Good day, ladies and gentlemen, and welcome to the First Quarter 2018 Financial Results and Corporate Update Conference Call. [Operator Instructions]. As reminder, this conference is being recorded. I would now like to introduce your host for today's conference, Danielle Keatley. You may begin.

Good afternoon, and welcome to the Ultragenyx financial results and corporate update conference call for the first quarter of 2018. We have issued a press release detailing our financial results, which you can find on our website at ultragenyx.com. I am Danielle Keatley, Senior Director of Investor Relations and Corporate Communications. With me today are Emil Kakkis, Chief Executive Officer and President; Shani Sharp, Chief Financial Officer; and Jayson Dallas, Chief Commercial Officer. I'd like to remind investors that this call will include forward-looking statements within the meaning of the safe harbor provisions of the Private Securities Litigation Reform Act of 1995, including but not limited to the types of statements identified as forward-looking in our 2017 annual report on Form 10-K that was filed on February 21, 2018, our quarterly report on Form 10-Q that'll be filed soon and our subsequent periodic reports filed with the SEC, which will all be available on our website in the Investors section. These forward-looking statements represent our views only as of the date of this call and involve substantial risks and uncertainties, including many that are beyond our control. Please note that actual results could differ materially from those projected in any forward-looking statement. For a further description of the risks and uncertainties that could cause actual results to differ materially from those expressed in the forward-looking statements as well as risks related to our business, see our periodic reports filed with the SEC. I'll now turn the call over to Emil.

Thanks, Danielle, and good afternoon, everyone, and thank you for joining us. The beginning of 2018 has been a pivotal moment for us with the approval of Crysvita in the U.S. and in Europe. I'll spend some time discussing Crysvita and our other recent progress before turning over to Shalini for an overview of our first quarter financial results and then to Jayson Dallas for an update on launch progress. On April 1, the FDA approved Crysvita for the treatment of X-linked hypophosphatemia or XLH in adult and pediatric patients 1 year of age and older. We were pleased to receive a broad label, which we feel is important for commercialization in both pediatrics and adults. To highlight the importance of label, I'll go over the key elements that have value for the launch. First of all, the indication statement text includes all XLH patients 1 year and older with no modifiers or limitations around severity or type of disease. This is the simple broad indication statement that we thought with respect to the fact that XLH is a lifelong disease, and the data shown is indicated for treatment for both pediatrics and adults. And this will be important in reimbursement decisions. Second, for pediatric patients, the label include improvement in rickets and serum phosphorus as expected, but it also includes data demonstrating improvement in volume and growth, which are both important assets of pediatric disease and are particularly not well treated by oral phosphate therapy. Third, for adults, the strongly positive serum phosphorus data is again labeled, but the FDA also included data showing a significant increase in healing of fractures and pseudo-fractures, a result which has never before been achieved effectively in, really, any disease. We greatly appreciate FDA's recognition of the importance and clinical means of these…

Thank you, Emil, and good afternoon, everyone. We issued a press release earlier today that included a financial update, which I will briefly summarize. Net income for the first quarter of 2018 was $30.3 million or $0.63 per basic share and $0.62 per diluted share compared with a net loss of $68.3 million or $1.63 per share, basic and diluted, for the first quarter of 2017. The shift to profitability in the first quarter of 2018 was primarily due to the sale of the MEPSEVII priority review voucher, which we sold for $130 million in January of this year. We expect to incur losses in future quarters and on an annual basis this year and over the coming years. Cash used in operations, which does not include any benefit from the PRV sale, was $89.5 million in the first quarter of 2018 compared with $61.2 million in the first quarter of 2017. This quarter, we had significant non-cash charges, including stock-based compensation expense of $18.8 million, $6.2 million in depreciation and amortization and $4.8 million in foreign currency remeasurement losses resulting from restructuring of some of our foreign subsidiaries, which is not expected to recur. We expect stock compensation expenses to continue to increase over time. We generally expect a proportion of our operating expenses that is noncash to be similar to 2017. For the first quarter of 2018, we reported $10.7 million in total revenue, which includes $1.3 million in product revenue from MEPSEVII and a modest amount of named-patient sales of UX007. Revenue also includes $9.4 million in collaboration and license revenue, primarily from our research agreement with Bayer. The revenue from Bayer is generally offset by operating expenses that we incur due to work performed on the collaboration. Our total operating expenses were $107.2 million for the…

Thanks, Shalini. I'll start with an update a MEPSEVII. The launch is progressing well, and we continue to receive new patient start forms on a regular basis, including start forms from newly diagnosed patients. We feel good about the start forms that are coming in from MEPSEVII and are happy to be able to provide an option for patients who previously had no approved therapy. Though we are not disclosing patient numbers, our progress gives us confidence in our estimate of 200 patients in the developed world. With the U.S. approval, we are now responding to requests from the inpatient programs in other countries, and we look forward to expanding our efforts into Europe with the potential marketing authorization later this year. I'll now turn to Crysvita. Our patient diagnosis activities continue in full force with 30 patient diagnostic liaisons out in the field focused on finding new patients with XLH. Their efforts continue to give us confidence in our U.S. prevalence numbers of around 3,000 children and about 9,000 adults living with XLH. Our commercial field force of 32 ultra-care liaisons has begun active promotion in the offices of physicians who are presently managing patients with XLH. Prior to approval, we began payer discussions, and we have continued these efforts post approval. Early payer reactions have been positive and have noted the responsible pricing for Crysvita. And we are cautiously optimistic about our uptake assumptions based on these conversations. As a reminder, we do believe that reimbursement will take 3 to 9 months to complete. But these efforts are underway, and we expect the penetration will ultimately reflect the true value of Crysvita. Between the incoming start forms for both children and adults, the new patients that the PDLs continue to find, positive feedback from payer discussions and the general excitement amongst the XLH community, we are off to a great start and look forward to sharing additional details on launch progress in the coming quarters.

Thank you, Jayson. As you can see, we made a tremendous amount of progress with approvals and recent launch of Crysvita and MEPSEVII. We know this is an extremely important year for the commercial execution, and we are intensely focused on these two critical launches. The transition from a development stage to a commercial stage company only happens once. We approach this transformation with deep thought and care to assure a successful transition and ultimately, assure the patients receive the therapies they need. In addition to our launch activities, we're advancing our clinical and translational programs. By the end of the year, we expect to have clinical data for our 2 lead gene therapy programs, Phase III data from our UX007 study in Glut1 deficiency syndrome movement disorders and clarity around the regulatory passage for UX007 FAOD. I look forward to updating you throughout the year on our progress. Let's move on to your questions. Operator, can you please provide the instructions for the Q&A portion of the call?

[Operator Instructions]. Our first question is from Eric Schmidt from Cowen and Company.

Maybe for Jayson on the Crysvita launch. Anything in these early days that surprised you either good or bad? And then you mentioned again that you expect reimbursement to take up to 9 or so months. What's different about the phenotype of these early adopting patients? What type of, I guess, insurance benefits do they have that allows them to get to therapy so soon in the launch?

Yes. Thank you, Eric. So the population that we've identified in our patient groups has pretty broad insurance coverage. So we're seeing a bunch of Medicaid patients, some Medicare patients and a bunch of private payers. As I said, we spent a lot of time speaking to payers in advance of the launch, and we're doing what we can to get them to have the P&T committee discussions, but of course, that just simply takes time. We are running 2 programs to provide free drug for patients once we get the start form and then to bridge the gap between the patients being approved to go on therapy by their physician and the insurance approving the reimbursement of that. So we're getting some early patients on by using those 2 free drug programs. But we're going through the normal processes of launch. That's why it just takes time for the payers to develop policies. We have seen a few policies come out in the early days, and we're so far very happy with what we've seen. We haven't seen anything that's a major diversion from label, and that's great. But it's a very, very early days in that regard.

Okay. And will the company sort of be providing metrics like number of start forms, patients -- number of patients who have reimbursement approval number on drug, those types of indicators?

I think, over the course of the next couple of quarters, Eric, we'll be able to start providing that kind of information.

Our next question is from Steven Breazzano from Evercore ISI.

On the gene therapy, GSDI alpha program, for the first dose cohort, do you expect to see efficacy in these patients? And where do you think you are on the dose curve based on some of the preclinical models?

Well, we do think we could see efficacy, although it's at the low end of the range because it's always a first-in-man experiment. But because gene therapy is a potentially one-and-done type of situation, we did want to have a low dose that had potential for efficacy. So we think that this dose at the lower end of the range from -- based on translation from the animal models, we could see an effect. And what we know from the GSDI patient data and other sources that perhaps something as low as 3% of normal might be sufficient to have an impact. So we do expect we could see an impact. However, to get to the optimal dose, our expectation is we may need to go to the second cohort or higher to get optimal impact. But it was based on what we think was the lower end of the range for effective dosing in the animal models.

Our next question is from Cory Kasimov from JPMorgan.

I guess, first one's also on the Crysvita approval and recognizing it's only been a few weeks. But I'm curious how aware physicians are at this point of the product or as your reps have kind of gotten in there and started to talk about it in the label and what's the awareness level like. And maybe what do you see is the main headwinds or tailwinds that you'll be dealing with at this preliminary stage outside of the typical payer hurdles you already discussed? And I have one follow-up.

I'll let Jayson answer that one.

Yes. So Cory, I think, in the early days, the awareness in the physicians who treat metabolic bone disease is pretty high, so in the pediatric endocrinology offices and the adult endocrinology offices who treat these kinds of diseases, and that's very encouraging for that first bunch of patients that we expect to see. And we are starting to see start forms come in, and they're picking up. We -- remembering, however, that tomorrow, we've only been approved for 3 weeks, so it is very, very early days in the launch. That being said, we've always said that the adult population offers for us an additional complexity in that many of the patients are not actually presenting in the offices of the metabolic bone disease docs. They're presenting in other specialties as a result of the complications that they're getting from their disease. And so we've got a little bit of extra work to do in terms of getting those folks referred into the offices of metabolic bone disease treaters so that they can get them on therapy. And that work has started since we got approval. So it's very early days for that, but again, we're encouraged by the desire that patients have to get treated for this disease.

Okay. That's helpful. And then the follow-up question is, and I recognize this is a little bit speculative, but as you're going through the Phase II FAOD data collection process on a patient-by-patient basis now, how is your confidence in that overall dataset and perhaps the potential to file early evolving if it is at all?

Well, we've gotten certainly feedback on a case-by-case base from each doctor. We had always a sense if there was a majority improvement. We certainly see the large majority of the patients, based on the doctor's opinion, were improved and had either improvement in clinical symptoms or improvement in symptoms and also in major clinical events and -- that they attribute to the UX007. So certainly, we're, I think, encouraged by the consistency of physician response. And certainly, not all patients respond, so -- but largely, they already did, and we feel it certainly supports our case. I don't think it changes the fact that the FDA still has to look at this and make their own determination, so we still have to put that to them. We'll provide that to them. We plan to have a discussion with them following that, and our expectation is still going to take until mid-year to get a result. But overall, Cory, we're encouraged by what we see there. It confirms what we believe the data show at the top line that UX007 is making these patients do better of reducing hospitalization and days in the hospital, and that's a real effect.

Our next question is from Tazeen Ahmad from Bank of America.

Mine is on Glut1. So you've got data from the Phase III movement disorder study coming up in the second half of '18. What would you consider to be a successful trial result there?

Well, remember in neurology, there's always a lot of heterogeneity in patients, so let's look back to what we found in the Phase II study. We had -- in a group of 6 patients that were carefully managed by 1 physician, they had a 90% reduction in movement disorder event, which is a very profound effect in neurology. We wouldn't necessarily expect it to be something like that because now in the heterogeneous randomized controlled study, it's not quite the same setting. So what we'd like to see thereby would be we would expect some fraction of the patients would respond. We haven't set a threshold. We certainly would want to see statistically significant result, and I would expect even a 30% to 50% overall reduction might still be meaningful, particularly since I would say it would be a mix of patients who have a very nice result and some patients who have less result. So I'd expect to see some heterogeneity in response, all right? So if you look at -- I would look at this just like you would look at a seizure study. And a lot of seizure studies show a 30% to 50% reduction in overall seizures, but some patients have a really good effect. Some patients have no effect, and that's kind of the way it would look. So I wouldn't get too hung up on the mean effect. I'd look and see how many patients have really good clinically meaningful effect, would might be another way of looking at the data in terms of its value for patients with Glut1 deficiency syndrome.

Okay. And then a follow-up to that, are you expecting that your drug would be applicable across the entire Glut1 population or for a subset?

Well, we're looking at patients who have the movement disorder, so I think from -- by inclusion criteria, we're talking and speaking to that particular phenomenon. We're not -- we're looking at other things in there, but we're focused on that. That's where the effect seems to be most distinguished. The truth is that the majority of Glut1 patients that are above the school age have movement disorders of some degree. So it's a pretty substantial fraction of the population we would see would have movement disorder problems, and we were able to enroll the study well looking for those patients. But it doesn't include those patients who have seizures only, and that would require I think -- look at seizures more specifically, which we earlier did not show good effect at this point. But at this point, we think it would be indicated for the patients with movement disorders.

Our next question is from Maury Raycroft from Jefferies.

First question is on 401. So we noticed that it was posted to -- the clinical trial was posted to ClinicalTrials.gov. And it shows time to first hypoglycemic event is one of the secondary outcome measures. Just wondering if this would be the eventual endpoint you would use for FDA approval.

We are considering that as the key endpoint because it has extreme clinical relevance to this population. One of the great dangers of this disease is that they go severe hypoglycemic during the middle of the night and die, and there are patients even who are well controlled that can have that happen. And so we think being able to tolerate going through the night without having a drop is a clinically meaningful benefit. And I think because FDA understands glucose and hypoglycemia, I don't think that's going to be a hard one to convince them of the relevance of it to its importance. So that is probably what might be the primary. We will look at all the things that we look at. We would, of course, look to see about some of the other secondary metabolic effects, the storage in the liver and other signs because liver injury is another part of the disease long term and what -- how liver injury causes complications, so -- but hypoglycemia would be the thing that potentially kills patients, and therefore, that's why it's probably the biggest driver of efficacy.

And then for the ASGCT update next week for 301, any comments on what we should expect for that update would be great.

I think the data presented are going to be about the first cohort. There shouldn't be anything -- second cohort, which is still in progress.

Got it. So just additional follow-up then probably.

Yes.

Our next question is from Adam Walsh from Stifel.

First one on Crysvita, I noticed in your comments that you talked about patient start requests from both adults and pediatrics, and there had been some talk kind of going into the approval and post launch that maybe the adult population would be harder, more difficult to identify and convert onto drug. Are you surprised that, this early in the launch, you're getting adult patient start requests? And then, if possible, could you give us the breakdown in terms of adults and peds for the start requests?

Yes. So I don't think we're surprised that we're seeing adult requests come in this early. We always said that there are a population of adults who are managed by the endocrinologists for the metabolic bone centers and who have a diagnosis of XLH, but that's a proportion of the overall adult population with the bulk of the adult population perhaps not having a definitive diagnosis of XLH and rather being treated for some of the complications. So there is this group of patients who have the diagnosis and have been sitting, waiting for therapy. So in the early days, we would expect to see roughly even numbers or equal numbers of adults and pediatric patients come in, but that gets a bit more challenging as we get into the broader adult population and as we start trying to get those referral patterns to happen. And that will take us a little bit longer to do.

Understood. And then just one follow-up. In the UX007 Phase III movement disorder study, I -- if I recall correctly and this was a while ago, Emil, there were some minor differences in the endpoint that was looked at in the proof-of-concept study and then the endpoint in the Phase III. Can you just refresh our memory on what the differences, if any, are between the 2 studies?

In the proof-of-concept study, we include all movement disorders, and it was a broader range of things. In our discussions with the FDA, they offered a request that we look more at disabling is one option in the program, disabling type. That is ones that interfere with the activities of daily living or whatever you were doing as opposed to any of the symptoms. We restricted that analysis, but when we looked at the proof-of-concept data for the restricted set of disabling movement disorders, the reduction was the same as the overall. So we don't think -- it just changed the number of events we're looking at, but it doesn't change the degree of effect that we were observing. So we're comfortable with the more limited set of disabling movement disorder not -- rather than all possible movement disorder.

Our next question is from Laura Chico from Raymond James.

One question on the Glut1 DS program. There's actually an interesting publication that just came out describing an AAV-based gene therapy product. It's preclinical data in a 1 DS knock-out mouse. But just curious, as you're looking at the opportunity, I'm just wondering if you think there is a role for gene therapy in this disorder and whether that might represent a more feasible approach.

Well, we've certainly -- now that we have a gene therapy technology, we certainly looked at it. I think that the disease in Glut1 is complicated because there's a transporter, and it's in the blood-brain barrier but also other places wherever the Glut1 transporter is. Keep in mind, in Glut1 deficiency syndrome, it's a defect that's autosomal dominant. That is 50% abnormal, gives you classic Glut1 deficiency syndrome, severe classic type, whereas 65% of normal, say, gives you a less severe non-classical type. So the problem with that means that in order to have a really big impact on Glut1, you need to get like close to 100% on the blood-brain barrier endothelium. You can't just have a few percent of the cells in the blood-brain barrier expressing. You need to have a high fraction doing a high amount. So there is a significant clinical challenge. Plus, it's not just blood-brain barrier. It's other places in the body where the uptake has to occur. I feel like this would be a complicating way to go. Of course, there's a lot of people trying to apply gene therapy to many things. At this point, we think that a small molecule substrate that can diffuse right across the membranes of the blood-brain barrier and across neuronal membranes will be a more expedient and more comprehensive way to distribute energy to the brain, which we think would be potentially superior and easier to apply.

And just one follow-up. On DTX201, I'm sorry if I missed this. Just curious, do you have an update on kind of the milestones for the program? Are you still planning on progressing towards an IND in the second half of '18?

Yes. In DTX201, we are progressing well, and in general, what we'll do with the new IND is we'll announce them when they've been cleared to progress, not when they get filed just as we did with GSDI. But we're on track to get that in play as we described.

Our next question is from Chris Raymond with Piper Jaffray.

I just want to understand the dosing dynamics with Crysvita. With the drug coming in, in 10-milligram increments, obviously, this is a weight-based dose. And I know, Shalini, you guys have guided to a net price for peds and for adults. But with the label sort of calling for treatment to -- in dosing to round to the nearest 10 milligram, as I do the math, there's a big step-up in cost as we go from a 30-kilogram patient, for example, in peds to a 32.5-kilogram patient. It works out, I guess, that there's a bit of a step function, I guess. And I know you guys have said weight is between 30 and 35, but there's a decent amount of variability there. So I guess, I'm just kind of curious, as you have guided to that $160,000 for peds, as we look at the weight, you could see a step function up by about $80,000. So I guess, maybe can you talk to how much of that is just variability? Or is there a built-in sizable discount like 30% plus that we should be thinking about here?

Sure, Chris. I think in the data we put out, we were trying to give you like an average piece. We took kind of an average in the range of the trial and that 30 and -- between 30 and 35, there is a step, and so we basically averaged what the 30 would be and the 35 would be to give you that $160,000. It's kind of in the middle of those two doses. Yes, there is step function of 10 milligrams per step and so therefore, the price or cost is going up in those steps. But I would also point out that the doctors are going to be looking at the phosphate in their patients to get to the low normal range. And so the dosing may be starting at a certain place, and they'll look and see how their kid's doing. If they're low, they may step up a little bit in their dose. The goal of giving you the $160,000, that was to give you kind of a rough average, including with that number you have for the gross-to-net calculation as well the compliance adjustment to give you kind of a more practical real-world estimate of what an average kid would be. But on an individual basis, there are a lot of questions about what your result is and whether you need to go up or not, and each step does increase the cost.

And maybe just to follow up, so I guess, then the logical sort of next conclusion is if, especially in peds, if these kids get -- are getting heavier as the time goes on, years go by, et cetera, you should see a corresponding step-up in revenue per patient.

That would be our expectation that they would be increasing the dose as they grow.

Our next question is from Dae Gon Ha from Leerink Partners.

So one quick one and then a follow-up for Jayson or Emil. First is with regards to Crysvita launch, can you talk to -- about the extent of different physicians and their specialties administering this drug, so specifically, metabolic bone specialists versus geneticists, endocrinologists and nephrologists? And then a follow-up is, with regards to some of the rare disease medicines and the access to Brazilian market, how are you guys approaching this as you contemplate and begin your process of launching Crysvita there. Can you talk to us about the -- your IP estate there?

I'll let Jayson take that.

So first of all, with regards to Crysvita and prescribing, our anticipation is that the vast majority of patients will get treated in an endocrinology office. So in the pediatric population, that would be the pediatric endocrinologists, and in the adult population, that would be the adult endocrinologists with a specific interest in metabolic bone disease. We do expect to see some nephrologists and maybe some geneticists treating these patients on their own. They do have a history of treating metabolic bone disease. But for the vast majority of other patients who are being managed for various complications of the disease, we expect to see the docs presently managing them for the complications refer them into the endocrine centers to get them treated for the underlying disease. That's our expectation. That's kind of what we're seeing in terms of where the patients are being managed at the moment. And as we identify patients in other clinical specialties, it's relatively clear that those specialties do not want to or do not feel comfortable treating them for metabolic bone disease. So that's the plan that we have, and that's how we expect it to unfold. In terms of your specific question around Brazil, now that we have approval in the U.S., we are able to start responding to named-patient requests for therapies in Brazil. We have actually filed MEPSEVII in Brazil. So we anticipate that we will be filing in these countries, and we will be going for approval. And then the other question around IP estate, obviously, IP is very complex in Latin America, and there'll be no more to say about that.

Our next question is from Arlinda Lee from Canaccord Genuity.

I had one on the 301 program. I guess, since you guys had already some efficacy in the initial dataset, I'm wondering what are you looking for in the second cohort to decide whether or not that should be expanded or go forward to the next dose level. And then maybe a follow-up one on -- for Shalini after that.

Okay. For DTX301, the second cohort, we'd be looking to see how many of the patients get to reconsider a normal ureagenesis or better level. We got to around 130% of 1 of the 3 patients with Cohort 1. We like to see that in all the patients, but that's kind of what we're looking for. And whether we -- what it takes to go to the next cohort may depend a little on the degree of effect, consistency effect across those patients as well as safety. But we're looking forward to a dose level that we think would reliably give normal ureagenesis in patients. With a sample of only 3, that obviously makes it kind of limited, but that's what we're looking for, for 301. And the second question?

On the R&D cost, for Shalini, is this maybe a good base to go forward? Or do you think this might fluctuate a little bit more?

Thanks, Arlinda. so what we've said for this year is that we expect R&D as a proportion of operating expenses to be similar to 2017, and in 2017, it was approximately 70% of operating expenses. So hopefully, that gives you a sense. We're not doing any kind of guidance from quarter-to-quarter, but at least for the full year, you can use that guidance.

Our next question is from Liisa Bayko from JMP Securities.

Just with respect to MPS 7, as you're kind of learning about this market and going out there with your drug, can you maybe speak to what you're learning about patient numbers and sort of what you're thinking of now in terms of how many patients are out there? And then any comments on sort of gross to net, too, would be helpful.

Well, I think the assumption always for MPS 7 that there are more patients than we had originally identified, but there -- we expect that number would be a couple hundred if you found everyone. And the reason is that patients diagnosed get lost or fall off, and then there are some patients who might not be properly diagnosed. We have seen a few new patients get diagnosed during this interval, some of which were older patients. We do think it's possible that we'll start seeing more patients getting diagnosed now that the drug's available, and that's the pattern I've seen in all MPS launches over time. People start paying attention. Labs start running it. It starts to become now important to pick it up as opposed to before it would be just an oddity or curiosity. So we're seeing more of that now. I think the two areas where growth could happen are in the hydropic young babies' situation, where they might not be always diagnosed and then also in the intermediate milder patients that have significant bone disease that might be in their teens or 20s where sometimes the diagnosis wasn't made currently. So we're starting to see some new patients diagnosed in that area, and we expect that to continue. It doesn't change our overall prediction around couple hundred. Right now we don't have any evidence to say that it's different. We still think that's probably a fair number.

And how many do you approximate today? Like, how many are identified today, I guess, as a start?

Yes. We haven't put it out because it's constantly changing, and we also haven't -- because you can't really carry identified information, we always can't deduplicate. So that's why it's kind of a number we have, and we don't have 200 found. I will say that. That has never been the case. That's our projection as if we were to find everyone in the areas where we're commercializing.

Our next question is from Gena Wang from Barclays.

Just one for movement disorder. Wondering if you can remind us the assumption on frequency of disabling movement events for placebo and the drug arm? And also for approval, in addition to hitting primary endpoint, any key secondary endpoints you have to show positive trend?

Okay. So we haven't really put out specifics around the movement disorder. What I can say to you is that in the pilot study, there were around 30 events during the 2-month period. We're doing a 6-week period. The requirement though is only having a few. It's more like 4 to get into the study. We are seeing a lot of patients with many more than that though in that range, so I think the pilot study around 30 is kind of a reasonable kind of estimate for what you might see. And majority of those were disabling really. Majority of the events are disabling. So hopefully, that's helpful to you.

Yes. And also for approval, in addition to the primary endpoints, any specific secondary endpoints you have to show positive trend?

We don't have any specific requirements in the secondary endpoints, but what I would say to you is regardless of what you may discuss with any regulatory authority, end of the day, they're going to look at the actual data you have on the primary. It's kind of a core thing. And there -- it is picked as a primary because we think it's the main clinical meaningful one. We don't have a requirement at this point for any particular secondary endpoint. The more secondary endpoint data you get, though, of course, the more supported the primary is, but -- so at this point, I -- there's no specific requirement, but we need to see what the data look like to determine how compelling it is for a single pivotal study and to be able to file.

Our next question is from Vincent Chen from Bernstein.

On the patients treated with the DTX301 gene therapy, what's your sense for why patient 1 showed a robust dose response -- well, rather robust response overall, while patients two and three did not. And is the late rise to over 130% seen in patient 1 in week 24 likely to be a real and sustained effect? What makes you think so? What makes you not? And if so, what might actually explain the relative late rise?

Well, the reason for the one we've talked about before but we have not proven the reason for, the one distinctive thing is that patient 1 received the largest total dose of AAV vector because the patient was 100 kilos. So he was almost twice the size of patient 3. So there is -- that is one factor. We haven't seen any other immunological or other criteria to explain why that is. And I would also point out that for a lot of gene therapy programs, there is patient-to-patient variability, so I don't think we're unusual in that regard. With regard to the response, it's pretty clear that patient response was very strong initially. They got put on steroids because of their -- their liver enzymes rose just above the threshold and that when the steroids were on, there was some -- there looked like some depression of the level. And then when steroids came off, it seemed to bounce back. So our impression right now is that the use of steroids may have suppressed the liver's expression or activity of the genomes that were present and that the removal of steroids released it back -- to bring it back to where it was because it was at 120% or something in that range. In the 6- or 12-week range, it hit 120% and then came down to the 80s and then went shooting back up. So we think that what you saw at 24 weeks was similar to what was actually early on before steroids were applied. So it could be a -- well, let's say a transitional event during steroid treatment that's causing that. But it's one experience of one right now so we can't be sure. But we do think that the effect you're seeing is -- now is maintained and seems substantial, and the patient has been doing well.

Our next question is from Yigal Nochomovitz from Citi.

This is Yang on for Yigal. So just quick one. Your press release has $1.3 million in product revenue from MEPSEVII and UX007. Can you say how much is from MEPSEVII and how much from UX007? And then where are you guiding revenue for 007?

Go ahead, Shani.

Thank you for the question. So the vast majority of that revenue is attributable to MEPSEVII, and a very small amount is attributable to named-patient sales for UX007. And I can tell you the UX007 revenue is ex-U.S. named-patient sales. Hopefully, that's helpful.

Our next question is from Matthew Harrison from for Morgan Stanley.

This is Ishmael on for Matthew. So one on Crysvita and the other on 401. Can you remind us on the opportunity in TIO with -- especially within the context of access to market and if really good data could drive off-label usage on 401. Could you talk more about how the preclinical data supports the opportunity and also some additional details on the likelihood the dosage you are evaluating are within the therapeutic range?

Okay. So for TIO, what we did say is we have some TIO data ready, which we've put out, and we're having additional data through 48 weeks, which is data that's similar to what we saw with bone quality looking at osteomalacia and bone biopsy improvements as well as phosphate and symptoms. The TIO is certainly not in the label, and certainly, we're not promoting for TIO at this point. The TIO opportunity is a relatively smaller opportunity. It's certainly, at most, about a couple thousand TIO patients, but maybe only half have non-resectable tumors. So it's in the range of 1,000 patient, so it's a relatively small fraction of the total 12,000 we're looking for XLH. Our expectation is to collect those additional data on the patients we have studied and to talk with the agency about what it would take to file a supplemental BLA for the product. With regard to 401, the preclinical data were in both mice -- a mouse model and dog model. Mouse model probably provides more of the detail on the exact dose of the vector, but in both models, a relatively small amount of expression was able to achieve a significant improvement and particularly in the dog in terms of glycogen storage in euglycemia. We have -- some of the data's already been presented, but rather than going through the deep titration of that, we think that within the 2e12 range that we're starting at to the 10, 13 range, which we expect to do in the study, that encompasses the middle of the mouse range to the upper end of the range. So we think, given where we are in the mouse and translating that to human, which is really increasing the dose compared to mouse, we're right in the middle of the range what we should see efficacy. That said, I think 2e12, the first dose is at the lower end, and we would expect it could need to go higher based on what's been seen elsewhere. So hopefully, that's helpful.

At this time, I'm showing no further questions. I would like to turn the call back over to Danielle Keatley for closing remarks.

Thank you, and this concludes our call today. A replay will be available shortly. If you have additional questions, please contact us by phone or at ir@ultragenyx.com. Thank you for joining us.

Ladies and gentlemen, thank you for your participation in today's conference. This concludes the program. You may now disconnect.