BioMarin Pharmaceutical Inc (BMRN) 2018 Q2 法說會逐字稿

Welcome to the BioMarin Second Quarter 2018 Financial Results Conference Call. Hosting the conference call today from BioMarin is Traci McCarty, Vice President of Investor Relations. Please go ahead, Traci.

Thank you, operator. Thank you, everyone, for joining us today. With me from BioMarin's management team are J.J. Bienaimé, Chairman and Chief Executive Officer; Hank Fuchs, President, Global Research and Development; Dan Spiegelman, EVP and Chief Financial Officer; Jeffrey Ajer, EVP, Chief Commercial Officer; and Robert Baffi, EVP of Technical Operations.

To remind you, this nonconfidential presentation contains forward-looking statements about the business prospects of BioMarin, including expectations regarding BioMarin's financial performance, commercial products and potential future products in different areas of therapeutic research and development.

The results may differ materially depending on the progress of BioMarin's product programs, actions of regulatory authority, availability of capital, future actions in the pharmaceutical market and developments by competitors. And those factors are detailed in BioMarin's filings with the Securities and Exchange Commission, such as 10-Q, 10-K and 8-K reports.

Now I'd like to turn the call over to BioMarin's Chairman and Chief Executive Officer, J.J. Bienaimé.

Thank you, Traci. Good afternoon, and thank you for joining us on today's call. So we delivered another very solid quarter and are on track to build our successes in the first half of the year.

Our accomplishments in May included the approval of PALYNZIQ, a very productive FDA advisory committee meeting of achondroplasia drug development guidelines and our 2-year update on expansion of the Valrox program. These accomplishments set us up for significant financial and regulatory achievements over the next 6 quarters.

With PALYNZIQ now approved in the U.S. and on track for an EU approval in the second half of 2019, we believe we can generate approximately $2 billion in revenues in 2020 with our currently commercialized products alone.

In addition to regulatory and clinical progress, we delivered strong revenues of more than $372 million in the second quarter, a 17% increase over Q2 of last year. We continue to see strong patient demand for our legacy products and are very pleased with the pace of the commercial launch of PALYNZIQ and Jeff is going to give you some details on that.

While we expect to incur a GAAP net loss for 2018, we remained on track to deliver non-GAAP profitability in excess of $100 million for the full year 2018 and total revenues of roughly $1.5 billion.

Having achieved key milestones across the late stage pipeline while delivering strong financial results, we are pleased with our accomplishments in the first half of the year.

And now I'll turn the call over to Jeff who will review the commercial business in more detail.

Thank you, J.J. I'm extremely pleased with the commercial team's execution in the second quarter of 2018, generating a 17% increase in total BioMarin revenues compared to the same period last year. Before reviewing specific product revenues, I would like to provide some commentary on the PALYNZIQ launch in the U.S. following approval in May.

As J.J. said, the early launch dynamics reflect the high level of interest from physicians, clinics and patients. All clinical trial patients who are being treated with up to 40-milligrams dose have been referred to RareConnections, which accounts for 156 subjects from the Phase III study. More than 50 naïve-to-treatment patients have also been referred since we received approval. So our launch expectations are in line with how the PALYNZIQ rollout is performing. These patients are smoothly working through the brand's enrollment process, and we are very pleased with that progress so far.

Keep in mind that the transition of clinical trial patients to commercial pace therapy is subject to an end-of-study visit for each patient, the timing of which is variable and subject to scheduling with trial sites. We anticipate the majority of patients to have completed their end-of-study visits by the end of the third quarter, and the vast majority of clinical trial patients to be converted to commercial therapy by year-end.

Another important measure of the initial launch is the pace of REMS certification by prescribers. Already, 112 physicians from 55 unique clinics are REMS certified and ready to prescribe, which includes a number of physicians at naïve-to-treatment clinics.

As we guided during the approval conference call, no revenue was generated in the second quarter as product supply only became available in the first week of July. We look forward to providing a revenue update on our third quarter financial results call as well as sharing 2 key launch metrics. Recall, they include total number of patients on commercial therapy, both those in the induction and titration phase, and those on maintenance dose; as well as the number of active treating sites defined as having one or more subjects on therapy with PALYNZIQ.

I would now like to turn to second quarter results for our other approved products, all of which contributed to strong revenues in the quarter. We saw growth across the commercial portfolio and want to share a couple of key trends behind the numbers. Despite unpredictable ordering patterns from Latin America, our MPS products, Naglazyme and Vimizim, benefited from new orders from that region quarter-to-quarter. The EMEA was also a large contributor to Naglazyme and Vimizim results due to favorable order timing in certain markets.

As a result, Vimizim revenue was $128 million, an increase of 24% compared to the second quarter last year and Naglazyme delivered $91 million in revenue for the second quarter, a 6% increase compared to the second quarter of last year. For the remainder of 2018, we remain cautious on Brazil quarter-to-quarter due to anticipated uneven buying patterns but reaffirm previously provided annual revenue guidance for both MPS brands.

Moving now to Brineura for the treatment of CLN2, which was approved in both the U.S. and the EU in 2017. We are pleased that our efforts to identify and treat children early in the course of their disease are progressing well. We are treating dozens of children globally across 12 markets with many new patients in the quarter from the EUMEA region. We are making progress working through pricing reimbursement processes in major EU markets and also gaining name patient approvals in smaller markets.

Even with the unique challenges presented by diagnosing and coordinating treatment, we are seeing a steady increase in patients on therapy. The second quarter contribution from Brineura was $11 million, and we are optimistic that it will become an increasing contributor to our total revenues as an important part of our diversified base business.

Finally, on to Kuvan. Solid revenue of $109 million in the quarter resulted in 7% growth year-over-year. Kuvan results were driven by strong growth in North America and continuing development of ex U.S. markets. Importantly, in North America, we saw a 10% growth in commercial patients year-over-year.

So in closing, the company generated strong revenues in the second quarter of 2018, and we look forward to updating you on specific U.S. commercial launch metrics with PALYNZIQ starting next quarter.

Now I'd like to turn the call over to Dan to provide more details on the financial results in the second quarter. Dan?

Thank you, Jeff. Please refer to today's press release summarizing our financial results for full details on the second quarter.

Starting with top line results: total BioMarin revenues for the first half of 2018 were a record $746 million, an increase of 20% year-over-year, driven by strong contributions from all of our marketed products. As J.J. mentioned, with our current commercial portfolio alone, we believe we can grow revenues 15% a year and generate approximately $2 billion in revenues in 2020 and then accelerate revenue growth after that based on Valrox and/or Vosoritide.

In addition to a strong and growing base business, we also have a diversified and therefore relatively secure product mix. To start, no single product is responsible for more than roughly 1/3 of our net product revenues. Moreover, there is a very diverse regional mix of revenue that reduces the reliance on any one territory.

For example, 4 years ago, the U.S. accounted for 43% of the net product revenues of the products we sell, and this year is expected to drop to 37%. In addition, Brazil accounted for approximately 14% of those product revenues 4 years ago, and this year is expected to contribute 5%. In the future, we expect the regional revenue mix to continue to broaden as we add new markets and new -- and more products, essentially building more of a hedge and reducing risk in times of uncertainty or volatility in any one country.

Turning now to particular items of note in the quarter, in June 2018, we received $20 million in milestone payments from Pfizer, Inc. triggered by the FDA's and the European Medicines Agency acceptance of Pfizer's submission of an NDA and a Marketing Authorization Application for talazoparib. The payment is being accounted for as a gain on sale of an intangible asset under operating expenses. Of note, if talazoparib is approved, we would be entitled to up to an additional $40 million in regulatory milestones, $100 million in sales milestones as well as a mid-single-digit royalty on sales.

Moving to operating expenses, R&D expenses increased to $176 million in the second quarter of 2018, compared to $143 million in the second quarter of 2017. This year-over-year increase was primarily due to increased spending on the Phase III programs with Valrox and Vosoritide as well as spending on the MPS IIIB program.

For the second half of the year, we expect total R&D expenses to be similar to first half levels while accelerating enrollment of additional payment -- patients in the global Phase III (inaudible) study. Last quarter, we announced an increase in planned enrollment in this study to 130 patients from the originally planned 40 patients in order to potentially demonstrate superiority on the gold standard annualized bleeding rate endpoint versus current standard of care. This expanded program has resulted in an increase in our full year R&D guidance to between $680 million and $710 million. So importantly, we expect to absorb this increased R&D spending while leaving all other guidance, including GAAP and non-GAAP bottom line results, unchanged.

Turning to bottom line results, GAAP net loss in the quarter was $17 million as compared to a GAAP net loss of $37 million in the second quarter of 2017. The decrease in GAAP net loss year-over-year was due to increased gross profit primarily driven by increased Vimizim and Brineura net product revenues as well as the milestone payment received from Pfizer for talazoparib. These increases were partially offset by higher R&D expenses.

Our non-GAAP income in the second quarter of 2018 was $20 million compared to non-GAAP income of $27 million in the second quarter of 2017.

As noted, we remain on track for full year non-GAAP income of between $100 million and $140 million.

In terms of cash, cash equivalents and investments, as of June 30, 2018, we had $1.6 billion as compared to $1.8 billion at year-end 2017. Looking forward, and as a reminder, we have approximately $375 million of senior subordinated convertible notes, which will mature in October 2018. As previously promised, we have elected to net settle any conversion of these notes. What this means is that if BioMarin's VWAP, or volume weighted average price, based on a 25-day observation period, exceeds the conversion price of $94.15, noteholders who have surrendered their notes for conversion will receive cash for the principal amount of their notes and also receive shares of BioMarin common stock representing the excess conversion value.

To offset a portion of those dilutions from the excess conversion value, BioMarin previously entered into cash call arrangements with counterparty banks at the time the notes were issued in 2013. Under these cash call agreements at maturity, BioMarin will receive shares of BioMarin stock equal to 50% of the excess conversion shares that we would be issuing.

By utilizing net share settlement and our cash call arrangement, the net dilution from settlement of the bonds at maturity will be very small. For example, if the observation period VWAP is equal to $100, the total net shares issued will be approximately 117,000 shares. At a $105 it will be approximately 206,000 shares.

For more precise information about the convertible note settlements, please refer to the indenture and first supplemental indenture dated October 15, 2013.

In closing, BioMarin delivered another solid quarter and excellent first half 2018 with strong Vimizim, Naglazyme and Kuvan sales, controlled operating expenses and our continued progress towards increasing non-GAAP profitability this year.

Now I would like to turn the call over to Hank. Hank?

Thanks, Jeff. Since J.J. has already walked through our key regulatory highlights in the quarter, I'd like to spend a moment on a more recent but equally important development positively impacting our Valrox program.

On July 11, the Food and Drug Administration and the Centers for Biological Evaluation and Research issued draft guidance for gene therapy product development. We were very pleased with the scope of the specific guidelines as they pertained to our global Valrox program for treatment of severe hemophilia A. For the industry, the guidelines illuminate the agency's focus on expediting the development and approval of innovative gene therapy treatments. For BioMarin specifically, we are pleased that the draft guidelines align well with the design and development path for our ongoing Phase III program. As a reminder, GENEr8-1, using the 6e13 dose, should be fully enrolled by the end of the first quarter of '19, and GENEr8-2, with the 4e13 dose, is expected to be fully enrolled 1 or 2 quarters after that.

Back to the draft guidelines, just to name a few of the most germane considerations that are already part of our development program. First, observing patients for 6 months through a lead-in period in a study to collect data for the annualized bleeding rates. Annualized bleeding rates based on retrospective data collection for medical records may be subject to recall bias and missing information and importantly, this is part of our GENEr-1 and GENEr8-2 study designs.

Second, sponsors should evaluate discrepancies between 1-stage clotting assays and chromogenic substrate assays early in the course of clinical development prior to considering whether to pursue accelerated approval using factor activity levels as a surrogate endpoint.

BioMarin's view on this point is that we have made tremendous progress understanding and delineating the basis of these findings, and we're confident that we can provide the Food and Drug Administration the information they'll need to evaluate Valrox's benefit and risk at the time of filing. Of course, specific details pertaining to our investigations remain proprietary and confidential. We've reported this discrepancy result between 1-stage in chromogenics at previous medical conferences and have published and would remind you that our previous reports indicate that the 1-stage result was approximately 60% higher than the chromogenic result. And as a further reminder, mean 1-stage activity at the end of year 1, when our pivotal study will be reading out, was approximately 100% in the 1-stage assay and 60% in the chromogenic assay, still well within the range of normal.

Further, other groups have reported such assay discrepancies before. For example, in the publication by George, et al in The New England Journal of Medicine in December of last year, these authors note that the 1-stage assay of Factor IX activity is roughly 60% higher than the chromogenic assay results. Traci can provide a copy of this reference if you'd like to dig this up for yourself. But importantly, the consideration is that there may be a generalized phenomena in transgene expression in humans with hemophilia of different types.

Also, we would note that recombinant products have been observed to have discrepant assay results and so these findings of discrepancies between 1-stage and chromogenic are by no means unprecedented, and we are well on our way to understanding these in a way that the FDA can take advantage of.

A third point pertains to efficacy endpoints for traditional approvals. Annualized bleeding rate is considered to be the primary endpoint to demonstrate clinical benefit and this is our primary -- this is our clinical endpoint in our program. And, in regard to accelerated approval, the agency comments that that factor activity may be considered as a surrogate endpoint for primary efficacy assessment under accelerated approval pathways. And as we've talked about before, factor VIII activity levels are one of our endpoints and while we are aware of the opportunities for early regulatory action this could present, and we have plans about this approval pathway, we won't provide specific details in regard to our accelerated approval endpoint strategy for competitive reasons.

However, to support this endpoint the Food and Drug Administration goes on to say, "We recommend that you resolve discrepancies in factor activity results for various assay methods, targeting factor activity as a surrogate endpoint," and as I said, we are working on this and the FDA goes on to then say, "Determine a factor activity level within the range of factor activity of normal populations." And clearly, we have demonstrated that we achieve a high level of factor activity representing a very high bar.

Taken together, BioMarin believes that the draft guidelines for the development of gene therapy products are very supportive of our Valrox study designs and approval pathways. The recommendation for factor activities normal range under an accelerated approval raises the bar for results outside that range and gives us an important competitive advantage. The draft FDA guidelines, along with our commercial scale manufacturing capabilities gives us greater confidence in our prospects with Valrox. We are more encouraged than ever that we have chosen the right development path, as well as the decision to develop both the 6e13 and the 4e13 dose, and we look forward to completing enrollment in GENEr8-1 in Q1 '19 followed by GENEr8-2 a quarter or two after that.

In addition to aligning well with U.S. health authorities, we're very encouraged by our ongoing conversations in development planning with the EU regulatory parties.

Turning briefly to another of our development programs, we're pleased to introduce the International Nonproprietary Name, or INN name, for BMN 250 for the treatment of MPS type IIIB. And going forward, we will refer to that product as (inaudible) alfa when we update you on the program. On that note, we look forward to providing you updates from the ongoing study that this year's SSIEM meeting in early September, so stay tuned.

Finally, I want to remind you to mark your calendars for our R&D Day on November 7. We look forward to sharing an in-depth look at research capabilities as well as how we're thinking about the future of BioMarin's development pipeline and beyond. We hope you'll attend. The IR team will be setting up save-the-date notices in the near future. Thank you very much for your support and I will now open it up to questions.

(Operator Instructions) Your first question comes from the line of Cory Kasimov from JPMorgan.

A couple of them, probably for Jeff, on PALYNZIQ launch. I guess, first of all, I'm wondering how we should think about modeling free drug or discounting for the product over the first few quarters here. And then the second question I have is, looking at this anecdotally for the roughly 50 naive-to-treatment patients that have been referred to your RareConnections for PALYNZIQ, do you have a sense yet whether these patients are primarily switching from Kuvan? Or if it's those patients that are in the clinic but not on drug at this point?

Cory, thanks, a couple of great questions. This is Jeff. I will try to cover both of them. Starting with the question about the free drug modeling. I do want to reinforce that, unlike Kuvan, we don't have a starter program, and we don't intend to have a starter program for PALYNZIQ. So when we have patients starting with their induction dose followed by titrating up to maintenance dosing -- and a reminder, we'll be reporting on those patient numbers for 6 quarters starting at the end of the third quarter -- so when you see those patients, those patients will all be commercial reimbursed therapy. The only place that free drug is coming in to play in a material way is for the clinical trial transition patients. So I already introduced the concept that we have all of those patients referred into RareConnections, our hub program. In parallel, those patients are managing through their end-of-clinical-study visits at their clinical trial sites. Once they get through that gate, they're available for commercially labeled drug. Through multiple experiences with transitioning patients from clinical trials on to commercial drug, we've learned that it's best to do two things in parallel: first is to work on getting patients transitioned on the commercial label drug, and if that's free drug, that's fine; in parallel, we work then on getting reimbursement approvals so that we can switch on the reimbursement. So free drug will really only play a meaningful role for PALYNZIQ in the U.S. in the first several months of launch. And again, when we're reporting those patient numbers a quarter from now, those will all be reimbursed patients. And then, relative to your question about the 50-plus naïve patients, I can't resist saying how pleased I am with that rapid uptake of naïve patient enrollments possibly exceeding our expectations at this stage. There is a mix of patient types in there, they're mainly naïve to therapy. There are a handful of Kuvan switch patients inside of there.

Your next question comes from the line from Salveen Richter from Goldman Sachs.

With regard to payer coverage decisions on PALYNZIQ, how is step edits with Kuvan being incorporated? And then just secondly on the heme A program, could you just give us an update on how enrollment is trending there?

Salveen, this is Jeff. I'll cover the first question and Hank will jump in on the second one. So we're starting to see payer coverage policies put in place and those are a reflection of our kind of national accounts effort getting in front of the major payers with PALYNZIQ presentations. And so far, these payer coverage policies look consistent with our label -- a reminder that our label indication is for patients greater than 600 micromolar fee levels on existing management. Existing management, of course, would include both diet therapy and Kuvan. That's wholly different than introducing a step edit or requiring a Kuvan failure specifically. So far, we haven't run into step therapy with Kuvan explicitly in the payer coverage policies, or perhaps more importantly, as we're going through the individual patient authorization with payers that hasn't cropped up yet. Wouldn't be surprised if it does in a minority of cases going forward, but so far, no sign of that yet. And Hank, on the second question?

Yes. On enrollment. As you know, Salveen, we don't give specific enrollment updates during the course of the clinical trial. But suffice to say, we're pleased by what we're seeing so far. There's an enormous amount of enthusiasm, as you might imagine, for a treatment that could really so profoundly change the lives of patients, and we're busy bottling that enthusiasm and getting patients enrolled in our trials.

Your next question comes from the line of Joseph Schwartz from Leerink Partners.

I was wondering, just given a current event in this space how you view the risk of an anomalous DNA fragment entering your production batches based on your processes and procedures for Valrox.

So Robert, you have plans answering the questions?

Yes. We followed that recent event very closely. And what I can tell you is that our starting materials are all GMP materials. They have been characterized, they're released by sophisticated analysis and specifications, and we have a very good understanding of the quality of that -- those starting materials and raw materials. So I -- we are vigilant to make sure that as we produce batches of our starting material that they meet those levels specifically, and we have appropriate methodology in place, but I would also mention is that our production system, while we have not talked specifically about it, makes very conservative use of those starting materials. Sort of a working bank, master bank-type mode. And so we are not routinely going from one batch to batch with our starting material for each production. So we have experience with the materials generated thus far. They've been used to generate the Phase I, II and III materials that we've been testing all along. We feel very confident in the overall quality of that starting material.

Okay, great. And then I also wanted to ask about your circular DNA hypothesis and this is probably for Hank. I know you haven't provided patient level data, but have you seen stabilization in the patients who've been treated for a longer period of time consistent with this hypothesis? And can you expand on your WFH commentary that the circularization leads to more stable expression? Is this something that you're able to observe or measure in any way in human hepatocytes? Or is it just theoretical?

Joe, so I can't provide any updates beyond what we provided already at WFH through 2 years and there, (inaudible) the pattern was the biggest difference in factor VIII expression occurred, I think, between 78 week and 91. Week 104 looked more similar to week 91 than it did to 78 and that gives us a little bit of confidence that we're nearing the plateau phase. The other key piece that gives us confidence about the plateau phase is preclinical studies that show the settling settles out in about half of the peak of the plateau -- half of the peak phase. So it will take a while longer to be able to look at individual patients and whether there's individual variation in the settling magnitude or phenomenon, and I'd say stay tuned for the promised 3-year update in the middle of next year as that will be the next time we're reporting activity data. And as regards assessing for the presence of the stable form of the DNA in humans, as you can imagine, that's going to require accessing human tissues, which in this particular case, means the liver of patients. Now ordinarily, hemophilia patients aren't typically -- don't typically undergo liver biopsy because of their bleeding disorder, that world has now changed, and we're in active discussions with investigators to initiate studies of the extent to which there is circularization and whether that does explain variations in factor activity levels. That substudy is not yet begun, but stay tuned for more information as we work with sites to ask the question. And then as far as how are we confident that -- or what is the basis for saying that the stable expression form is in fact a circular form? As we summarized at WFH, there's really an abundant literature about that. That if you look out many years past transgene delivery in a variety of different systems, the only recoverable form of DNA is circular. And there is no evidence that persistent expression can be mediated, particularly in the AAV nonintegrating case, by forms other than circular forms. So taken together, and combined with preclinical studies that we've done demonstrating that Valrox can in fact make circles, can express stably in immunosuppressed animals and can express for a good durable period of time in humans, we believe that the transgene product is taken into liver cells and is now capable of sustainably producing factor VIII message.

Your next question comes from the line of Robyn Karnauskas from Citi.

Not that far from the (inaudible) for the PKU, just a couple of questions, like are you seeing -- do you have any sense now of patients who are getting dosed are being dosed (inaudible) for not just one dose but for multiple. So is it easier to give or more difficult to give than what you thought, given the prophylactic or anaphylactic concerns? And then on Vosoritide, you had this call. I know you wrote in your press release, but (inaudible) in line with your expectations the FDA, post (inaudible) the clinical trial and what are your thoughts on this time, the big takeaway from that meeting?

You're breaking up a little bit. It's difficult to understand your question. I think we figured out the first one, which is -- or I can't speak around for Hank or Jeff, which is what's our early experience in terms of naive -- I guess you're talking about naïve patients in terms of starting and titrating them. I think we've done as I mentioned before (inaudible).

Is that indeed the nature of your question, Robyn?

Yes, the titration and -- the first one (inaudible) seeing patients -- stay in there long, I'm just curious about the whole trend of like (inaudible) patients and worrying about their anaphylaxis risk and whether or not that's going more smoothly than what originally we anticipated.

So based on the number of patients that have gotten all the way through patient referral, reimbursement approval, shipment of first product, scheduling of first visit in the clinic or first injection, which I can say I can count those on 2 hands. The anecdotal reports that I've gotten back from our team that had been in or around those clinic visits has been very positive so far, probably not enough of an end to draw any firm conclusions, but positive signal so far. (inaudible)

The second question was on Vosoritide, that you had the panel and you wrote in the press release that it's in line with what you expected. Have you had a dialogue with the FDA since the panel? Anything else you've learned regarding the trial that you've designed?

Yes. Just to give you a little bit more detail on our take on the panel. The kinds of things that we heard from the panel discussion line up incredibly well with the Vosoritide development program. No surprise that. So just to call out a few things specifically, they want to see randomized placebo-controlled trials. They want to see more of than a single clinical trial and, to remind you, we now have ongoing 2 randomized placebo-controlled trials. They want to see trials being conducted in different age groups. So we have 1 randomized trial in 5 to 14-year-olds and another randomized trial in patients who are under 5 years of age. The panel wanted to see average growth -- annualized growth velocity as the primary endpoint, we have annualized growth velocity is the primary endpoint. The panel recommended that more contemporaneous data on the current natural history of untreated achondroplasia individuals be evaluated. They encouraged sponsors to collect additional data on other endpoints of achondroplasia. For example, sleep apnea or functional measures. And so we left that meeting incredibly encouraged that our development program was very much well on track. And as I said, no surprise because we have done an enormous amount of work with academics and opinion leaders on the critical elements, and I have to say that we're also incredibly pleased that the agency referenced on several occasions data that was published by members of our study steering committee or principal investigators of our study. So I think that means that we're talking to the right people about the right things. As regards the one outstanding item of the duration of placebo-controlled trials, we noted that there was some discussion at the panel about the duration, but the caveat around that was that durability of effect can be ascertained by measuring what amounts to final height gains in treated patients through adulthood by evaluating long-term treatment in a cohort of patients and that also is very much ongoing. And the durability question rests on whether there is expected to be any meaningful attrition in the effect, which will obviously be assessed through final adult height gains. We believe our preclinical biological data as well as our clinical data as well as our biological marker data all support the thesis that there should be no expectation of attrition of treatment benefit, and so we believe the 1-year study is a feasible, ethical, responsible approach to establishing the effectiveness of Vosoritide over the long term. We continue to be in active dialogue with the agency and wherever possible, to make sure that we are on the same page and I think as you saw from that advisory opinion if you look at our program, you see we've done a lot of work to establish the page and get everybody on the same page.

And the follow-up. Durability (inaudible) that affects approvability? So it's not like you get approved and there's a few more issues with reimbursement. Anything that you think -- you actually think FDA would be fine with approving on 1-year follow-up even though you don't have (inaudible) -- what is your thought on that? What does that tell you about the feasibility and reimbursability?

Well, just to remind you, the first 20-something patients that have been treated with Vosoritide have now already been treated for 3 years as of the last R&D -- 30 months since our last R&D day, 42 months as of the next R&D day in a couple of months from now. And if you just sort of try to line up the time lines of when the readout of the Phase III trial's going to be and imagine when the submission is going to be, while we haven't given precise timings for all that, we're going to have many years of durability data on that cohort of patients. And that, together with our natural history data, we anticipate will be sufficient to establish a durable effect of Vosoritide. We're encouraged by the data through month 30. Obviously, stay tuned for month 42, 54, et cetera. And the biomarkers -- J.J. is reminding me, and I kind of take this for granted because we've seen so much of this, it's been such a long time being established that there is no evidence of attenuation of treatment effect. When you look at biomarkers, cGMP, urinary excretion of (inaudible) cGMP or the biomarker that's more precisely related to bone, which is the (inaudible) bone biomarker. So taken together, we have a lot of biological and human data that indicates that the treatment effect should be maintained. That, together with the data that we would be providing in the NDA at the time of submission, should substantiate a durable treatment effectiveness of Vosoritide.

Your next question comes from the line of Chris Raymond from Piper Jaffray.

Just 2 questions. So first, for Hank maybe, on the issue of FDA draft guidance and especially your calling out a target factor activity that should be in the normal range. I guess, Hank, this would seem to be a little bit at odds with at least a couple of decades of experience using factor, where normal is not even close to the range of possibilities. Some folks we've talked to have speculated that this targeting guidance, factor targeting guidance, might not stand. Just kind of wondering if you have any insights on that, especially given that this is not recombinant factor, it's obviously gene therapy. Any thoughts on that? And then on PALYNZIQ, I know it's still early days here, but I think you'd expect hypersensitivity reactions, when they do occur, to happen I think in your clinical experience like 8 weeks or so in the titration phase. Just remind us, I think you saw something like a 15% or 20% dropout rate in the clinical trials. What would you expect in the real world and what are you guys doing to sort of bend that number down lower?

Okay. Chris, we'll let some clinicians wonder if the agency's guidance for factor-based activity regulatory assessments to not stand. Let me remind you that the context in which they talk about targeting normal factor activity levels is in a context of enabling accelerated approval. Full approvals in their language still need to be based on lead rates. The motivation behind that from an agency perspective is that to accelerate the approval of something, they want to believe that the treatment effect is robust, reliable, substantial and provides a reasonably likely evidence -- a benefit that's reasonably -- they want to measure an endpoint that's reasonably likely to commit clinical benefit. Now as I said, when you're looking at factor activity levels that are in the lower range and not understanding the biology of what those factor activity assays are telling you, one could imagine that there are situations where a former 5% -- using recombinant factor replacement therapy -- doesn't have the same meaning as transgene factor expression. So to those clinicians who wonder if that's going to be the sustaining bar or whether the bar is going to be lower, it will take some time to elucidate that as we learn more about the relationship between transgene factor expressions that are under the normal range and bleeding. So I think it'll be a while before -- a long while before that bar could be lowered and it will require fairly substantial sample size investments for companies who are targeting lower factor levels to prove that they have the same clinical efficacy as a product which produces factor activity levels in a more normal range. We like where we are, and it'll be a while before anybody can really rationalize the factor VIII activity level based on a lower range of -- than normal.

So Chris, as Hank (inaudible) this is a regulatory hurdle. It's not a clinical one (inaudible). But it's a regulatory hurdle specifically for accelerated approval.

Got it. And then on PALYNZIQ?

Chris, you kind of quoted what we were telling you on the launch call regarding hypersensitivity reactions and dropout rates through the clinical trial and I think the real gist of your question was, what are you seeing so far? What can you do about lowering those rates and what do you expect in the real world? The first thing I would point out is that we learned a lot with the patients and how to manage through both the induction and titration of patients in the clinical trial. So by the time we wound up coming out of the clinical trial, there had been adjustments made to the protocol for that portion of the study, and we had what we thought was a pretty good induction and titration protocol and that's what was printed in the U.S. labeling, and we reviewed that with you on the launch call. Just a reminder that patients started 2.5 milligrams once a week, they accelerate -- and that's for 4 weeks -- they accelerate to twice a week dosing, then they move to a 10-milligram dose. They do that more frequently, they get to a 10-milligram once a day dose and then they transition up to 20 milligrams once a day. The fastest pace is 10 weeks to get all the way through that process, but our expectation in the real world based on feedback from the clinical trial investigators is that the dose adjustments are managed very carefully and based on tolerability to the patient. So I would expect that the induction and titration phase will go deliberately, carefully and appropriately slowly for patients to reduce the incidence of hypersensitivity reaction and ultimately to reduce the dropout rate in the clinical trials. Now how do we effect that in the real world? The first thing that we're doing is, beyond the label language that I already mentioned, the first thing that we're doing is we're really leveraging the experience of the clinical trial investigator. We've had a number of webinars run by clinical trial investigators for other PK prescribers who don't have experience with PALYNZIQ through the clinical trials. So they're guiding on how to use PALYNZIQ, particularly through induction and titration, and how to monitor for tolerability to manage through the dosing changes. In addition, we've had a fair number of clinical trial investigators doing individual speaking events at clinics that are looking to learn how to use PALYNZIQ. Now we follow that up with rigorous training on the use of PALYNZIQ. That would be done by BioMarin team. That's a mix of sales personnel and MSLs, when there are deeper questions. So this is all under way. Too early to tell other than the anecdotal report I mentioned earlier that so far, for the naïve patients, very small in, early days, but so far we haven't seen any problems pop up.

One thing I would add if I could. We have an active REMS program also. One way to absolutely mess up your discontinuation rate would be to have new sites go rogue on induction and titration. And our REMS program puts our eyes on exactly that, so that we don't have any outlying behaviors by physicians. And just have amplify a little bit on this, to amplify the comments Jeff made on the call. This is BioMarin's first REMS program, so we hired a bunch of REMS specialty types of people and they put in place systems that are acceptable for the physicians. They're not barriers to them wanting to participate or prescribe. And we get a lot of data back about performance in this context. So if anybody's having particular problems, our side of the equation is going to see this very quickly and react very quickly. In the interest of patients, but also in the interest of the commercial business. So I think that's going to be a key to also bending down the -- as Jeff said, having the good experienced people teach the new people, but then also watching the new people that they follow the track record of the good people and don't have to relearn things that have already been learned.

Your next question comes from the line of Matthew Harrison from Morgan Stanley.

Hank, probably 2 for you. I know you said you didn't want to talk about details about your potential plans to accelerate hemophilia filings. I guess maybe if I could ask the question another way and see if you're willing to comment. Can you talk about if you're going to employ an accelerated strategy and if you are, how much earlier do you think that could file? It wasn't clear from your comments to me whether or not you were actually going full in on accelerated strategy.

Well, I think what we can say is everybody can recognize the virtue and value of getting Valrox in the market as quickly as humanly possible. And I think you can read for yourself the FDA guidance document and assess for yourself whether it's likely that BioMarin can pull together the relevant information to support an accelerated approval strategy. And we've been fairly transparent about our findings. For example, we were talking about 1-stage clotting in chromogenic substrate assay discrepancies more than a year ago, more than a year in advance of the FDA issuing its guidance documents. It's not just not news to us. I think we are, you could say we are key drivers of these being the considerations. We want to stop short of talking about issues related to timing and scope and methodology because as I said on a number of locations, this is highly proprietary, confidential information. It's an enormously valuable and competitive market. Our shareholders have invested a lot of capital in making sure that we get the maximum benefit for our patients and our enterprise for them, and we are duty-bound to protect that interest. And so just stay tuned for more.

But at the same time obviously, those guidelines do offer a new opportunity here that we are seriously considering. Is that (inaudible)?

That's very helpful. And then can I just ask a follow-up. Just maybe on your other gene therapy programs. As we think about time, I guess -- in terms of initial data sets from some of those, I mean, how many patients do you think you need from some of those initial programs to maybe see a signal specifically maybe for your, I'll call next generation Peg-PAL program?

That's good. Next generation phenylalanine hydroxylase. (inaudible)

Just to remind you that pegvaliase's a substitution. It's not a naturally occurring human enzyme. And the cool thing about this is we would love to have delivered the (inaudible) the missing human enzyme, but it really would've required an enormous amount of sapropterin for it to work. So the ability to put the PAH gene in the liver is really the central advantage of the gene therapy approach. And so a little too early to talk about patient numbers. What I can tell you is the FDA has approved on more than one occasion an intervention whose primary benefit is to lower blood phenylalanine levels. And I speak here specifically of BioMarin's Kuvan and BioMarin's pegvaliase. Second thing is, is that the effects on blood phenylalanine levels that we observed, for example, with drugs like pegvaliase are quite substantial so that it doesn't require an enormous amount of sample size to definitively prove a reduction in blood Phe levels. And the third thing I would mention is that we've got great, great efficacy data of our PKU gene therapy in preclinical models of the disease. Models that we've run with Kuvan, pegvaliase, and know quite well. Come to R&D Day and we'll talk a lot about the data that we've generated so far and I think that would give you a little bit more insight into what the development path could be. I think what you can tell from the hemophilia gene therapy guidance document, BioMarin is an active and energetic contributor in a proactive way to how the FDA thinks about introducing innovative therapies and I can assure you that we will put ourselves in a similar position with PAH gene therapy, (inaudible).

Yes. And also maybe, Robert, you should add a few things about also the fact that on the manufacturing front, at this time, we're planning to start our PKU gene therapy program basically with our Novato facility here in California. So this is not -- the plan then will be, so we'll start up phase (inaudible) and the first-in-human should be with commercial scale, potentially commercial scale, (inaudible) won't guarantee that potentially commercial scale product. Do you want to add a few words to that, Robert?

Yes. Thanks, J.J. One of the great advantages of having built our own manufacturing capability and demonstrated that in the production of Valrox earlier this year, material that is currently supporting the Phase III study that Hank has alluded to, it gives us a tremendous knowledge base to leverage platform technology for production of other products specifically for gene therapy of a PKU product. What that allows us to do is to, in this case and that is our plan at this point, is to produce the initial clinical material using what we would anticipate to be the commercial-scale production in the commercial facility. That simplifies the manufacturing production quite a bit because we can leverage the experience we have with Valrox. It aligns very well with what Hank has talked about in that seeing a profound potential clinical benefit early on will be not encumbered by having to transition manufacturing or scale up manufacturing somewhere else. The upside of all of that is, I think, this program, like many other programs at BioMarin, has the potential for very rapid clinical development going hand-in-hand with the manufacturing development to produce material to support pivotal studies as well as the commercial launch.

Your next question comes from the line of Phil Nadeau from Cowen.

One commercial and then one follow-up for Hank. On the commercial side, Jeff, can you talk a little bit about how the treatment-naive patients, the non-clinical trial patients that have been recommended for therapy with PALYNZIQ, where they're coming from. Are they coming predominantly from clinical trial centers? Or are there new treatment-naïve physicians, for lack of a better term, who are prescribing? And then second, Hank, for you on hemophilia, curious to get your thoughts on your commitment to filing both the 6e13 and 4e13 doses at the same time. Just seems like 4e13 could be 6 months behind and given the design of the single-arm study with the factor levels not actually quite getting up to normal, at least on the factor levels that doesn't seem to fit as well as with the new guidance document. So is it possible that you file 6e13 first and then figure out what to do with 4e13 later?

So, Phil, this is Jeff. I'll cover the first question, which is for PALYNZIQ in the naïve patient referrals, where they're coming from. We have just to reinforce a little background. We now have greater than 100 prescribers at 55 PKU clinics that are through the REMS certification process and able to prescribe PALYNZIQ. That includes, let's say, a majority of the 32 clinical trial sites, some from the other first-targeted 30 large PKU clinics that are naïve to treatment experience and a handful of clinics that are in that kind of broader group of 125 total PKU clinics. But by the time you get past the first 62, you're starting to get into some smaller clinics. Most of the naïve patients are coming from the clinical trial sites, some from the other larger PKU clinics that have not participated in the clinical trial and at least a few patients from that other larger group of smaller PKU clinics. That's the experience so far.

On your second question, Phil, I think, it's a reasonable assumption to make that we're going to file quickly after getting what we believe will be registration-enabling data for the first dose group that crosses the finish line. Based on the comments that I gave, the 4e13 group enrollment is -- even though it's a study that's approximately threefold smaller, we still fully expect the 6e13 dose study to enroll much faster. The interest level in the 6e13 dose group is extremely high. So unless some major dynamic were to shift, I would anticipate that 6e13 will be ready to be filed well in advance of 4e13 and therefore, wouldn't hold up availability of Valrox in general waiting for the second study.

(Operator Instructions).

(inaudible) You made a good comment that 4e13 might not qualify with the new guidelines for accelerated filing because it's potentially below the 50%. And that would also apply to our competitors.

(Operator Instructions). Your next question comes from the line of Ying Huang from Merrill Lynch.

Just one on the hemophilia program. Hank, I think I heard you saying that the next update from the patients in the Phase I/II study for Valrox would be at maybe sometime next year. So does that mean you're not talking about this at your November 7 R&D Day? Or we won't get any clinical update at ASH? And why is that?

So yes, that's what it means. And why? Because we're now moving into a more chronic phase of gene expression. And the issue really is durability and that's really going to be decided by clinical outcome. As I mentioned, it's not crystal clear what factor activity level translates to what reduction of bleeding using the transgene, that's why we want to get it as high as possible, take that question out of the equation. But in x years from now --and I don't know when x is and it may not be the same for every patient -- we may start to learn about what the relationship between lower factor activity levels and breakthrough bleeding events are. As J.J. just pointed out, part of the reason that lower factor activity level studies, including our own 4E13, are not going to be eligible for accelerated approvals because of the uncertainty about what lower factor activity levels mean for bleeding. So what we want to do is we want to observe bleeding frequency and relate that back to factor activity levels, not the other way around. That's the principal reason for running the experiment the way that we're running the experiment and reporting the data the way we were reporting it.

Your next question comes from the line of Tim Lugo from William Blair.

This is Ashiq Mubarack on for Tim. So as you -- to follow up on the last question, as you continue to look at long-term Valrox durability, if Factor VIII activity continues to contract kind of moving forward, at what point would you consider a multiple-dose strategy? And would you reconsider your dosing levels? And in that scenario, how would you be limited by your manufacturing capacity?

Well, the subject of redosing is important, not so much driven by potential contraction, because the data that's been collected in Factor IX hemophilia in humans or -- hemophilia in dogs, does indicate that once you get the transgene in there, it stays in there. People talk about liver dilutional effects and we don't really see that as being an adult phenomenon, but it is an important pediatric phenomenon. And that's where redosing could become really important. And so for that reason, we are keen to develop some concepts around redosing, but that's, for us, a bit down the road. And I don't have really concrete thoughts to pass along about what you'll see and when you'll see it just yet because we're really most focused on running our Phase III program and getting to registration. And I should remind you -- I should have said this when Ying asked the previous question -- that it's not just that you're not going to see the Factor VIII data from either the Phase I/II study in the interval time points or from GENEr8-1 and GENEr8-2. We're in lockdown mode. And so none of us see the data either. It's a really important point as it pertains to study integrity and protecting the interpretation of the results. I hope you all understand and bear with us as we go through this phase of our program. We are keen to know, too, but we need to run the experiment in a high quality way. This is going to be a landmark study of its kind. It's going to change medical practice. These data must be high quality.

Okay. That makes sense. And just quickly on the FDA draft guidance. Did you hear anything or see anything of interest related to your potential PKU studies and is there anything that kind of made you reconsider how you might design those studies moving forward?

Not particularly. Like I said, I mean, it's a gene therapy for hemophilia guidance and other rare disease guidances and whatnot. For a company like BioMarin, this has been our living for 20-something years, expediting the development of drugs for patients with terrible rare diseases. So, honestly, I feel like we see our language in these guidance documents is the real answer. So I suspect as we get closer to the development strategy and whatnot, we'll be working closely with the agency to identify the most facile path. And as I said, it won't surprise me if we see more and more proclamations coming out that look a lot like BioMarin's programs.

Your next question comes from the line of Marty Auster from Crédit Suisse.

Probably going to be obsolete in a few years as you develop gene therapy, but curious about, post the US approval of PALYNZIQ, if you've got any update or comments on the conversations with European regulators. And then curious, for Jeff, is PALYNZIQ's efficacy enough to kind of create a very meaningfully different value proposition ex U.S.? Obviously, Kuvan has been kind of a majority, a majority of that value has been in the US. I'm curious if you think PALYNZIQ will have a larger proportion of its value kind of recognized globally.

On the regulatory process, what I can say is we've publicly communicated when we filed and if you're familiar with the European regulatory process, you know there's a critical milestone about 4 months after a submission, and we're poring over the review comments that we have seen to date. I would say that we're not done, on the one side. On the other hand, we are very optimistic that as we got through the FDA, so too shall we get through the EMA. As we got through the FDA in the first cycle of approval, we are optimistic that we will get through the EMA in a first cycle approval. Now, in the future, if I use the term cautiously optimistic, please don't overinterpret that because we're in the middle of the process here. Our expectation and belief is that we'll be able to bring PALYNZIQ to European patients through an approved MAA, I think we're saying the second half of '19. So nothing to report that worries us that we're going to be on that time line.

And with respect to value proposition for PALYNZIQ ex U.S., that's a big subject, could talk about it for a long time. But I think what you're pointing out is the environment ex U.S. is substantially different than where we're currently launching PALYNZIQ in the U.S. For BioMarin's portfolio, the U.S. market is the best and most receptive market in terms of a payer environment, on the one hand, and it's also where we spent over a decade developing the PKU market with both payers, prescribers and professional associations taking a pretty important stake in how PKU is treated up to and including using therapeutic intervention. So U.S. is the best. As it relates to Europe and other markets, it's been great for the last 2.5 years. So we've recovered rights to both Kuvan and PALYNZIQ. I can tell you that we've used that 2.5-year period to develop how other markets are thinking about PKU, both prescribers, patients and payers. It's notable that the E.U. treatment guidelines were published after BioMarin recovered rights to the PKU franchise. And also worth noting that the European systems tend to have more developed and specific programs that you need to go through for price and reimbursement approvals. So we are working hard to prepare. Different systems behave differently. So U.K. has an incremental cost-effectiveness program, France assesses value based on innovation and clinical -- degree of clinical response, for example. So there are different systems and we are working to develop our strategies for each of the major EU systems. But fundamentally, PALYNZIQ has a really powerful value proposition. It's rooted in the burden of illness and remaining unmet need for adult PKU patients and the second piece is the profound reduction of phenylalanine. So we're moving from a world in Kuvan, in which we seek to impact phenylalanine levels based on response, to the world for adults with PALYNZIQ, that we're really moving towards phenylalanine control, and that's the power of PALYNZIQ.

Your next question comes from the line of Laura Chico from Raymond James.

This is actually Timur Ivannikov for Laura today. And we have another question about Valrox. How should we be thinking about the minimum level of duration for Valrox? We don't know if you have had payer discussions and based on your market assumptions, what do you think is the minimum durability of Valrox treatments?

So, your question is related to payers? Specifically to payers?

Not necessarily specifically to payers, but maybe about your market assumptions in general, what would be a feasible minimum duration of this treatment?

I think payers are the end of the line. It starts with what kind of durability are you seeing out of your clinical development program. It next moves to what kind of durability will regulators require. And then finally, what are the expectations around durability from payers. So payers are saying, "Look, we will pay based on an expectation of durability, which will be built into an eventual price tag." I'm not suggesting this would happen, but if you went to a payer system, and you said that "our gene therapy will last for exactly 36 months at 100% and after that it will go down to 0", payers would be able to say, "Fine," and we'll negotiate a price based on an expectation of 36 months of response. But I think Hank...

(inaudible) because we are doing a lot of payer research, evaluating, and actually no payer has actually said something related to, "Well, if your product doesn't work more than x months or x years, we're not interested in covering it." Nobody ever says that. At the same time, I think we have communicated in the past that with respect (inaudible) by the time we get approval we are probably going to have probably 4 to 5 years of historical data on several patients and so we'll be able to document whether we need to know (inaudible). By the way, the effectiveness will be mainly demonstrated here by bleeding rates not by Factor VIII. Assuming we still have no bleedings after 4 to 5 years, we'll have visibility on that by the time of launch. We have the feeling also based on earlier payer research that it might be difficult anyway to potentially base the reimbursement negotiations on more than 5 years of value of the current treatment. So that's another way to kind of answer your question. Although it's not really expressed that there is a minimum, there is no real minimum expressed so far.

There was a regulatory or clinical component to the question. I think the thing I'd say about that is we presented our 2-year data to a major health authorities, and one of the striking comments that one of the health authorities made to us was something at 2-year mark -- was something like, "No breakthrough bleeding, all target joints resolved, give me some of that if I'm a hemophilia patient."

Your next question comes from the line of Kennen MacKay from RBC Capital Markets.

I was wondering, with the recent approvals of Brineura and PEG-PAL, if there was any chance at some point you can help with longer-term guidance sort of beyond the Kuvan 2020 initial staged (inaudible). Caveating, obviously, that there are some moving parts with sort of Vosoritide and Valrox, and if not, sort of what maybe when we could see that or what the moving parts are sort of preventing that?

I will start and probably Dan can step in. First of all, we only give 1 year guidance when we report Q4 the previous (inaudible) during the first quarter of the actual year. We will continue to do that. I think today, we have given a bit more longer-term guidance speaking in terms of revenue since we say we anticipate that our currently commercialized products will generate around $2 billion of revenues by 2020. So that's some -- beginning of long-term guidance. And (inaudible) Dan (inaudible) just as a reminder that the generic (inaudible) for Kuvan is in Q4 of 2020, is only in the U.S. market. Ex U.S., we are predicting year 2024.

I don't think I have a lot to add to that at the moment. 1-year guidance is the guidance we feel most comfortable and appropriate. We will think about others down the road. But don't anticipate doing it at the moment.

Your next question comes from the line of Gena Wang from Barclays.

This is Parag on for Gena. Just a couple of very quick questions since most of the questions were already asked. Number one, on your trial for the AAV5 patients with AAV5 antibodies, could you give any additional color on where that is? And number two, if you could give additional color on what kind of data how many patients are we going to see for the MPS asset at the SSIEM.

For the study in AAV5-positive patients, also known as the (inaudible) patients, also known as the [203] study is underway, but it is a slow study. It's slow because we are treating patients who have the weakest titer first followed by patients who have stronger titers and then stronger titers yet. And we also want to measure factor activity level for several weeks before treating subsequent patients and this is designed so that in the event that pre-existing immunity does affect transgene expression in a complete way, that we don't expose a lot of patients to Valrox when there is no chance for it to work. So 203 is a slow study and all I can say is we have started, not really give you an enrollment update or time line for completion anticipation. And SSIEM, we have enrolled a full natural history cohort just to remind -- MPS IIIB. Just to remind you, the study has a run-in period essentially of a year -- that's the part that we have fully enrolled. And then after that year of observation -- so that completed in April of '18. A year of observation will complete at '19, at which time all patients will roll over to the active drug and be treated for a year. So as of April of '20, all -- roughly 20 patients have been treated for a year -- they'll have a year of their own natural history data untreated, and a year of treated data. Therefore at SSIEM we're still relatively in the early side of all of that having just completed enrollment in the run-in study. We reported on 3, I think, patients at the World Congress in January of this year -- we reported on 3 patients, complete normalization of heparan sulfate levels, normalization of organ size, some signs of stabilization of brain cognitive function. We'll update that with a few more patients, but we're not really in a position to be definite about the direction that 250 -- (inaudible) is heading. So I hope to see you at the conference and we can talk about the data.

Your next question comes from the line of Vincent Chen from Bernstein.

Couple of quick ones on Valrox. One on manufacturing, one on the development path forward. So on manufacturing, were all the patients in the initial 6e13 group treated with the same level vector? I don't mean just one manufacturing run, but one vat of product made it in multiple runs and mixed together prior to dosing? Or did different patients get product from different manufacturing runs?

Robert, you want to answer that?

We produced a number of different runs and patients have been treated with different production batches, both from our initial clinical process, which was the genesis of the Phase I/II results we've presented in May. And we have produced several batches of material that will be used throughout the Phase III clinical trial, so that we will have, at the end of the day, clinical experience with a number of different batches made by different production processes and scales and facilities and our analytical characterization gives us a strong sense of the comparability of that material. And so we will not be basing our entire clinical outcome on the production of a few batches. There will be a significant number of batches produced and patients treated with that material.

And then could you provide some additional color on the data supporting your plan to move into a trial treating hemophilia A patients with pre-existing antibodies. What is the data that supports that? And, I guess, what are the implications that this data would have on the ability to re-dose gene therapy?

We presented at ASH last year a poster, I think it was, on the treatment of nonhuman primates who have demonstrable titers of antibodies to AAV5 and we showed that there is an attenuation of transgene expression in patients with pre-existing humoral immunity to AAV5, but not a complete and total elimination of transgene activity. And so that's the scientific basis. And underneath that concept is that we're presenting a massive amount of vector and capsid protein to a swampable immune system and the idea is to break past the defenses of the immune system to get the gene in the liver. And so that's the scientific basis. And the second part of the question, remind me?

Just what will be the implications of this data with (inaudible) ability to redose?

So at one level, one implication could be Valrox for everybody. At another level, it could be that there is a requirement for a companion diagnostic to identify patients to whom benefits and risks of Valrox is not the same. And for example, AAV5-positive patients versus AAV not -- naïve patients. In the event that a companion diagnostic is required for development, we're well underway in that regard. We've already contemplated the possibility that, that 203 study may not work, in which case we would have to have a companion diagnostic, by FDA guidance (inaudible) registration and we're well on our way towards having that as a backup if we need it.

There are no further questions at this time. I'll turn the call back over to CEO and Chairman, Mr. J.J. Bienaimé.

Thank you, operator. Thank you all for participating in our call today. So to conclude, we believe we delivered another very solid quarter, and we are on track to reach approximately $1.5 billion in full year revenues in '18. We remain on track for full year non-GAAP income between $100 million and $140 million. And with PALYNZIQ now approved in the U.S. and on track for an EU approval in the second half of next year, we believe we can generate approximately $2 billion of revenues in 2020. And that's a little bit of long-term guidance. We currently -- commercialized product only. So we look forward to providing you an update on specific metrics on the US commercial launch of PALYNZIQ in the third quarter and also at R&D Day a few weeks after that. So thank you for your continued support and goodbye.

This concludes today's conference call. You may now disconnect.