for the genetic epilepsies, which are a focus of our collaborations with Deep Genomics and the Allen Institute. In each of these areas, we will build upon existing expertise or toeholds that we've been building over the years, all are supported by experts who have been involved in the development and registration of therapeutic assets. As a result, we're confident in these carefully selected areas of focus. Looking ahead, we plan to build on this earned expertise through dynamic collaborations and partnerships that will bring new capabilities and assets to BioMarin, further building our momentum. Another useful way to view our efforts by therapeutic area is to examine the progress of assets in each area are making in clinical development. This is the slide that Hank shared with you earlier which gives you a glimpse of some of the more than 20 programs currently being pursued by BioMarin, many of which remain undisclosed. This pipeline graphic does not include important platform investments we continue to make or wealth of potential assets that are still being defined in earlier discovery stages. What I want to underscore with showing you this snapshot is that the dual strategies Hank communicated to you of tackling genetically defined forms of common disease and finding rare disorders that teach us about biology that can be harnessed for broader therapeutic utility have been extremely fruitful. With the ongoing discovery of genetic contributors to disease through ever-improving genetic methods, these strategies will sustainably bring efficient initiation of programs and significant value to our portfolio. We will be excited to continue to update this snapshot of a very dynamic process in which a study flow of inputs will continue to make its way on to the table. After a short coffee break, we'd like to give you a snapshot of 4 of these early development assets, which we haven't said much about previously, namely BMN 255, 331, 351 and 349. [Break]

Hello. My name is David Lomas. For the past 30 years, my research team and I have focused on the condition called alpha-1 antitrypsin deficiency. Indeed, in the early '90s, as a graduate student, I purified the abnormal protein from the livers of children affected by this condition. We work to have exactly how the abnormal molecules linked together. And in the intervening years, we've shown that this unusual protein-protein linkage account for a whole range of different diseases. Over the past 10 years, we focused on developing small molecules that prevent the abnormal molecules of alpha-1 antitrypsin linking together and so offer the prospect of a novel therapy for this condition. I'll tell you more about that now. So alpha-1 antitrypsin is produced from the liver. And it base all the tissues of the body. It's there to protect against enzymatic damage. 95% of severe deficiency results from the Z allele, which is the substitution of a positive lysine for a negative glutamic acid at position 342 in the polypeptide chain. Approximately 1 in 1,700 Caucasians of North European descent are homozygous, i.e., have 2 abnormal copies of the gene. That's about 100,000 people in the United States, with about 6 million people being carriers, having only one copy of the abnormal gene. In the homozygous, antitrypsin levels are low. That's because although the gene is normally transcribed and translated the abnormal protein is retained within the liver. This gives rise to 2 major conditions. The first is that the lack of circulating antitrypsin means that the lungs are exposed to enzymatic damage and so develop emphysema. And you can see this on an x-ray of a patient of mine with the bottom of the lungs being black as they've been destroyed by an excess of neutrophil last days. The retention of protein in the liver causes liver disease. And this is a 3-year-old child that we transplanted for antitrypsin deficiency. This liver at the top is shrunk and scar nodule. And you can see the areas of fibrosis with a healthy liver underneath. If we section this liver and then look at it under high power, you can see the abnormal protein, the abnormal antitrypsin stuck within the hepatocytes given rise to cell death and ultimately to cirrhosis. Well, I purified these inclusions, I showed that there weren't random aggregates, but were made up of chains of polymers. And these have a beautiful feeds on a necklace appearance that you can see here with each white blob on this micrograph antitrypsin molecule. Over the years, we've worked out exactly how one molecule is linked to its neighbor. And we've presented this a couple of years ago, showing this linkage, what we call a C sheet linkage between one abnormal death antitrypsin molecule and its neighbor. And of course, understanding the linkage gives us the opportunity to develop small molecules that can block these molecules binding together. If we look at the current therapies for antitrypsin deficiency, the lung disease is treated in some countries with augmentation therapy, replacement of plasma purified antitrypsin. But this is of limited efficacy and in many countries, including the United Kingdom, it is not licensed because the efficacy is not yet proven. There is currently no therapy for the liver disease other than transplantation for the most severely affected children and adults. If we look at the therapeutic landscape, Arrowhead have now partnered with Takeda to develop an siRNA that knocks down all antitrypsin production in the liver. This is our latest data taken from the EASL meeting in 2021, where they reported 2 cohorts of adults treated with siRNA that knockdown antitrypsin, and they reported efficient knockdown of monomer, polymer and some clearance of the inclusions within the liver. And I think this is impressive, but the problem with the therapy is that a single dose lasts for a long time. It can't be titrated. And of course, it affects all antitrypsin, whether it be M or Z, so it's impossible to treat heterozygous. We developed a range of small molecules that will bind to a cryptic cited antitrypsin near the Z mutation if they negate the Z mutation and they prevent the protein from undergoing [indiscernible]. We reported these molecules last year, and it's this series of molecules that have been licensed by BioMarin for translation through demand. The beauty of oral small molecules is that we can titrate the dosing, so we can give people dosing for shorter or longer periods of time. And also, they're likely to be effective in MZ heterozygous. Because we can pick out the abnormal set earlier, whilst leaving the wild-type M protein untouched. I'll now hand over to Dave Jacoby in California, who will give an update on our latest results.

I want to thank Dr. Lomas for the overview of both the disease and the program. As you heard earlier from Hank and from Kevin, BioMarin has expertise in multiple therapeutic areas and multiple therapeutic modalities. This is important because it lets us choose the appropriate therapeutic modality for the unmet need in the disease as we are chasing down early development leads in research. Dr. Lomas has given you a great overview of why choosing the right modality for the right patient population is an important concept for drug development at BioMarin. By leveraging the mechanisms of genetic illness, proximal in the pathway, BioMarin expects to seize large effect sizes. Effect sizes that will be transformational to our patients. This is what we refer to when we talk about targeted therapies and transformational effects Dr. Lomas has given us a real good head start in terms of understanding the mechanism and the rationale for this program. The disease is caused by a mutation that allows the protein to misfold and get trapped in the endoplasmic reticulum. This mutation is called the Z mutation in alpha-1 antitrypsin. The difficulty hear is the protein is trapped. It polymerizes, it forms aggregates within the cell and ultimately leads to cell death and cell stress. The deposition of protein polymers overwhelms the ability of a cell to degrade the misfolded proteins and the inclusions drive the pathophysiology of the disease. As we see in the panel on the left, BMN 349 is an oral small molecule that preferentially binds the Z-mutant protein and allows its transport through the endoplasmic reticulum and secretion into the plasma. As we can see from the slide, the cell then has the ability to degrade the existing polymers using its protein degradation pathways, resulting in liver regeneration and restoration of synthetic function of the liver. The characteristics observed in preclinical studies of BMN 349 include oral bioavailability, high potency, rapid onset and titratability of action. These have strong implications and the potential improvement of current management, particularly around the burden of treatment, medical management, population age and treatment of severe cases. This is particularly true for pediatric populations with alpha-1 antitrypsin deficiency. This is the most common genetic cause of pediatric liver transplants. And the ability to interact in a pediatric population with rapid onset shows the therapeutic advantage of an oral small molecule in this disease. This slide demonstrates the potent biological activity of BMN 349 in a mouse model of alpha-1 antitrypsin deficiency. This mouse expresses the Z-mutant form of the human A1AT protein and develops a molecular, cellular and clinical phenotype that is very similar to human disease. BMN 349 has been tested successfully in treatment paradigms early and late in the course of the disease in this animal model. If we look on the right of the slide, the histogram shows that there is already a deposition of liver polymers of the Z protein at baseline and vehicle treatment for 4 weeks results in a rapid and severe accumulation of further polymer deposition. This, again, is the pathologic driver of disease. If we look on the right 2 columns, BMN 349 treatment reduces the polymer burden below baseline levels to near normal, near absolutely wild-type levels. If we look on the left, we see liver histopathology of a mouse that has been allowed to live for 12 weeks and has a severe globule burden in their liver. And you see the resulting treatment reduces the polymer burden in liver histopathology slides dramatically. These 2 sources of data clearly show that 349 doesn't simply [indiscernible] disease. In other words, it doesn't just return the liver polymer burden to baseline, but reduces them below baseline levels showing that removing the formation of new polymers allows the liver to degenerate the existing polymer burden and allow livers to increase their synthetic and biological function. This slide demonstrates the potent biological activity of BMN 349 in the liver. This mouse expresses human Z alpha-1 antitrypsin protein and develops a molecular cellular and clinical phenotype that closely resembles human disease. BMN 349 has been successfully used as a treatment paradigm early and then late in the course of liver disease in this mouse model. The histogram on this slide shows the treatment paradigm in 5-week old mice, in which BMN 349 is given daily for 4 weeks. At baseline, there is already deposition of liver polymers suggesting the onset of disease with vehicle treatment for 4 weeks results in rapid accumulation of liver polymers. And as we'll see later, deficits and synthetic function of the liver. Treatment with BMN 349 reduces the polymer burden in the liver below baseline levels to near wild-type levels. If we look on the left panel, we see the effective treatment in mice that have been aged and have severe liver polymer deficits in their liver. So this histopathology shows the reduction in liver polymers that occurs with just 4 weeks of treatment in elderly mice. These 2 sources of data clearly show that 349 doesn't simply halt the disease progress, but treatment allows the liver to clear existing polymers. So this data shows that treatment doesn't just bring the polymer level down to baseline, but actually reduces it to near wild-type levels. Beyond reduction in aggregation of polymers, we want to see that the overall liver health has improved. The next 2 slides show the treatment results in reduction in cell stress, an increase in normalization in liver function associated with BMN 349 treatment. The histogram on this slide shows a marker of endoplasmic reticulum stress in mice that are developing the disease. We see at baseline that their ER stress levels are above the wild-type levels in animals and at the 4-week vehicle treatment period results in a significant increase in the amount of ER stress. Remember, this is a protein that hangs up in the ER and preliminaries at that site. BMN 349 treatment reduces the level of ER stress below baseline and near wild-type levels in untreated mice. Implicit in this data is that the mechanism of Z protein stabilization results in significant biological changes rapidly within days to weeks of the onset of treatment. This is a time frame in which genetic therapies have yet to reach maximal effect. So far, we've shown you that BMN 349 treatment and a relevant mouse model of disease results in the reduction in the pathologic driver of disease, the formation of polymers in the liver. And it reduces the amount of ER stress seen with the ongoing course of disease. We'd like with treatment to be able to show that liver function is restored, and this slide demonstrates that as we interrogate liver function by measuring clotting factors synthesized in the liver, the BMN 349 normalizes liver function going forward. Here, we show clotting factors measured at 5 weeks of age that are reduced compared to a wild-type control. And the 4 weeks of vehicle treatment also reduces the synthetic ability of the liver to below baseline levels. However, when we treat with 2 doses of BMN 349, clotting factor synthesis has increased above baseline levels to near wild-type levels, restoring the normal liver function. So we have seen that the mechanism of action reduces the pathologic insult, the pathologic drivers of disease and allows the liver to restore normal cellular function. As a potent, rapidly actable and titratable medicine, this is the logical treatment in situations in which modulation of dose and rapid onset are key determinants of clinical success. An example here is the ability to intervene in infant cases in which the titration and time to onset may be key attributes of treatment. As we know, alpha-1 antitrypsin deficiency is the most common genetic cause of pediatric liver transplant, and the ability to intervene rapidly may have important clinical outcomes. As shown here, we're currently finalizing the form and formulation of BMN 349 and plan to initiate IND-enabling studies next year and be in the clinic in 2023. The clinical plan will be centered on a rapid proof of concept in adults with alpha-1 antitrypsin deficiency, but will rapidly pivot to populations for which the oral formulation and titratability may have unique and differentiating use, populations of which the modality and the time to affect have therapeutic advantages. I'd like to also point out that this is an example of BioMarin's pivot to more prevalent disorders. There is an estimate of the ZZ prevalence of alpha-1 antitrypsin deficiency greater than 250,000 individuals in the world. And as Dr. Lomas pointed out, the ability of this molecule to act preferentially on mutant proteins allows us to think about applications in patients with liver disease that are heterozygous for the Z mutation, which is a much broader, larger population. The principles in leveraging genomic understanding for a targeted approach that is ratably accessible in clinical trials and transformational to patients does indeed apply to all of our programs. Further, genetic insights into the mechanism of illness provide evidence of effectiveness in more common disorders driven by similar pathophysiology. Our next program, BMN 255 for the treatment of hyperoxalurias, exemplifies this principle. The genetic form of hyperoxalurias are quite rare, but they inform the medical risk of other forms of systemic oxalosis. Shown on the left panel of this slide are the morbidities in children with primary hyperoxaluria type 1, PH1, which is the most severe genetic form of the disease. Excess oxalate exposure to the kidney results in the formation of microcrystals. This sludge in the collecting apparatus of the kidney results in progressive renal damage. This has a positive and negative feedback loop in which the renal damage allows less oxalate excretion and drive secondary deposition into tissues such as the bone, heart, retina and skin. Kids with PH1 suffer from progressive renal disease, but they also have pathologic bone fractures, cardiomyopathy, loss of vision and skin ulcers. Historically, BioMarin has developed medicines that are replacement therapies, but this pivot demonstrates how genetically informed discovery and therapeutic applicability go beyond the genetic form of the disease as oxalate has been clearly identified as a primary risk factor for both renal and metabolic diseases beyond the genetic forms. BMN 255 is a potent oral small molecule that blocks the formation of oxalate. On the panel of the left in this slide shows the metabolic formation of oxalate and the mutation that causes primary hyperoxaluria type 1. The loss of function of the AGXT transferase results in formation of oxalate rather than glycine. BMN 255 works upstream in this pathway. It's a reversible and potent inhibitor of glycolate oxidase. It removes the upstream substrate such that there is no oxalate formation, even in the genetic forms of this disease. Plasma and urine glycolate here are now measurable biomarkers of target engagement and inhibition and inversely related to urine oxalate excretion and oxalate formation. BMN 255 has undergone extensive preclinical investigation and models of activity and safety. The totality of evidence shows a potent enzyme inhibition, a subsequent block in all of the metabolic oxalate formation, a rapid onset of action and kinetics that support once-daily dosing with a clear safety margin. The data in this slide shows the increase in urine glycolate as a result of BMN 255 treatment in a mouse model of genetic hyperoxaluria. There was a large increase in a dose proportional fashion in urine glycolate, which occurs within 7 days of the onset of treatment with significant effects seen at day 3. This increase in urine glycolate is associated with a reduction in the oxalate formation in the kidney and reduction below wild-type levels showing complete inhibition of metabolic oxalate formation. In sum, BMN 255 is an oral small molecule that inhibits the formation of oxalate. It is potent, has a rapid onset of action and is titratable to effect. High levels of urine oxalate excretion is a risk factor for renal disease progression and has systemic complications. As noted in our discussions of BMN 349 for AATD, an oral therapy with titratable and measurable effect with onset in days may have significant benefits in the setting of acute pediatric or more severe forms of hyperoxaluria. BMN 255 is currently clinically enabled in a Phase I/II study now being assessed in safety and biomarkers in healthy human volunteers. We plan to transition to hyperoxaluria cohorts in the second half of next year for similar assessments of multi-dose safety, pharmacokinetics and measurements of biomarkers that show target engagement and inhibition of oxalate formation. Hyperoxalurias, again, are much more common than genetic forms. If we look at estimates from the Oxalosis and Hyperoxaluria Foundation, there are at least 250,000 patients with oxaluria in the United States. I'd like to switch gears and give a brief snapshot of the BMN 331 program for hereditary angioedema. Once again, this program exemplifies BioMarin's ability to match the therapeutic modality to the patient population with a genetic targeted approach. This will be studied with protocols reliant on readily accessible biomarkers, clinical endpoints that are relevant to the disease. And by intervening in the critical genetic disease pathway, we expect to have a significant effect size and a transformational benefit in patients. The patient experience with hereditary angioedema informs our pathway for clinical development for gene replacement therapy. The left panel here shows dramatic pictures of acute angioedema attacks these attacks are unpredictable, unprovoked and can be severe and life threatening. Those with HAE are limited in activities of daily living, at work, at school and at home. They face symptoms of anxiety and depression and have insights that they face potential life-threatening complications with laryngeal edema asphyxiation. The literature constantly reports very low health-related quality of life indices with HAE. Current therapy is effective at reducing the frequency of attacks. Yet, there is a constant burden of repeated injections and 60% to 70% of patients still have recurrent breakthrough attacks with the risks of severity and hospital visits remaining intact. Furthermore, half of patients report injection site pain or reactions. Patients, therefore, have to choose between the burden of therapy with chronic injections and are stuck in a place where they still have the risk of recurrent attacks, potential life-threatening consequences and hospital visits. The goals of development of gene replacement therapy is to address the burden of the continued need of chronic therapy and to address the burden of the risk of repeated recurrent attacks. HAE results from mutations that cause the underproduction of C1 inhibitor protein, which leads to loss of control of kinin homeostasis and the activation of the bradykinin system. It's the activation of the bradykinin system that results in unpredictable acute attacks of angioedema. BMN 331 is an AAV5 liver-directed gene replacement vector. Reconstituting the expression of C1 inhibitor allows control of the bradykinin system at multiple control points so that we're augmenting the biology that is missing by the mutation. C1 inhibitors made in the liver, it's secreted into the plasma and plasma levels are measurable and relevant to the likelihood of clinical efficacy. Small increases in the levels of C1 inhibitor are predicted to have large clinical benefits. And the ability to constantly express C1 inhibitor may result in a higher proportion of patients that are attack-free. This data is taken from the clinical experience of prophylactic administration of recombinant C1 inhibitor protein in HAE subjects. This correlates the relative risk of an acute attack with trough levels of C1 inhibitor activity. Bear in mind, levels of C1 inhibitor in patients typically vary between 25% to 35% of wild-type levels. Hence, this curve shows a relative risk of acute attack at about these levels. If we can augment the levels to about 65% of total C1 inhibitor activity, this confers a 90% reduction in the risk of acute attacks. This 65% of C1 inhibitor level becomes a meaningful personalized target for clinical development. This curve also demonstrates that the risk of attack is not completely addressed by repeated troughs with current therapies. We see that even at 100% of activity, there is still some conferred risk of attack. We propose the constitutive expression with gene replacement therapy at levels above 65% will increase the proportion of patients that are attack-free. In animal preclinical studies BMN 331 has shown a robust dose response of C1 inhibitor expression at doses predicted to normalized C1 inhibitor activity in humans. Here, we show that the mid-dose results in the synthesis of C1 inhibitor that would restore greater than 65% activity in HAE patients. Remember that the amount of C1 inhibitor that we need to make in order to augment the levels to 65% are about 30% to 40% of novel expression. Here, we also show super therapeutic levels of C1 inhibitor that are expressed at the high-level dose. Importantly, these mice had no safety consequences, including no risk of thrombotic events. These studies support BMN 331 can express C1 inhibitor levels that will allow patients to be attack-free over a long and robust period of time. BMN 331 has an open IND, and we will enroll our first patient in the near term. The protocol will assess safety and C1 inhibitor levels with a decision tree to expand or escalate based on reaching near therapeutic levels of C1 inhibitor activity. We expect that the assignment of dose and the expansion and extension will occur in 2023. Again, to reiterate, the goals of development are to alleviate the burden of chronic therapy and the fear for the risk of breakthrough attacks. As we have seen in our clinical experience in the hemophilia program, the development of new agents that increase the time intervals between doses are useful, but do not change the underlying paradox of the burden of chronic therapy and the risk of recurrent medical episodes. Our goals for 331 are that we can reduce the burden of chronic therapy, but also increase the proportion that are attack-free. With that, I would like to reintroduce Kevin Eggan, who will summarize our Duchenne program at BioMarin.

As a scientist, it's rewarding to work for a company like BioMarin, which closely adheres to supporting programs that exemplify its core principles. We focus on using genetics to identify targeted therapies, which can be readily assessed in the clinic and are designed to transform patient lives. These core programmatic attributes inspire us and guide the curiosity of our research teams to important, yet tractable problems. They invite our scientists to think about fresh opportunities presented by emerging discoveries, like those recently created through the rapid progress in decoding the genetic factors underlying the epilepsies and cardiomyopathies. But they also challenge us to revisit widely studied genetic conditions where major medical unmet need remains unmet and a new approach to targeting a well understood disease biology might bring transformative impact for patients. It's very much with this last motivation and inspiration in mind that I'm excited to share with you the science behind our new IND candidate for Duchenne's muscular dystrophy, BMN 351. As a reminder, Duchenne's muscular dystrophy is a devastating disorder caused by loss of function mutations in the dystrophin gene. Due to its X linkage, Duchenne's disproportionately impacts boys, who present with a rapid decline in muscle function that confines them to a use of a wheelchair by the time their adolescents. As they become young men, boys with Duchenne's increasingly suffer from respiratory and cardiac distress, which generally takes their lives before they turn 40. While there have been substantial efforts to develop new therapies for these boys, they're still waiting for a transformative medicine. Restoration of dystrophin expression offers an attractive and targeted strategy that can be quantitatively assessed in muscle biopsies. Genetic studies of boys with the less severe Becker muscular dystrophy suggest that restoring even 20% of near full-length dystrophin expression could transform the life of a boy with Duchenne. Our aim with BMN 351 is to surpass that. With sustained dystrophin expression above that benchmark, we feel clear improvements in motor, cardiac and respiratory resilience will surely follow. Since our initial attempt to register drisapersen, we've really thought carefully about the lessons that we've learned from our experience and have been quietly laboring on research efforts to develop a next-generation oligonucleotide with greater efficacy for inducing exon 51 skipping. ASO in the field sophistication and understanding the mechanisms that govern exon inclusion in a final mRNA have advanced, we began to appreciate that early thinking, which led the field to the site what drives a person and eteplirsen act, might not have led us to the best opportunity for acting with an ASO to induce the exclusion of exon 51. This realization guided by emerging knowledge of the sequence specificity of RNA binding proteins that regulate splicing guided us to an unappreciated and very potent splicing enhancer far from the intron exon junction for all other molecules in Clinical Development Act. Tests with many comparable oligos around these 2 sites support the notion, the new enhancer we identified can play a 100-fold more important role in regulating splicing than the site where all other molecules in development currently act. Our discovery of the superior biological approach for modulating exon 51 skipping is the basis for BMN 351. We are so excited to collaborate with the Duchenne's community to test whether BMN 351 acting at this novel site can improve dystrophin expression while employing well-understood phosphorothioate chemistries. As I will now detail, we are investing in 351 because our preclinical data suggests that restoring as much as 40% of dystrophin expression might be within reach. This level would far surpass our programmatic goal of 20% dystrophin restoration. And if achieved, would place children in the lower end of the range observed with a normal muscular phenotype. We anticipate restoration to this range would provide a transformative effect on muscle function. Our excitement for developing BMN 351 emerges from our experience evaluating dozens of oligonucleotides in the humanized mouse model of exon 51 addressable Duchenne's. Again, results from the team's experiments consistently show that in vivo, BMN 351 far surpasses the performance of many other oligos acting at the original exon 51 site, increasing dystrophin expression by more than 100-fold over molecules like [indiscernible] in short-term treatment paradigms. Here on this slide, I'm pleased to share with you the results of the long-term study in which we chronically administered BMN 351 in dose response and compared it to eteplirsen. Over the sustained treatment period, we observed that the highest dose of BMN 351 completely restored normal dystrophin levels. We again found in this study that BMN 351 provided a very significant improvement in dystrophin expression over at a person. As you can see, even substantially lower doses of BMN 351 either outperformed or provided comparable restoration of dystrophin to that observed with the highest dose of eteplirsen. Experiments like these in pharmacological modeling, which is based on our experience with phosphorothioate oligonucleotides, have informed our goal of restoring very substantial levels of dystrophin in Duchenne boys. Our belief that restoring significant levels of dystrophin will transform the lives of these boys is largely informed by human genetic studies, but they are even further fueled by the phenotypic improvement that we observed in these humanized DMD mice treated with BMN 351, and that's depicted here on this slide. As you can see, humanized DMD mice show gradual disease progression as measured by an automated composite score of motor performance. Treatment of these animals with a high dose of eteplirsen did not protect them from this motor decline. In contrast, animals treated with BMN 351 were significantly protected from this decline and even showed a modest improvement over the baseline. Needless to say, these preclinical findings show improvements in muscle function that clearly differentiate BMN 351 from first gene therapy is to target the initial splicing enhancer. We found that the high dose, which completely restored dystrophin in mice, was well tolerated by NHPs during our short-term dose range finding study. And our long-term GLP tox study is currently underway and slated for completion by the end of the year. We intend to have our first engagement with the FDA concerning our program in the first quarter of 2022 and are targeting an IND in the second quarter with the first patient enrolled thereafter. As Hank outlined to you, a core component of our pipeline strategy going forward is to tackle recently identified genetic forms of common diseases like the cardiomyopathies and genetic epilepsies. In order to efficiently and rapidly bring pipeline assets forward for these conditions, we are growing our approach to collaboration and partnerships. Today, I'd like to introduce you to 3 exciting collaborations we've initiated since our last R&D day and that are showing very exciting signs of progress. Collaborations with Deep Genomics and the Allen Institute for Brain Sciences are bringing new technologies, new approaches and new promising assets to our neuroscience therapeutic area, while our ongoing collaboration with DiNAQOR is focused on gene therapy for treating genetically defined forms of cardiovascular disease. In recent years, we've been collaborating with Invitae to use targeted exon sequencing to identify children who would benefit from treatment with Brineura. As you can see in the upper right-hand corner of this slide, our experience with this effort has informed us that a surprisingly high portion of children that are diagnosed early in life with epilepsy have a genetically defined form of that condition. Our careful analysis of this space indicates that these forms of epilepsies are well suited for targeted therapies, which could transform patient lives by putting these children back on a normal developmental trajectory. Our own experience with BMN 351 has really informed our thinking concerning the importance of the subtlety of identifying mechanisms for therapeutically regulating gene expression in children like these. We were really impressed upon our first encounter with them by Deep Genomics' efforts to deploy the field's ever-growing gene expression data sets with their machine learning methods to identify new opportunities for therapeutically regulating gene expression. These proprietary approaches and their expertise in setting human neuronal systems as well made them attractive partners, and we're currently pursuing 4 programs with them across genetic forms of epilepsy and ataxia. We're particularly excited for our lead program, which is currently on track for an IND in 2024. Another feature of the epilepsies that have drawn us towards developing treatments for genically-defined forms of this condition is the synaptic function of the genes that are altered in many of these children. We're particularly interested in targeting these genetic forms of epilepsy because they offer great potential for disease reversal if addressed early in life. One of the challenges with correcting the underlying cause of disease in these children is the exquisite selectivity, the expression of the affected genes show in very specific subtypes of cells in the nervous system. Restoring their function appropriately will require an equally targeted approach in order to be successful. To achieve this, we've embarked on an exciting collaboration with the world leaders in understanding the regulation of gene expression in the brain the Allen Institute. Together, we're harnessing Allen's discovery of genetic elements that regulate cell type specific gene expression to develop 4 highly targeted gene therapies for nervous system disorders. We've already achieved a stunning proof of concept in this area with plans for first gene therapy IND in 2025. Finally, I want to update you on our collaboration with DiNAQOR. As we work to efficiently and effectively advance multiple programs in genetically defined cardiovascular disease, we have benefited from access to DiNAQOR's human engineered heart tissue platform for functional characterization of gene therapy candidates. In particular, we've made strong progress in our program for hypertrophic cardiomyopathy caused by mutations in MYBPC3. We have recently selected our clinical candidate vector. And as you can see in the upper right, IV administration of this vector in the early experiment delivered correctly localized expression in more than 70% of cardiomyocytes in the mouse heart. We are moving forward with IND-enabling studies and anticipate this being our next gene therapy IND in 2023. Over the last 3 years, we have completely pivoted our pipeline, and evolved the R&D organization to more efficiently and expertly deliver high-quality and high-value INDs. As you can see in this projection, we are currently on track to deliver on our corporate goal of an average of 2 INDs per year in 2023. With line of sight to multiple INDs in each therapeutic area, we are already realizing synergies between programs within each of our therapeutic focus areas. Momentum is gathering in our early pipeline where our scientific strategies are proving to be a rich substrate for nominating and advancing programs. Thank you for your attention. Hank back to you for summary remarks.

Let's dive in. We're making a lot of progress Valrox and enabling patients to have the best possible outcomes from gene therapy for hemophilia A. We're bringing VOXZOGO to the market with expansion opportunities into both younger patients and broader populations. We believe that insertional oncogenesis is an important finding, but not an important issue in the development of AAV therapies. Debate remains. Is integration a cause or simply a coinciding observation in mice? Reassured that no cancers are currently associated with AAV gene therapies. We acknowledge that the significance of these findings remains uncertain. The pipeline has pivoted. We have added more gene therapies, but we've also added other modalities to our portfolio of product developments with the idea of creating a sustainable pipeline and further compelling opportunities. Partnerships expedite, sustain and augment our existing capabilities to continue feeding our growing pipeline. And it's with great excitement that I turn this over now to J.J.

So thank you, Hank. And before I begin my closing remarks, I want to thank all BioMarin employees for their contributions to building the leading rare disease company in our industry. Having successfully pivoted the pipeline, our future growth is grounded in applying the same rigorous approach to more common conditions that we successfully established with the ultra-rare disorders. As you have just heard, the R&D engine is BioMarin's business. And with our global infrastructure, with our established capabilities in place, we are poised to drive significant growth through the consistent introduction of innovative and transformational products for patients. Leveraging our proven capabilities with 7 commercially available products, we will use our deep foundational know-how to pursue the development of medicines to treat larger genetic diseases. With a track record of discovering, developing, manufacturing and ensuring access to our essential medicines, the path to addressing more presently affected genetic conditions is before us. Our focus on thoughtful investments as we turn the corner towards GAAP profitability balances the plan to reinvest in our highly productive R&D engine with a focus on continuing operational excellence will be leading to meaningful shareholder returns. From where we stand, the future looks very bright. We have faced headwinds over the last 2 years that have helped refine our focus and mission. With 2 key approvals in the last 3 months, VOXZOGO in Europe and in the United States, we achieved what we said we would. The demonstrated clinical benefit of VOXZOGO for children with achondroplasia kept us focused during unpredictable times. We hope to have the same outcome with ROCTAVIAN soon. With potentially 2 consecutive major product launches in 2021 and 2022, we are posed to sustainably grow while sustainably building out the next generation of innovating products, as you have just heard. Our track record of success, deep foundational knowledge, the financial flexibility to pivot the pipeline successfully, all combined to put us on a path to accelerate drug discovery and developments, product approvals and successful commercial launches. So this is how we view our journey to date and to the future state of BioMarin. As you learn today, having successfully pivoted the pipeline, we can now leverage our approach to studying rare genetic diseases to address larger genetic conditions and multiple therapeutic areas. Kevin described the multiple toeholds in our defined areas of therapeutic focus, which proved a range of opportunities, mainly with direct overlap to our currently approved or late-stage programs. The therapeutic areas are established in the call it rectangles on the slide and below are the cone of opportunities, which show our existing toeholds and below that, the additional indication, which we touched on today. There are many other opportunities under each therapeutic area, a number of which are currently being explored. Each class of assets in our portfolio can be linked to numerous other indications. For example, other genetic short statural conditions as Dr. Dauber shared with us today. In the CNS category where we have Brineura approved for CLN2, we have an open gateway for broader pediatric genetic epilepsies. The leading indication may be ultrarare, like CLN2, but the adjacencies are all higher prevalence indications as shown by the prevalence histograms at the bottom of the slide. The core of what we do is not changing, but our expertise, capabilities, targets and impacts are changing with the aim to provide transformational therapeutic options to address larger and more numerous genetic conditions. Not only are the opportunities to expand the pipeline based on the many unmet needs across the therapeutic areas I just described. but it is BioMarin open approach to using the right modality for the right indication that enables maximum flexibility and innovation. As described by Kevin earlier, our experience with BMN 307 does not change the way we are looking at ROCTAVIAN or gene therapy in general. Let's say, we are not monolithically committed to gene therapies as shown through the distribution of our portfolio across modalities in the image on the left. So we remain focused on multiple modalities, and we continue to exercise nimbleness in selecting the best modality to -- for each mechanism. We have the opportunity to leverage our internal capabilities and expertise as well as partner with others to fill gaps and share risk and costs. Product identifying the best candidate for a given condition is based in a comprehensive understanding of the current therapeutic landscape, including the criticality of route of administration, the role of compliance in ensuring that patients experience the full magnitude of potential therapeutic benefits, titratability and dose response, therapeutic window and the ability to potentially reverse the damage of a given condition. Sustainably innovating and growing has been the theme of the day and translating genetic discoveries into transformative medicine is and will always be BioMarin's mission. Having successfully pivoted our ultra-rare pipeline to larger indications, BioMarin is uniquely positioned to harness our expertise to address more prevalently affected genetic conditions. We are fortunate that many common diseases are genetic in origin. This is enabling BioMarin to renew the next-generation pipeline through genomics, as Hank described earlier. With our research now focused on broader expanse of therapeutic areas and the opportunity to leverage our deep foundational knowledge, we are set up for exponential growth, as you have heard today. The financial proof points that support our accelerated growth plan are in place. On the left of this slide, our actual financial results from our established base business, which have been steadily improving while we have concurrently invested in significant opportunities like VOXZOGO and ROCTAVIAN. What we see to the right of the slide is our vision for growth over the next few years. Leading to the middle of the decade, with revenues, contributions from VOXZOGO, the only approved treatment for achondroplasia and potential revenue from ROCTAVIAN beginning in late 2022, we would expect to realize operating profitability next year. In addition, continued fiscal discipline and the reduction of expenses as a percentage of revenues will lead to expanding operating margins. This growth is driven by the innovation and productivity from our R&D engine, as you have heard today. Our focus going forward is on sustainably innovating to accelerate growth of the business. We want to thank you all for your support, and we will now take your questions in a live Q&A session. Thank you for your attention today.

The patience that has played an essential role and BioMarin's ability to develop innovative and transformative medicines. Thank you both for participating in today's R&D Day. To my colleagues, Geoff Nichol, Harold Bernstein, Kevin Eggan and Dave Jacoby, thank you for your contributions to the content share today. The presentation shared today provide just a glimpse of the scientific strategy being executed at BioMarin to drive pipeline renewal and sustainability. We hope you'll agree that the expansion of the pipeline, as described today, presents numerous opportunities across multiple therapeutic areas, and we'll now open the event to questions and answers. Please e-mail Traci, if you would like to ask a question. Starting now with the very first question and regarding BMN 255, there are other programs that are more mature than 255 in hyperoxaluria and target the disease with other mechanisms, specifically, for example, RNA interference. Describe the unmet need that you see in the market that drove the decision to proceed with this program? Dave, maybe you can kick us off there.

Sure, Hank. Happy to. So I think one of the things that we tried to emphasize in the presentation about 255 is that the genetic form of the illness really allows us to understand what the pathophysiologic drivers is in a broader patient population. So we know from the genetic form of hyperoxaluria that hyperoxaluria is an independent and severe risk factor for progressive renal disease. We also know that hyperoxaluria is a very prevalent disorder. We gave the example of the enteric forms in the presentation. But if we look, for example, in recurrent stone formers, there's about a 10% lifetime risk of having a renal stone, 30% to 50% of them have recurrent stones and about 25% to 45% of those have hyperoxaluria, so that we know that hyperoxaluria is not only a risk factor for progressive renal disease but also recurrent kidney stones. So we think there's a large prevalent disorder for which inhibition of the formation of metabolic oxalate is going to be a key potential contributor to disease modification. In terms of the question about mechanism of action, I think it's a theme in all of the disease -- the programs we discussed, 255, 349 and 331 that we think that route of administration and mechanism of action are going to be key competitive differentiators in developing medicines for these patient groups. For hyperoxaluria, 255 is an oral small molecule. It acts directly at the inhibition of the formation of metabolic oxalate. There's real value in working at that particular target rather than the genetic knockdown therapies, which currently exist on the market. Those types of interventions lead to a degree of heterogeneity response. And obviously, if you can't completely knockdown the expression, you're going to have enzyme formation that does lead to a subsequent formation of oxalate. Here, we have a small molecule. It works on the enzyme. We have nonclinical evidence to suggest that we can dose to completely inhibit the enzyme activity. And importantly, we have a plasma biomarker that allows us to dose to effect, which is something that the oligonucleotide knockdown therapies cannot do. So we have a titratable oral molecule, we can dose to effect, we can completely inhibit the enzyme, we can reduce the formation of metabolic oxalate in these diseases. And that's a clear differentiator, we think, going forward in development. We also have another oral go inhibitor that's in about the same stage of clinical development as we are. And so it is a prevalent disorder. We have a lot of confidence in the pharmacologic activities of BMN 255. It's very potent. It's bioavailable. It's going to be, I think, safe and tolerable in the clinic. And so we're really confident in the therapeutic potential of BMN 255 in this disorder.

Thanks, Dave. And when you talk about complete knockdown, I think I've heard you talk about the proportion of patients that can be completely normalized with interfering RNA therapy. Can you just remind us about that?

Yes. So the data in OXLUMO, their pivotal study shows that at their prescribed dose, only 43% of patients were actually able to achieve urine oxalate excretions below the upper limit of normal. So normalization only occurred in about 40% of the patients taking that dose. So 60% were reduced, but they weren't reduced into the normal range. We know from the genetic forms that even small amounts of hyperoxaluria lead to a risk of progressive renal disease. So there is a therapeutic advantage in completely knocking down urinary oxalate.

So that's pretty exciting. So there's definitely room for doing better in the hyperoxalurias. Okay. That's great. So switching gears a little bit, VOXZOGO seems to be the first opportunity for BioMarin with one product to address and many potential indications with numerous opportunities to extract value. I first want to congratulate my colleague, Geoff Nichol, and also send a big congratulations to our colleague, Brad Glasscock. Our clinical and regulatory teams just did a spectacular job. Now of course, they are supported by hundreds, if not many hundreds of BioMarin employees. But while I had the chance, I wanted to thank Geoff and Brad for the outstanding work they did. And it gives me also an opportunity to put Harold in the hot seat and ask Harold, give us a game plan for the future of -- how are you seeing things unfold?

Yes. Thanks, Hank. I'm happy to. So we're laser-focused on delivering VOXZOGO to patients and children less than 5 years old as well as obtaining global approval for what we believe is a really transformational medicine. We're very excited with the results that we've had. And as you know, the EMA recognized this with an approval down to 2 years old. We're also very interested in helping to define the additional health benefits that come with treating the skeletal dysplasia with our CNP analog. This is something that's important not only to us but to patients. And so we're working very, very hard on that. Thirdly, what you heard from Andrew Dauber earlier, we're very interested in the potential of using VOXZOGO to treat other genetically informed statural conditions. This is a great opportunity for us to look at the FGFR and C-type natriuretic peptide pathways in a much broader way. And it's something that Kevin and I have been actively discussing. Finally, I just want to note that we do have active research ongoing to address the convenience of administration. So I think that not only for VOXZOGO, but this is really a great example of how we're looking at programs across the portfolio to really extract the most value for patients and really bring them the maximum benefit.

So this is really cool because the 255 program, genetically enabled large metabolic problem in the population. Vosoritide genetically proven at this point, but amenable to address a very large fraction of the population. So I'm starting to see the threads that we -- get us really excited about how to leverage genetics into larger and more transformative medicines. Let's shift gears a little bit again. And Kevin shared -- Kevin, you shared that 307 results are not consistent with the notion that vector integration was the event that initiated tumor formation in the study that was put us on hold. Can you walk through the next steps, additional experiments needed to have the hold removed and why you believe the oncogenesis finding is unrelated to BMN 307 specifically?

Yes, certainly. So we've reviewed the information request from the FDA and have prepared our responses. We intend to submit them at the beginning of the year, and we hope to be off clinical hold by the end of the first quarter. Now of course, we're still awaiting our discussion and engagement with the FDA. So we remain uncertain about precisely what will be required. But they have not, at this time, requested further experiments. Now one of the reasons why I have increasing confidence that we will come off of hold is that our careful quantitative analysis of the tumor samples indicates that, in most cases, only a few tumor cells, a few percent of tumor cells in each one of those samples actually shows evidence for integration. If integration was a driver event for these tumors, we'd fully expect all or the vast majority of cells within those tumors to carry a clonal integration event. That is to say, we'd see that in the vast majority of tumor cells the same integration event in all of those cells. But interestingly, it's not the pattern we see. Instead, we see just a small number of cells have those events, suggesting that after the tumor began to form the vector may have hopped in. As a result, we're still looking for the causes of transformation within those animals, and we remain uncertain as to whether or not they will translate into humans. As a result, we've remained focused on the safety studies within those individuals as part of our normal clinical development processes. So from here, we're really awaiting that first interaction with the FDA following our responses, and I'm sure we'll be providing more information as soon as we've done that.

And in regard to the potential for read-through to ROCTAVIAN I think, Kevin, in your prepared comments, you made some reflections and it might be worth recapitulating how we think the 307 results to read on the 270 applications, both in Europe, where we are in process in the FDA where we are yet to be in process.

Right, right. This is a very important issue. It's clear that the entire gene therapy field is rightly focused on this issue of integration. And thus far, we've not had any interactions with the European health authorities that would make us believe that this is an issue that would hold back the approval of ROCTAVIAN. We continue to transparently provide them information concerning our ongoing studies of 307. And we intend to do that as we move into our pre-submission meetings with the FDA, and I'm sure following those meetings, we'll be eager to share their further feedback. But again, if one looks back over the course of the summer, at the FDA AdCom, while there is focus on this issue and a clear desire to see long-term follow-up of patients who have received gene therapies, we don't believe it's an approvability issue at this time.

And one of the things that I thought was encouraging in some of the dialogue that I heard was the nature of the questions being asked by the FDA and the EMA make it look like they are working together to get their arms wrapped around this issue. I don't know if you wanted to add anything to that.

Yes. I think having had a chance to engage with the EMA through the day 120 from the ROCTAVIAN process and also the information requests, I would say there's a very strong overlap in the questions that they're asking. And so if there's not direct communication, there certainly is very strong, aligned thinking on this issue. And I'm looking forward to their feedback on our responses early in the new year.

Okay. Let's see if we can take another question. I think we've covered this question, the potential for read-through. So let me shift gears now again a little bit more, and I want to go to the readout of the Phase II 0- to 5-year-old study. But first, I want to also congratulate the team for getting the 2- to 5-year-old label claim in the European approval. I think that was a better than expected, whereas the FDA was an expected outcome. Having said all that, I also want to extend congratulations to the team because I believe we're on file in Australia already. I believe we're on file in what's the second or third -- third largest pharmaceutical market, Japan. And I'm very proud to say that we've obtained an approval in Brazil. Now as Geoff like -- our colleague, Geoff likes to caution us all the time, it takes a long time to go from approvals to reimbursements, but it's certainly a fantastic positive step. I think it reflects some innovation by the resilient health authorities, and I think it represents an alertness by our regulatory team. So with that, and Harold gave us a nice outline of the overall plan for VOXZOGO. But Geoff, the 0- to 5-year-old study, give -- maybe you can give us an update on where that is.

Yes. So Hank, as you said, that's designed to take us below the age of 5, which was the lowest age that we treated in our large Phase III study and extend actually all the way down to those young children over 2, which I've already talked about, as well as toddlers and even down to neonates and babies. And it's taken us a while to advance that study. It's now fully enrolled. And you need to proceed carefully when you're studying young children. They're not just the same as older children, necessarily. And we did find in the toddlers and babies that there was a change in the relationship of exposure of vosoritide to dose. And those children required a slightly higher dose than the 2 to 5s and older children. And we certainly anticipate that any sponsor who's wanting to take their product down into those younger age groups is going to have to go across that hurdle as well. That being said, very happy to say that, that study is fully enrolled. We anticipate breaking the blind and delivering the top line results in the first half of next year, submitting our package to health authorities in the second half of next year and hopefully an approval in 2023.

What's really exciting to me about that, Geoff, is that I think we're the only entity that has started to treat children 0 to 2 years of age, and you've completed the enrollment in that cohort. That's pretty...

Correct. Yes.

That's a great accomplishment. We're way ahead there.

And we're also treating in a separate study very young children in the first year of life who have foramen magnum narrowing, which is a major problem in the very first months of life for children with achondroplasia. That's enrolling well. So we're also very optimistic for the outcome of that study.

And I think an underappreciated fact about infants with achondroplasia is that the sunrise mortality rate for those children is something like 50x the age-adjusted normal children, average stature children. So that has the potential to be a huge and profound benefit. So I can't wait for that result to come on the plate. So shifting gears again. How does BMN 331 gene therapy compare to currently marketed standard of care products, which is a monoclonal antibody? What does a potential gene therapy product offer in terms of better overall effectiveness? Dave?

Thanks, Hank. So I think, again, one of the themes that we talked about in the presentations is being able to tailor the modality of administration to the current unmet need in the patient population. And we feel that although there are currently good therapies for the prevention of attacks in HAE, there's still significant medical unmet need that might be addressed by gene therapy. There's a clear burden of therapy that comes with chronic injectables. And we can see that patients are looking to maybe switch to oral inhibitors, kallikrein inhibitors, even though the efficacy data in clinical studies doesn't look nearly as strong. So we think that a single administration therapy that expresses C1 inhibitor, so the actual product that is missing in the genetic form of the disease, really still has massive potentials in terms of treating unmet need in HAE. C1 inhibitor works at multiple biological checkpoints in the bradykinin pathway. So we think that there is potential that constitutive expression might have a more durable -- and it is a feature of the prophylactic therapies for HAE that there's still a fair proportion of attacks that are moderate to severe. And there are still a number of patients for which they don't have disease-free survival. So they have still a few attacks a month. So one of the things that we're looking at is whether constant expression of C1 inhibitor -- and C1 inhibitor itself because it works at multiple biological checkpoints, can actually have both a durable, but it may be a potentially more therapeutic benefit in terms of proportion of patients with moderate to severe attacks and proportions of patients that have disease-free or attack-free survival for long periods of time.

Great. So lots of reasons to believe that the gene addition with C1 esterase inhibitor is going to provide a more profound benefit to the patient. But I wanted to focus in on something you mentioned. So attack-free is a better physiologic state to get to, and also, treatment-free is a better physiologic state to get to. And I wanted to spend a second on the physiologic -- on the pharmacology-free aspect of this. And I remember coming into work one day or Zooming in one day. And Geoff was very excited. For those of you who aren't as intimate with these sorts of things, Geoff is the industry rep on the Cell and Gene Therapy FDA Advisory Committee. So he gets to sort of see and participate in really important discussions that happen there. And one day, I came to work and Geoff was all exorcised about the advisory community that was going on. And so, Geoff, I wonder the -- if you can talk to us about that Cell and Gene Therapy Advisory Committee Meeting that you went to, and what emerged in terms of the relevance of sparing rescue medication. What did committee present? And what was recommended?

Well, that was an interesting one. It seems maybe a little orthogonal to what we're doing at BioMarin, but actually, in some ways, landed right sort of in the center of what we're doing with several of our gene therapy programs. The product at issue at the April Cell and Gene Therapy Advisory Committee was actually pancreatic islet cells that are sort of infused into the liver, and they had been trialed in brittle diabetes, this form of transplant. And it turned out that the product did not actually improve the brittle diabetes. It did not actually either improve hemoglobin A1c compared with the insulin treatment that the subjects in the study were on. But the FDA reviewer noticed that a very significant proportion of the subjects were basically free of the use of insulin. They were insulin independent. So they had gained control of their type 1 diabetes through this transplant without the need for any further insulin treatment. And that is to the heart of the matter. And I think I mentioned earlier, in the context of ROCTAVIAN that this is one of the striking benefits that the subject in our ROCTAVIAN studies speak to, that sense of freedom from the shackles of constant need for therapy. And that's something that we think is extremely important and lies behind several of the programs that we are advancing. So thanks for [indiscernible] that up, Hank.

Tremendous. Yes. Yes, yes. And so you can really see the read-through to 331 where you have the same physiologic state. You have a person who's desperately fearful of having an attack but desperately bothered by the unpredictability of it and the need for chronic medication, unbuckling yourself from that and having confidence that you're not going to experience an attack is really, I think, what I hear from opinion leaders is to the gold standard of therapeutic outcomes of gene addition and C1 esterase inhibitor. And certainly stands up why the benefit of ROCTAVIAN is seen as so profound by the patient community, namely, being not just bleed-free but independent of prophylactic therapy. It's really a tremendous advance for people. So let's move back to Duchenne muscular dystrophy, an area of near interest at BioMarin, BMN 351, different from our prior work and product for Duchenne's. So why do we think we can win with this next-generation target? Kevin?

So Hank, obviously, since our experience with drisapersen, we've reflected a lot on our experience and what we've learned from that. And I think the field has advanced quite a lot as well. And as we look back on those early molecules like drisapersen and eteplirsen, we recognize that when those molecules were selected, really, our understanding of how the dystrophin gene is truly regulated. It was very limited. And for those drugs, really, the first site that happen to be identified for regulating the splicing of exon 51 with the goal of restoring dystrophin in boys with exon 51 addressable mutations was based on that first glimpse into the regulatory biology. Since then, the field has really gone in a couple of different directions. One has been to try to take that early understanding around the regulatory site where drisapersen and eteplirsen work and change the chemistry of molecules or their conjugation to increase uptake into muscle. And I think the field will learn a lot from that. But of course, there are also uncertainties within that as well around new risk profiles that can arise, around uncertainty about whether or not those conjugated molecules will be released after they go into cells and access the nucleus. And we have been watching that carefully, of course, and trying to learn from that, but we ourselves took a distinct approach, which was to step back and say, is there something more that we can learn about the regulation of the dystrophin gene? Is there an approach that we can take to better improve this skipping of exon 51 with the chemistry who's strengths and liabilities we really understand and can work around? And I'm really happy to say that our team succeeded in doing that. We identified a site within the dystrophin pre-mRNA which is very important for regulation of inclusion and exclusion of exon 51. And we've designed oligonucleotides which target the regulatory biology of that RNA in a way that is distinct from all other oligonucleotides that are in the clinic today. That site seems to have an extremely profound effect on the regulation of exon 51 inclusions in multiple assays that we've run, both in human muscle cells as well as in vivo in human -- in humanized transgenic models. I'm really excited about that because what we found particularly in those humanized models is that the biology of drisapersen actually seems to translate very well between the original model and the human patients. That is to say, if you look at the amount of skipping that we get per microgram of drisapersen within that humanized mouse model, it's actually very similar to what was observed in the pivotal Phase III studies. Now what's really exciting about this new regulatory biology is that when we obtain the same concentration of BMN 351 within this humanized model of rodent, DMD, we actually see 25% skipping within that model -- or sorry, 25% dystrophin expression within that model relative to just the 1% or 2% that we see with drisapersen. And so based on that, we're quite hopeful that if we can just achieve that same concentration with BMN 351 that we did with drisapersen that we might also achieve 25% dystrophin expression in patients. And it's very clear from the genetic studies of Becker muscular dystrophin that really anything over 10% expression could be transformative. And with BMN 351, we have the opportunity to go well beyond that. And I think we're seeing excitement within the community around that possibility and I think excitement around being able to test a complementary hypothesis about dystrophin biology that could be coupled with the work that's being done in other chemistries and conjugation.

So the -- my oversimplified understanding of control of expression came from the lac operon, and I think there have been some updates since then. So is this like a fluke that the different exons have different regulatory control over gene expression? Or is -- do you think this is a...

No. No, no. Not at all. I mean if you look at just the breakthroughs that have been made in sequencing technology within the last decade, it's given us this incredible quantitative view into how RNAs are regulated in cells. And the complexities of that, I think, are staggering. And while we're making great progress in understanding the huge numbers of proteins that act within the nucleus to regulate the metabolism of an RNA and to turn it into a fully mature transcript, we're finding that computers are starting to do a better job than people are of predicting outcomes. And that's really one of the reasons why we've turned to Deep Genomics now as a collaborator. Now this journey that we began from the lessons learned from drisapersen to better understand regulation of RNAs to restore dystrophin expression is now something that we're really expanding in many of our programs throughout the portfolio. We've established a great team of [ gelics ] experts and genomics experts, which are really helping us with that, and now teaming up with Deep Genomics is giving us an opportunity to say, what can deep learning approaches, AI approaches bring to the conversation for better understanding these events? And we're seeing already in a short period of time real benefits from that. If you look back, it took really 3 years to go from these lessons that we were beginning to learn around drisapersen to the identification of this new regulatory site and substantiation of early molecules that we could begin to develop. In just 9 months, we've gone from target selection with Deep Genomics to the identification of large numbers of regulatory opportunities for therapeutically modulating the genes of interest that we're targeting together in the genetic epilepsies. And in a very short period of time, we've been able to now substantiate the activity of those in the first target across multiple mechanisms. And what we're seeing is that that's much more fruitful, much more efficient than just tiling across a gene, looking for activity, and it's giving us an opportunity to look at places that we never would have thought to look in the past. And I think that's incredibly exciting for the future of this area.

I've said to colleagues that you are the human equivalent of machine learning. And I'm looking for -- Kevin made the mistake once of telling me what his dream was, which was to particularly understand the circuitry of control of gene expression throughout the body. And I'm so excited that that's going to be what we're working on together. I think that's going to be an amazing contribution to patient care. So we covered Deep Genomics. Now let's move to a related sort of topic of interest is the Allen Institute and our collaboration there. And Dave, I know you've been a particular champion. Again, I remember the day you came in and said, wow, this is really exquisite, and I was like trying to keep up with you there. But tell us a little bit more about why you're so excited about the Allen Institute collaboration.

Sure. I think that CNS disorders have sort of been the last frontier of drug delivery. And one of the reasons is certainly the fact that it's an anatomically sequester tissue. But the other -- and we actually have great experience to showing you can directly deliver therapeutic molecules to the CNS. So we're beginning to overcome that kind of anatomic barrier. And so now, there is a complexity of the tissue that needs to be addressed. You can't just put a gene into the brain and expect that it's going to have the therapeutic benefit because it needs to go to a specific type of cell and have a specific regulated function. So the Allen Institute is a type of science generator that has basically curated huge amounts of information on cellular phenotyping, transcriptional regulation, transcriptomics and the functional capabilities of neurons throughout the brain. And so now, we have the ability of understanding how we can deliver genes to specific types of neurons. Actually, we're taking delivery out of the equation a little bit because we can deliver genes now to multiple cells, but we can express the transgene in specifically regulated, very tightly regulated cells. So it's not just putting it into an inhibitory neuron or excitatory neuron. It's the actual subtype of neuron for which we're going to have large and transformational clinical benefits. So the collaboration with the Allen Institute has given us this access to understanding how we can bring in regulatory elements that will allow us to tightly and specifically control gene expression in a number of genetically defined CNS disorders. But it goes beyond that. That kind of like knowledge database is going to be helpful with our Deep Genomics collaborations and with the oligo platform that we have because now we're learning about isoform expression, temporal expression, cell type-specific expression that will enable us to also design oligos with the Deep Genomics collaboration that will be much more specific to the genetic deficit that occurs in these diseases than you could with just sort of blunt instruments and just an intrathecal administration of an oligo. So the Deep Genomics and the Paul Allen Institute, these are kind of like all now matrix information databases that will allow us to think really carefully and thoughtfully about which modality, which cell type, which disease we can address with these types of new CNS therapeutics.

That's amazing. Just amazing. So let's now go back to BMN 349. So how is that different from knockdown candidates currently under development, Dave?

Well, this is analogous to the 255 discussion. Again, the mechanism of action here is very different than the oligonucleotide-mediated genetic suppression. Here, the target is the actual protein, the mutant protein that's made. And I think the reason why there's a real advantage to targeting the protein rather than the gene expression is, again, this kind of idea that we can dose to effect and that we can clearly dose to a level where we're getting all of the protein that's mutant and being deposited pathogenically into the liver, out of the liver. Whereas we know that there is a range in the response to genetic suppression when you use an oligonucleotide. Some people get 90% expression knockdown, some people are 80% or 70%. That just means that 70 -- sorry, 20% or 30% of the mutant protein is being made and deposited in the liver. And so we think that there is real value in having a mechanism that selectively binds the mutant protein. The other real key component here is that we've got a plasma biomarker to follow, so that we can not only titrate the dose with a neural molecule that has a rapid onset of action, but we can titrate to effect because we can follow the plasma monomer as it's being sequestered by BMN 349 and secreted out into the plasma. So again, we have an oral molecule, but we also have a mechanism of action that gives us the ability to dose to effect and probably a modality that allows us to go into patient populations for which that titratability is really a key component. So for example, we know that alpha-1 antitrypsin deficiency is the single most common genetic cause of pediatric liver transplants. And so one can imagine in a pediatric population that has portal hypertension at risk for transplant, having something that you can titrate to effect and is rapidly active, would be a major therapeutic benefit. So we really, again, think that the reason 349 is really a compelling development opportunity for BioMarin is the mechanism of action, really targets the pathologic driver of disease.

So that -- it's a little bit analogous to the vosoritide experience in the sense that you have this very young, very severely affected, in the case of vosoritide, it's the foramen magnum population; in the case of the alpha-1 population, it's the kids who are destined to either undergo liver transplant or experience lifelong and severe liver disease. On the other end of the spectrum, when Professor Lomas was first talking to me about the concept that was behind 349, he mentioned something about MZ, which, at the time, a little bit, I was like, wow, that's like way too big. Maybe you could say a little bit about the rationale of why this molecule in something as large as heterozygous patients could make sense?

Certainly. So again, one of the features of BMN 349 is it selectively binds the mutant protein. So it doesn't need to be a homozygous liver disease patient. It could be a heterozygous liver disease patient. We can test the hypothesis that the Z protein in heterozygous with liver disease is really a pathologic driver because we can selectively bind that protein and clear it from the liver. And that is the carriers of Z protein is really a large patient population. And there's a lot to learn about what that is as a risk factor for subsequent liver disease going forward.

That's very exciting. My eyes almost bugged out of my head when Professor Lomas set up, but now I'm getting a little bit more conditioned to this concept that these molecules with these unique properties can be leveraged into much larger indications. You talked about 255 before. We're experiencing that with vosoritide. Now we're talking about alpha-1 being in that direction as well. So that's pretty exciting. Let's now shift a little bit more and talk about potential for retreatment in gene therapies. It's an area of some interest, and I think people would be keen to hear Kevin with what are our thoughts currently about retreatment.

Well, so first of all, we're still watching with great interest the levels of Factor VIII expression in patients in ROCTAVIAN clinical studies. And so we are still waiting to learn how long that expression will last. But we certainly acknowledge that in some patients, levels are gradually dropping. And so we're mindful of this, and we've launched a number of exploratory programs in order to be able to provide an opportunity for redosing if that is eventually needed for some patients. Broadly speaking, that's -- that activity is broken down into 2 areas. One is that we've had for many years, at BioMarin, an extensive capsid discovery program. We, to my knowledge, have now isolated and filed intellectual property around more capsids from more sources than any other academic or a private group. And we're using that new resource judiciously to identify capsids that are immunologically distinct from AAV5. And it's our hope that some of those will provide an opportunity for patients. But I'm also very excited about the experience all of us have had from the shots that we've had in our arms around COVID-19, about the promise that nonviral activities -- or nonviral means of delivering nucleic acids are happening -- having. Now we're deeply committed to the AAV platform at BioMarin, and we will continue to invest and drive new programs into that space. But in order to balance out the opportunity in the genetic medicine space with an increasing number of INDs per year, we have to balance our investment across multiple modalities to allow those INDs to go forward efficiently and judiciously. And probably many have read about the collaboration that we just announced with Entos for nonviral gene delivery. And we have other programs underway. And certainly, administering Factor VIII through gene transfer with these nonviral approaches is an exciting way to think about overcoming the existing immune response post treatment.

Harold, hold the line. So far, we have not had a lot of need for retreatment. So let's hope that we can keep this in the Kevin research workshop space.

Yes, I'd be more than happy to shut this down.

Now -- so that now is a good pivot also then to a collaboration with DiNAQOR. And before we get into the question here, I did want to give a brief shot-out to our colleagues, Johannes, Valeria and Thomas, who have been just outstanding partners with us in the DiNAQOR collaboration. So Kevin, maybe you can give a thumbnail on where we are in development because I think an amazing amount has been accomplished?

Absolutely. As someone who has worked for most of my career in developing human cell models of disease, I was deeply impressed and very excited as I came into BioMarin about the collaboration with DiNAQOR. They've developed really an exciting human tissue model, which physiologically recapitulates many aspects of heart contraction and when genetically manipulated can recapitulate aspects of pathological contractions. Of course, they were very logical partners for our work on MYBPC3 genetically mediated forms of HCM. One of the things that I've certainly learned from my own work in the past, and we see here at BioMarin, is that although it's very useful to have in vivo models in animals in order to look at certain physiological activities of the treatment, the factors that regulate gene expression are often very unique to the human-differentiated cell type that, that gene functions in. And so to have the opportunity to actually use functional cardiomyocytes of a human source to test which vectors give us the best expression was really exciting and a great collaboration with them. Now with them, we've been able to select our candidate vector and associated capsid. We've also finalized the selection of the production scheme, and we're well on our way to generating tox material with the hope that we will have an IND in 2023.

I want to give a huge shot-out to our technical operations colleagues in Greg Guyer's group because they've really rallied around enabling rapid progress of this program. I think it's been fueled by the work that you just mentioned. And I'm really thrilled that Harold joins us as a cardiologist stem cell biologist. So boy, are we in a lucky position for that program. So congratulations, and thanks for all that progress. Geoff, another exciting day when you came in and you said to me, I think I got a game plan for how to talk the Europeans into VOXZOGO under 5. And you went through some of that data, but maybe we can go in a little bit more clarity about specifically the contribution of Cmax on growth outcomes versus its contribution to safety. And how do you view the strategy around the next generation of CNP projects?

Okay. Well, the first question, good question, very short answer. Cmax and AUC, which will, let's call it, exposure, highly correlated, and those 2 both highly correlated to dose, except, as I already mentioned, in the toddlers and infants for a given dose, you get somewhat less exposure. So you need to adjust the dose in that age group. The relationship of exposure to growth, whether we measure that by AGV or whether we measure it by collagen X, is clearly exposure-related, but it plateaus at exposures round about the dose that we're using in the majority of children of 15 micrograms per kilogram, and there appears to be no greater benefit above that level of exposure. In terms of safety outcomes, cardiovascular effects, we don't see them at the exposure ranges that we've looked at in our dose-finding studies and in the Phase III study, doesn't look like we are significantly on the dose response curve with that yet, with really essentially a flat response in terms of any relevant clinical effects that would impact safety. Where we're going next, I'd perhaps throw that back to Kevin or Harold. What's next?

What's the biologic instinct?

Yes.

Yes. Well, I mean, I think the genetics is really teaching us the profound effect that genes in this pathway have on human height and skeletal biology more broadly. And so I think, really, the work that we've been doing has been to do the nonclinical work to substantiate and support the type of clinical studies that are ongoing by Andrew Dauber and others and that we intend to follow up with a robust program here. And maybe you want to say a word about that, Harold?

Yes. So as we talked about before, we have this opportunity with VOXZOGO to really explore the converging pathways between CNP signaling and FGF receptor signaling. And it turns out that there are a number of experiments of nature that manifest as diseases that refer back to those pathways. So not only are those diseases in front of us as potential therapeutic opportunities, but it's also -- there's a way to leverage that, and this is the kind of thing you and I have been talking about, to think about other diseases that are within that pathway and how we can use the analogs that we've been inventing to potentially approach those.

It's an exciting thing for BioMarin in the sense that we have this achondroplasia opportunity, and then we have these other opportunities to develop medicines. And we clearly have to do right by our achondroplasia patients, and we want to seek the right pathways for patients with other forms of mutation or other statural conditions that might be amenable. So this is going to be some really important work that we get to in the first part of the year now that VOXZOGO has passed both the U.S. and the EMA regulatory hurdles and, as I mentioned, is starting to accumulate success around the globe. So we -- Kevin, you talked about the latest on PKU gene therapy from an FDA response perspective, and the thumbnail on why we're optimistic is there's really nothing new in terms of what we demonstrated as compared to what was discussed at the FDA Advisory Committee. Actually, through 2016, basically, these types of results have been seen in the field, and we see continued clinical investigation approval, licensure decisions, et cetera. So let's assume that we get back on track in the clinic, Harold, what's next in the clinic for 307?

Sure. So as Kevin stated earlier, and I just want to reinforce, I think we're in a good position to answer the questions posed by the FDA, and that we will go off of clinical hold. Our experience to date, we essentially met the criteria for phenylalanine lowering with our experience to date. And so we're now ready to expand the cohort at the 6e13 dose. And we're poised to do that as soon as the clinical hold is released.

Good. We'll have our skates on. This next question has led me to want to wear sunglasses around BioMarin, the future being so bright that I have to wear sunglasses. But the question is, okay, in the past, we've discussed an IND a year, but now we have these 24 undisclosed programs, which, by the way, are all BioMarin in their fundamental nature. So what's the thinking around INDs per year? Kevin, do you want to start?

Yes. Sure. Well, we remain on track within the stated LRP of having 2 INDs on average per year by 2023. And if the programs continue to go forward in the way they are, we're quite excited on delivering about that. But we're mindful that the desire of shareholders to see growth is substantial. And we also see that the opportunities that are laid out by these 2 core strategies that you described at the beginning of today's presentation, Hank, are very, very rich. And so to try to take advantage of that opportunity, we feel urgency to move forward in that space. But we're doing it with discipline by having multiple modalities to be able to deploy, so that we're not overloading any portion of our manufacturing group as we have more INDs per year. And we also will have these therapeutic area focus teams, which will allow us to have multiple teams working in parallel as a strategy for gradually increasing the number of INDs per year beyond that.

And I think one of the things that we've talked about is sort of this notion of a prospective understanding of biology and forming the expected outcomes of the clinical study. And if we see them, we want to fund that and progress aggressively. If we don't see that, we have to be fairly disciplined about saying, well, we didn't have the biology exactly as we thought we did. Or at least in humans, it's more complicated. We can send that back to the drawing board and let's not have that opportunity cost get in the way of advancing one of the other 23 ideas in the portfolio. So that discipline that your teams are going to be wielding is going to be really, really, really beneficial.

Absolutely. And so in all 24 of those programs all have as a characteristic an aspect of their biology where we can readily assess very quickly in the way that you described the outcomes. That's going to be exciting to collaborate with you around that, Harold.

Well -- and Dave, you mentioned this, like I don't think that people thought that you could put a protein into the brain directly and have as profound a benefit. And so given Batten's disease and gene therapy experience and the fact that we have a very rich CNS program, you made a reference to this briefly in the previous answer, but maybe say a little bit more about what we've learned about delivery, and how that's going to inform this huge panoply of preclinical ideas that we have.

Well, sure. I mean the brain is anatomically sequestered, but we've learned an awful lot, both with our clinical experience with Brineura in terms of delivering enzyme replacement therapy to the brain, but also, there's now a new generation of therapeutics in terms of oligos and gene therapy treating diseases like spinal muscular atrophy. So we think that we're really at the frontier now of thinking about how the biology and specific biology can really unlock a whole panoply of genetic CNS disorders.

That's amazing. And because I remember at that particular time in the development of Brineura, people are administering the medications, enzyme replacement therapies through the lumbar intrathecal space, and we were the weird outliers who decided, let's go where the money is, like the top of the brain, and that turned out to work really well. And so for conditions which are severe enough, it does make sense that [ Berthold ]. And we've had good many years of experience by now at the Brineura children.

Absolutely. Absolutely.

A lot of reassurance that, that can be accomplished.

No. It -- a brain is a challenging place, but actually, the clinical experience with Brineura is that giving 300 milligrams of enzyme every 2 weeks is remarkably well-tolerated. And I think the lessons learned here is that we -- again, we match the route of delivery to what the needs of the disease were, and we didn't compromise with distribution going intrathecally to the patient actually and to the surgeons having something that was permanent, and the ventricles turned out to be a clinical approach that was well-tolerated. But we got enzyme to the areas of the brain for which we really needed to get it there. And we'll take the same approach with gene therapy and with oligonucleotides. We need to really get the therapeutic to the right parts of the brain, and we'll think about creatively what's the route of administration to do that.

And I think the thing you mentioned earlier about the sort of secondary insurance policy is that by targeting specific either promoter enhancer elements in brain tissue as unfolded by the Allen Institute collaboration or specific regulatory elements for those specific tissues as identified in the Deep Genomics collaboration gives us that further shot at selectivity and specificity of control of expression. Another question. What is the likelihood the FDA may want to see the results of the 303 steroid study before filing and approving ROCTAVIAN? Well, we haven't had pre-submission discussions for the next cycle of review. And what I can say about that is it hadn't come up in the previous cycle review. And as you know, we're well into the European cycler review, and this hasn't specifically come up as an area of concern. I think the way that we look at the prophylactic steroid study, as Geoff outlined, is that its main purpose is to try to simplify where possible the patient's journey through the early phases of gene therapy. So hopefully, this will be an improvement in the regimen more than it is something essential to demonstrating a positive benefit. As we talked about before, the benefit here is really fairly significant in terms of what it offers for patients. So I think it's -- we're coming up on the last question, very last questions here. The BMN 351 for Duchenne, how are we doing toxicologically, guys? Are things progressing there?

Yes. So I can give maybe a quick update there. We've completed a dose range finding study within nonhuman primates and suggests that BMN-351 is in those studies very well tolerated at doses that go beyond our therapeutic target dose. And one of the things that's been very gratifying, as many people know, phosphorothioates, and in particular, drisapersen, were associated with complement activation in humans. Now we don't see that even at high concentrations of BMN 351 in human serum or in nonhuman primates, which are sensitized species. This has informed and enabled our long-term tox study, which actually should be reading out very, very shortly. And we expect to have that data by the first quarter of next year, which will enable an IND at the -- in the summer next year in 2022.

Great. Great. Well, with that, it's a good opportunity to ramp up, and nothing makes me happier than talk in science, and so you guys have made my day. In fact, let's stay around for another 3 hours, and we can just continue the conversation. So -- but for those of you who are out there, I want to thank you all for your attention today. The R&D organization has never experienced more momentum, opportunity, excitement for what lies ahead. We look forward to keeping you apprised of our many future clinical development events over the coming quarters. Thank you very much.