福泰製藥 (VRTX) 2017 Q3 法說會逐字稿

Good evening. This is Michael Partridge, Vice President of Investor Relations for Vertex. Welcome to our third quarter 2017 financial results conference call. (Operator Instructions) This call is recorded. A replay will be available later tonight on our website.

Dr. Jeff Leiden, Chairman and CEO; Dr. Jeff Chodakewitz, Chief Medical Officer; and Ian Smith, Chief Operating Officer, will provide prepared remarks this evening. Stuart Arbuckle, Chief Commercial Officer, will join us for Q&A.

We will make forward-looking statements on this call. These statements are subject to the risks and uncertainties discussed in detail in today's press release, our 10-K and other filings with the Securities and Exchange Commission. These statements, including those regarding the ongoing development and commercialization of KALYDECO and ORKAMBI, Vertex's other cystic fibrosis programs and Vertex's future financial performance, are based on management's current assumptions. Actual outcomes and events could differ materially. Information regarding our use of GAAP and non-GAAP financial measures and the reconciliation of GAAP to non-GAAP can be found in the financial results press release and in tonight's webcast slides, which have now been posted on our website.

I will now turn the call over to Dr. Jeff Leiden.

Thanks, Michael. Good evening, everyone. 2017 has been a tremendous year for Vertex. We've made significant progress across all parts of our business and moved closer to realizing our long-term vision of treating all people with CF. During the year, we've increased the number of people eligible for and being treated with our approved medicines. More recently, we showed data from multiple triple combination regimens that demonstrate that we may be able to treat up to 90% of all CF patients. As a result of treating more people with our approved medicines, CF product revenues have increased quarter-to-quarter throughout this year. And tonight, we are increasing our full year guidance for total CF product revenues. Our financial strength has substantially increased, and we expect this revenue and earnings growth to continue going forward, creating significant value for shareholders.

Let me now briefly review our progress over the last 3 months. First, KALYDECO. On August 1, the FDA approved KALYDECO for use in more than 600 people with CF ages 2 and older who have 1 of 5 residual function mutations in the CFTR gene. KALYDECO continues to be a transformative medicine and is now labeled to treat approximately 6,000 people with CF globally. And we continue to increase the number of patients initiating treatment with KALYDECO.

Second, tezacaftor/ivacaftor. Based on the positive Phase III data we announced earlier this year, the FDA and the European Medicines Agency have accepted our applications for the use of the tezacaftor/ivacaftor combination treatment studied in 2 different CF patient populations: one, people with CF ages 12 and older who have 2 copies of the F508del mutation; and two, people who have 1 F508del mutation and 1 residual function mutation that is responsive to tezacaftor/ivacaftor. In the U.S., the FDA has granted priority review of the NDA and set an action date of February 28, 2018. We currently expect approval in Europe in the second half of 2018. Our commercial team in the U.S. is currently preparing to bring this medicine to the many patients who are eagerly awaiting its approval.

Third, our triple combination regimens. In July, we announced clinical results for 3 different regimens, which included next-generation correctors, VX-152, VX-440 and VX-659. Across these regimens in 2- and 4-week studies, we saw significant clinical activity in both het/min and F508del homozygous patients across multiple measures: increases in FEV1, increases in sweat chloride and in the one study in which it's been evaluated thus far, increases in CFQ-R. We also saw good tolerability. We expect to continue to collect Phase II data for next-generation correctors, VX-152, VX-659 and VX-445, in triple-combination regimens into early next year. And pending data from these studies and discussions with regulatory agencies, we expect to begin pivotal development in the first half of 2018 of up to 2 triple-combination regimens. We expect that each of these triple regimens may have a different next-generation corrector and one of the regimens may be dosed once daily. These triple regimens may provide the opportunity for 90% of the CF population to be treated with a medicine that addresses the underlying cause of their disease.

We continue to invest in scientific innovations to bring forward life-changing medicines. We believe this creates the most value for patients, employees and shareholders. We're seeing good progress both internally and with our collaborations, and we expect to advance new medicines into clinical development in 2018. We look forward to updating you on this progress as well as helping you understand how we select our disease areas and potential investigational medicines.

We're very proud of the significant progress the company has made in increasing the number of people eligible for and being treated with our medicines around the world. And we are well positioned to meet our goals of consistently creating transformative medicines for patients and delivering significant and sustained revenue and earnings growth for our shareholders.

I'll now turn the call over to Dr. Jeff Chodakewitz to make specific comments on some of our development programs.

Thanks, Jeff, and good evening. We're very pleased with the clinical progress we've been able to make across multiple CF studies this year, which are bringing us closer to our goal of treating all patients with CF.

First, I will review results from 3 different studies in CF that recently completed, then I will make comments on our ongoing triple-combination programs. I will start with the ORKAMBI clinical results. We are pleased to have announced tonight Phase III results of ORKAMBI in children 2 to 5 years of age with CF who have 2 copies of the F508del mutation. The primary endpoint of the study was safety, and the results showed that the treatment was well tolerated. In addition, the study also showed improvements in CF-related disease measures, including sweat chloride and nutritional status. These are the first findings to show the safety and benefit of ORKAMBI in children as young as 2. CF is a devastating and progressive disease, and our goal has been to evaluate our medicines in younger and younger patients so that we may establish a basis for intervening earlier with medicines that change the course of the disease. Based on the results from the ORKAMBI 2 to 5 study, we expect to submit for regulatory approval by regulatory authorities in the U.S. and Europe in the first quarter of 2018.

Now to tezacaftor/ivacaftor. We announced top line data from a Phase III study evaluating the addition of tezacaftor in people with CF ages 12 and older with 1 copy of the F508del mutation and a second gating mutation who were already receiving ivacaftor monotherapy. The study did not meet its primary endpoint as there was no difference in FEV1 for those who added tezacaftor in addition to ivacaftor that they were already taking compared to those receiving placebo in addition to ivacaftor. Importantly, safety data from the study showed that the combination was generally well tolerated and the safety profile was consistent with that observed in prior studies of the tezacaftor/ivacaftor combination.

For secondary endpoints, we also looked at sweat chloride and CFQ-R. In sweat chloride, we saw a decrease of approximately 6 millimole compared to ivacaftor monotherapy that was statistically significant, and no difference in CFQ-R. Based on these results, we do not plan to seek approval for tezacaftor/ivacaftor in gating patients who also have an F508del mutation. The vast majority of all patients with a gating mutation around the world are receiving KALYDECO, and we expect that they will continue to do so.

Now to our ENaC inhibitor, VX-371. We've also announced the results from a Phase II 28-day study of an inhaled ENaC inhibitor, VX-371, when added to ORKAMBI. The study dosed 142 CF patients ages 12 and older who were homozygous for the F508del mutation. The study did not meet its primary efficacy endpoint. In these patients who were already receiving ORKAMBI, the addition of VX-371 did not produce changes in FEV1 when administered with or without hypertonic saline. Safety data from the study showed that the addition of VX-371 to ORKAMBI was generally well tolerated and the safety profile was consistent with that observed in prior studies of VX-371 monotherapy. We continue to conduct a Phase II study of VX-371 monotherapy in patients with primary ciliary dyskinesia.

I will conclude with some remarks on our triple-combination regimens, as they continue to advance. We are well positioned with 4 different triple-combination regimens in Phase II, 3 of which have already demonstrated significant clinical activity and good tolerability in 2- to 4-week studies. We recently announced that we have included the experimental potentiator, VX-561, formerly CTP-656, into the Phase II studies of the next-generation correctors VX-659 and VX-445. VX-561 could play an important role in a future once-daily combination regimen that treats the underlying cause of CF. We remain on track to begin pivotal development for up to 2 triple regimens in the first half of 2018, and I look forward to updating you on our plans.

With that, I will now hand the call over to Ian.

Thanks, Jeff, and good evening to everyone. 2017 has been a year of significant financial growth. We are poised to deliver continued revenue and earnings growth into 2018 and beyond. Tonight, I will discuss our third quarter 2017 financials, review our 2017 full year financial guidance and make further comments on the strength and the trajectory of our business.

Revenues first. Total CF product revenues of $550 million in the third quarter of 2017 reflects strong performance based on the demand for our medicines. This represents a 34% increase compared to $410 million last year. Our product revenues have grown each quarter throughout 2017 as we have increased the number of patients treated with the approved medicines. We expect this to continue in 2018 as we gain additional reimbursements in countries outside of the U.S. and we also have the potential to launch new medicines such as tezacaftor/ivacaftor.

Both ORKAMBI and KALYDECO have demonstrated strong growth in the quarter. For ORKAMBI, we reported third quarter 2017 product revenues of $336 million, an increase of $102 million compared to the third quarter of 2016. The quarterly growth this year has been driven by the continued uptake of the medicine globally, most specifically in the children ages 6 to 11 in the U.S.

Third quarter KALYDECO sales were $213 million compared to $176 million for the third quarter 2016. The quarterly growth this year has been driven by the strong uptake for the medicine by patients in the U.S. with residual function mutations following the recent FDA approvals.

Our third quarter 2017 non-GAAP combined R&D and SG&A expenses were $334 million compared to $295 million in the third quarter 2016. This increase was primarily due to the continued acceleration and advancements of our portfolio of triple-combination regimens for CF and the investment to support the launch of our medicines globally. Strong growth in product revenues is driving an expansion in our operating margins. Operating margins were 26% for the third quarter 2017 compared to 16% for the third quarter 2016. And with our anticipated revenue growth, we expect our margin to continue to expand in the future.

Non-GAAP net profit in the third quarter 2017 was $136 million compared to non-GAAP net profit of $43 million in the third quarter of 2016. The significant growth in non-GAAP net profit was largely driven by the strong growth in total CF product revenues.

We have a strong balance sheet as we ended the third quarter with approximately $1.8 billion in cash, cash equivalents and marketable securities. This compares to $1.4 billion at the beginning of 2017. And during the third quarter, we paid $160 million to Concert Pharmaceuticals upon completing our asset purchase agreement for CTP-656, now VX-561.

Now turning to the full year financial guidance. Based on the strong underlying demand of our CF medicines and recent label expansions, we are increasing our total CF product revenue guidance to $2.1 billion to $2.15 billion in 2017. This was previously increased to $1.87 billion to $2.1 billion in August of this year. For ORKAMBI, we now expect $1.29 billion to $1.32 billion in net product revenues, which reflects the strong underlying demand for the medicine throughout the year among the people with CF ages 6 and older in the U.S., and is based on potential revenues in countries where ORKAMBI is currently reimbursed. I would note that this guidance does not assume the recognition of any ORKAMBI product revenues in France in 2017.

As to KALYDECO, we now expect $810 million to $830 million in net product revenues based on recent label expansions and the strong underlying demand we are seeing by eligible patients. As we think ahead to potential 2018 revenue guidance, we are mindful that the expected approval and launch of tezacaftor/ivacaftor will make it challenging to forecast revenue for individual products. Therefore, in early 2018, we anticipate providing full year total CF product revenue guidance but no individual product guidance.

Now to operating expenses. We continue to expect combined non-GAAP R&D and SG&A expenses of $1.33 billion to $1.36 billion in 2017. As we plan for 2018, a key investment driver will be the initiation of pivotal studies for up to 2 triple-combination regimens, including clinical drug supply and the supply chain investment for potential commercial success of a triple-combination regimen. Therefore, non-GAAP operating expenses will see growth in 2018. At the same time, we do expect operating margins to expand in 2018 as our revenue growth will significantly exceed any increase in our operating expenses.

In conclusion, we are well on track to deliver a financial profile that includes high operating margins and sustainable earnings growth. Our success in increasing the number of CF patients that we treat with our medicines is driving the strong financial performance we have reported today. We are well positioned to realize our goal of treating many more patients and enhancing the benefit for those we treat with the CF pipeline assets that we are rapidly advancing in development. We look forward to continue to share our progress with you in the weeks and months ahead.

And with that, I open the line to questions.

(Operator Instructions) Our first question comes from the line of Michael Yee of Jefferies.

Two-part question. One was, as you think about the triple and the strategy going forward, is there an expectation that the overall clinical Phase III program should pretty much look like ORKAMBI in both het/mins and homozygous in terms of a duration of the study? Or do you think you could figure out a way to accelerate that? And then the second question is around the uptake of 661 next year. While I appreciate it's early, are there opportunities there to significantly grow the overall doublet franchise? Where are the opportunities to grow that? Do you think it cannibalizes a bit? How do we think about that market?

Michael, thanks for the question. Maybe given my comments, we expect to collect all that -- the data on the triple-regimen portfolio. We expect to have a really good view into that in early 2018. At that point, we can have a more, I'd say, detailed discussion. But for the strategy of how we're approaching the pivotal studies, the time lines, Jeff Leiden and maybe Stuart can talk to you about how we think about tezacaftor/ivacaftor launching in 2018.

Yes, Michael. It's Jeff. Let me sort of give you a high-level view. Again, we're just in regulatory discussions now, so obviously we won't be able to give you the final point of view until we finish those. But a couple of points, I think, worth making. One is, as I said and Jeff Chodakewitz said, our current plan, obviously, which will be based on the data, would be to take up to 2 programs forward. And those 2 programs would have different next-gen correctors in them. And our hope, again, based on the data that we'll see from Phase II, is that one of those programs might be a once-a-day regimen that would incorporate VX-561. That's one point. The second point is sort of to your question directly and the way you phrased it, which is we've obviously learned a lot as we've gone through multiple Phase III trials in CF. We've learned what the length of those trials needs to be, for instance, let's say, 8 to 12 weeks for collecting acute data and up to 24 weeks for long-term data. We've learned about the size of those trials, how they have to be powered to get the effects we want, and we've learned a lot about how to collect those endpoints.

So our strategy overall is to use all of that knowledge and, particularly, based on the strength of the results that we're seeing with the triple, to streamline these trials with the regulatory agencies as much as possible, so we can get these medicines as quickly as possible to patients, particularly the het/min patients who don't have anything right now. So more to come as we finalize those discussions. We'll certainly be able to explain those to you in the first half or the beginning of next year. But our intention is to streamline those trials and to think about how we get these drugs to those patients who are waiting for them as quickly as possible.

And Michael, it's Stuart here. On the tezacaftor/ivacaftor, whether we think it's got the opportunity to grow the CF franchise, we certainly do. We think given the benefit-risk profile that it has, it really provides a tremendous new treatment option, which is going to allow more patients to be treated with a CFTR modulator. There's probably a couple of different kind of groups of patients I'd point you to. The first would be in the homozygous population. As you know, with ORKAMBI there, we've had a number of patients who've discontinued ORKAMBI, often due to adverse events. Given the benefit-risk profile of tezacaftor/ivacaftor, I think that's a population which is likely to see high demand for tezacaftor/ivacaftor. We also know that in the homozygous population, there's a number of patients who've never been initiated on ORKAMBI, often because of their views of the benefit-risk profile, and I think that's another population that potentially could be excited by a new treatment option. And then the third one would be in the residual function population. We obviously have the indication here for KALYDECO for those ivacaftor-responsive mutations. Not all of those patients are yet on KALYDECO, and so for those who are naïve, again, tezacaftor/ivacaftor could be a good option for those patients. And obviously, outside of the U.S., further down the line, the same in the homozygous population, but the residual function patients there outside the U.S. don't have any approved treatment to treat the underlying cause of their disease. So we certainly see tezacaftor/ivacaftor as a very important new treatment option, which can grow the number of patients being treated with a CFTR modulator.

My final follow-up is just, is there anything magic about 6 months that they ask for ORKAMBI or why it couldn't be 3 months? Is there any magical reason that those were the numbers picked?

Yes, Mike. I should have been a little clearer about that. There are 2 different -- or 3 different considerations. So there's the acute benefit. And as I said, we know we can see that pretty quickly within, certainly, 8 to 12 weeks. There's the long-term benefit in things like pulmonary exacerbations, hospitalizations, antibiotic use. When you look at that, that typically is going to take 6 months. And then there's, of course, the safety database, which we need to provide an adequate one. Having said that, obviously, what we're going to try to do is to come up with a regulatory strategy which fits that point of view, which means that we don't necessarily believe that we need to wait all the way to the end of all those measurements to file regulatory submissions, particularly for the patients who don't have anything today.

Our next question comes from Matthew Harrison of Morgan Stanley.

I was hoping to ask sort of a non-CF pipeline question, and maybe I could ask it in 2 parts. So I guess, one, could you talk about some of the milestones that we might see over the course of next year for the non-CF pipeline and maybe specifically about your spinal cord injury asset? And then just help us think about how you plan to invest either internally or externally into the non-CF pipeline.

Yes, Matt. This is Jeff Leiden. I'll take the first part, and maybe Ian and I can work on the second part together, sort of a BD question as well as an internal question. First, I'd remind you of something we've said many, many times, which is that we believe, at Vertex anyway, that investment in scientific innovation is where the real value is created in this industry, and that includes both our internal investment in R&D and our external investments in BD. With respect to our internal pipeline, we do expect to bring those programs -- and we've talked about 4 or 5 of them. We've talked about AAT. We've talked about adrenoleukodystrophy, sickle cell disease, pain. We've talked about 4 or 5. Those programs are actually advancing quite nicely. And what we've told you before is that as they enter the clinic, and we expect some of them to enter the clinic in 2018, you'll hear more about them. So I think you can anticipate hearing more about those programs next year. With respect to the external investment, you've seen us do a number of deals that both strengthen our technology platform, if you will, and also bring in early-stage products. You'll see us do more of those. But maybe -- Ian does a nice job of sort of summarizing our strategy for BD investments, which hasn't changed for the last several years and won't change going forward, except the fact that we have more financial firepower now.

Yes, sure. So I would actually start by, obviously, the company is in very different position today than it was a couple of years ago. One, where we were in capital preservation mode as we invested to create our medicines, whereas today, we do have significant cash flow. And I pointed in my prepared remarks that we have approximately $1.8 billion of cash on the balance sheet and without taking out any debt. And so we do have a significant access to capital, and to Jeff's point, I believe that we have to invest to create. And so externally, we have a strategy where it's got 3 points of approach. First is we do look at everything in CF. We should and we try to identify the approaches that are complementary to our own approach, where you have the disease-modifying agents, and they may be more orthogonal approaches as compared to CFTR modulation. But we really look at everything in CF. And we've done a couple of deals in that area, and most notably was the one we've done in Q3 with Concert, where we have the opportunity to potentially improve a regimen by going to a once-a-day regimen. And we'll see how that study plays out.

Secondly, as Jeff mentioned, scientific footprint. Vertex has been tremendously successful with small molecules over the last 20 years or so. But there's been a lot going on outside the world of Vertex, and we think it's time for us to plug into that. And we've already done transactions to give a different kind of modalities of actions to approach the diseases that we're interested in. And most notably, we've done collaborations with CRISPR Therapeutics, which is going very nicely, and we'll probably touch on that a little later on this call. And also Moderna. And then the third area is kind of more opportunistic, looking at disease areas -- still looking at disease areas that are consistent with who Vertex is and the disease areas that we focus on. And we look at those opportunities as well and consider whether to apply our capital. They tend to be earlier stage. I don't want people getting carried away with a large-scale deal at this point, but they tend to be earlier stage. So we can match our scientific approach and creation to developing these medicines, consistent with how we've done things in CF. So that's our approach, and applying enough capital as we generate it is important for us to grow our business.

Okay, perfect. And then could I just ask specifically, I mean, will we see data on the spinal cord asset next year?

Matt, it's Jeff Chodakewitz. Really, we don't expect that. I just would remind you that these are long-term studies and that the patients get treated acutely, and then we follow them for quite a bit of time. It's all going to depend on the rate of enrollment. And then even once the patient is enrolled, it takes some time to get data. So it's just really too early to tell, but I wouldn't expect any next year.

Our next question comes from Geoffrey Porges of Leerink.

Ian, a quick financial question, which is you mentioned that the revenue in France is still excluded from your updated guidance. Can you give us a sense of when -- what the cumulative cash from France is now and when you think that, that might actually be converted over into reported revenue? And potentially -- presumably, that will go through your non-GAAP numbers. And then secondly, maybe Jeff could help us understand the failure of tez/iva in the gating mutations. And how does that fit with the observed change in the sweat chloride and with the activity on the preclinical assay? And does that increase or decrease your confidence in the basis for assessing the triples?

Thanks for the question, Jeff. Let me just -- it's a complex answer, and we can get together after the call as well, so I'll just put my CPFA hat on. So first of all, France, we do not record revenue. We are getting paid for our medicines in France because -- through an early-access ATU program, and we accumulate that cash on the balance sheet. But because we don't have a definitive price in France, we have to keep that on the balance sheet because we may have to give some of it back if we agree to a price that's lower than the list price they're currently paying us now. If we were to receive an agreement for reimbursements in France in 2017, revenue recognition would allow that amount that's in the balance sheet to be recorded in the quarter that we got the reimbursements approved, and that would be a bolus. And it's about $130 million, $140 million that would actually be recorded on top of the revenues you would record going forward.

Now, the complexity in all of this is that if December 31 passes and we're now at January 1, there is no bolus in 2018 because the revenue recognition rules are actually changing. And this applies to all companies, not just Vertex, and you're probably aware of it already. But we will have to make an estimate, as of January 1, of what the value is we will be reimbursed at. That bolus of revenue that now sits -- or bolus of cash that now sits on the balance sheet actually goes through retained earnings. It does not go through the revenue line in 2018. So what happens in 2018, you don't get to record the backlog of cash. What you have to do is, going forward January 1, just start recording revenues as if you're getting paid for your medicines in France. Hopefully, that -- tried to keep it simple, but that's how we proceed. There will not be a bolus recognition in 2018. It will go through retained earnings, unfortunately.

So Geoff, this is Jeff Leiden. I'm glad I got the easier part of the question. You really have 2 questions in there about the tez/iva results. Obviously, we've looked at this data very carefully. We believe that there are 2 reasons that sort of fit together, as you'll see, that explain why the study didn't show a clinical benefit. The first reason and maybe the most important reason is these patients who entered the study in KALYDECO were extremely well treated, as you might have expected. You can see that in their baseline characteristics. So the average patient has been on the drug for more than 3 years. They had a sweat chloride of 50, which is actually quite close to carrier levels, and they had a BMI of 24.5, which is normal, essentially. And so the very high bar that KALYDECO had set with those patients made it difficult to identify incremental acute improvements in FEV1 simply because when they entered the study, they were doing so well. The thing that goes along with that and the other part of that equation is, in this study, we're obviously adding a single first-gen corrector, tez, which only incrementally improves CFTR function. And you can see that clearly from the sweat chloride benefits, which, in the study, were about 7 -- about 6 to 7 millimolar. And so the combination of that, that the bar was so high when the study started and we're adding a single second-gen -- first-gen corrector which only improves sweat chloride a bit but only by about 7 millimole, just made it difficult to detect that incremental difference in acute FEV1.

Our next question comes from Geoff Meacham of Barclays.

When I look at kind of the development of your CF program, you've obviously developed very specific medicines for very specific mutations. And I guess, when you look at the triple data, it's entirely possible that, regardless if the patient is homozygous or heterozygous for Delta F, they could be treated with the triple just based on what we see today. So A, do you agree with that? Do you agree with sort of a one-size-fits-all Delta F508 allele single medicine? And then, B, what would be the -- what would you wait on the data in homozygous Delta F, a little bit more data from different triples than what you have already in place, to get that going? Or would you just go ahead and go with the strategy of, obviously, going out for het/mins first and then homozygous comes after? And I have one follow-up.

Geoff, this is Jeff Leiden. Great question and really important question because I think you're pointing out something that we've talked about a lot internally, which is that our sort of whole concept and strategy here around how we're approaching CF has changed over the last, say, 12 months or so, particularly as we've gotten more triple data. We originally thought of this disease as 3 or 4 buckets of patients, each of whom might require different medicines. So KALYDECO monotherapy for 1, maybe doublet therapy for another, triple for another. That's changed dramatically with the results that we're seeing from our triples. And we do believe, just as you said, that whether you have 1 508 allele and you're het/min, let's say, or whether you have 2 508 alleles, you're a homozygous patient, you will be best treated with a triple regimen, and that's certainly what the Phase II data suggests. So if you take all that together, what it really means is that up to 90% of all patients, we believe, will be on the single triple.

There will be a small set of patients, likely, who remain on KALYDECO monotherapy because those are the patients who might, for instance, have a gating mutation on 1 allele, but not 508 or not a second-gen corrector-responsive mutation on the other allele. And so we believe those patients may be fully treated with KALYDECO, but we're talking a very small number, a few hundreds of patients, in that bucket. And then you're left with the final 10% of patients, and those are the patients who don't make any protein, largely because they have stop codons. And we believe those patients, obviously, will not benefit from any of the CFTR corrector strategies. They're going to need a genetic therapy, whether that's gene therapy or gene editing. We're obviously working on those, but those are farther away. So I think the way to think about it is just as you said, we're rapidly moving towards a world in which we believe that up to 90% of patients will be on a single triple regimen and getting maximum effect, hopefully, post the carrier effects. 10% are going to be waiting for genetic therapies and there will be a few, a small number, that will remain on KALYDECO monotherapy. That's how we see the world a few years from now.

Okay. That's helpful. And then, when you -- you mentioned, Jeff, the nonsense mutation patients, and that still is a fairly large segment of the population and unaddressed. But are you comfortable with the gene editing technologies today and does that -- is that less of a priority for you than the full 90% kind of segment? And do you feel like the gene delivery systems today are good enough for you to address that population? Or do they need further optimization?

Yes. So first of all, for your first part of the question, we have been very consistent and insistent that we're committed to treating all patients with CF. So that's our priority, and one population is not higher than the other. I think what's different, just as you point out, is that we're actually quite close, I think, to that 90% of patients as we move our triples through. You're correct that the gene editing or gene therapy approaches are further way, and interestingly, not because of the gene editing or gene therapy technologies, but simply because of the difficulty of delivery, and particularly delivery to a lung that's occluded by mucus, that's full of nucleases. We already have to get to a live surface area. And then, of course, remember that most of those, if they're going to be inhaled deliveries, won't reach the other organs. And because CF is a systemic disease, we're going to have to make sure that we're treating all the organs.

So those patients who have stop codons, that is a difficult patient population to treat. We're absolutely committed to doing it. We're working hard on gene editing and gene therapy or Moderna types of approaches, but I do -- I don't want to give you the impression that that's an easy problem, that it's coming quickly because that's a longer-term problem. You also asked in the first part of your question whether we were going to wait to do homozygous after we had done het/min patients with the triples, and the answer to that is no. Our intention is to move forward with all of the eligible patients. Whether it's exactly on the same day timeline or week or month timeline, we plan to move forward simultaneously because we think that all of those patients are going to benefit from the triple.

Your next question comes from Alethia Young of Crédit Suisse.

Just one on Europe actually. Kind of as we think about the process that you're going through with ORKAMBI and potential approval for teza/iva in second half 2018, I guess, I just kind of want to think about the dynamic of like both of these events kind of being somewhat relatively close to each other. Like how does that kind of drive the discussion? Or is there any kind of correlation at all?

Yes, thanks for question, Alethia. It's Stuart here, I'll take that one. Our private negotiations on ORKAMBI are ongoing in multiple countries. As you know, we've had significant success through 2017 in places like Germany, Ireland, Italy and Denmark. And I'm pleased to say, just today, we reached an agreement in principle with The Netherlands, which is another important country for us. There's patients there who have been waiting a long time as well. So we're pleased with the progress we're making with ORKAMBI. It's never as fast as we or the CF community would like, obviously. We're also pleased with the prices that we are achieving in Europe. We think they reflect the value of this transformative medicine. And importantly, they're very consistent across the different countries in Europe, which is an important principle for us because we know that the value of ORKAMBI is the same for a patient, no matter what country they live in. So we're continuing to progress those and we want to conclude those pricing and reimbursement agreements as quickly as we possibly can because we know that CF is a [relentlessly] progressive disease and these patients are getting sicker and sicker, whilst they could be benefiting from ORKAMBI.

Obviously, the filing of teza/iva and, indeed, the high level of awareness there is about the triple combination data has led a number of countries to inquire about portfolio-type arrangements like the one that we have struck in Ireland. And if we could strike those kind of arrangements, I would certainly do that with any country, obviously, at a reasonable price. Because I think that's a win for us as a company, and I certainly believe it's a win for patients and physicians because it allows access to the best Vertex medicines as soon as they're possibly available. So we'll continue to pursue pricing and reimbursement agreements with ORKAMBI with as much speed as we possibly can, but are certainly open to any government who wants to enter into a more portfolio-type agreement as well, but obviously, we don't want that to slow down access to ORKAMBI because those patients are waiting today.

Our next question comes from Brian Abrahams of RBC Capital Markets.

Two questions. First off, I recognize that you can't comment on ongoing FDA interactions around the triple combo path forward. But just wondering if you've done the same kinds of preclinical analyses for the triple combos, for instance, in FRT cells that had met the threshold for in vitro supportive approval for KALYDECO in the residual function setting. And then, secondly, you made some comments on the potential to move 1 once-daily triple-combo cocktail forward. I'm just wondering if we should take that to mean that you wouldn't likely move forward with both 659 and 445, or just that if you do take 2 cocktails forward, you'd probably mitigate the risk by having one be KALYDECO-based and one be 561-based.

Two good questions. Thanks, Brian. The second question is our ABS, you're right on the money there. We would take, for instance -- or could take 659 and 445 forward, but almost certainly 1 of those would be once a day, one of them would be twice-a-day KALYDECO, for exactly the reason you said, just to mitigate the risk. So that's an easy answer. And then your other question about FRT cells, yes, we have and are continuing to accumulate data on the triple in FRT cells with all -- many of the different RF and other mutations. And certainly, those are part of our discussions with the regulators.

Our next question comes from Cory Kasimov of JPMorgan.

This is Shawn on for Cory. I have 2, if I might. First, kind of just a follow-on and clarification. I just want to make sure I'm understanding correctly. So regarding the recognition of revenues from France, now, in the event that the reimbursement does not come online in 2018, and then -- sorry, does not come online in 2017, and we get into 2018 where we're subject to the new rules, would that backlog of revenue then be recorded by necessity in the first quarter of next year? And then, secondly, kind of in regards to the planned triple program, was the intention here to include a separate trial for gating mutation patients? And if so, kind of how do the results from the recent teza/iva trials change or not change your thinking around that?

So thanks for the questions, Shawn. Obviously, I'll take the first one, and Jeff Leiden can take over. The -- I'll try again on the revenue recognition. So if we do not get reimbursed in 2017, we still have that bolus of cash we received for providing medicines in France sitting on our balance sheet as we go into 2018. At the point -- starting January 1, we will record revenues going forward in 2018, and that will be at an estimated price that we anticipate that we would receive in France, and that will be recorded on the revenue line. That bolus of revenues from the prior period still sits on the balance sheet. At the point that we actually get reimbursement approval in France, that bolus of revenues will go through retained earnings. It will not go through revenues. And so we continue to record revenues at an estimated price on the top line in 2018, with or without approved reimbursement, but that bolus will no longer go through our revenue line. It's a function of the change in the GAAP for revenue recognition.

And Shawn, this is Jeff. Thanks for the question on gating because it's an important one, one that we've thought a lot about. As I said, what we learned from this trial was that the addition of a single first-generation corrector to KALYDECO in these very well-treated patients isn't enough to drive acute FEV1 benefit. However, we definitely want to test the hypothesis that with a first-generation and a second-generation corrector, which obviously drives a lot more activity from that single F508 allele, that we could potentially drive increasing acute FEV1, even in these well-treated patients. So our intention is to include those patients in our Phase III program. Just to be really clear, you said, would it be a separate trial? That I can't comment on because, obviously, the trial design we wouldn't talk about. But the fundamental question, are we going to include gating patients in our Phase III triple programs? Absolutely, because we really want to answer that question of can we produce even more acute FEV1 improvement?

But equally important, I just want to remind you that this study really only looked at the acute improvement, and there's a whole different, very important aspect to these CFTR-corrector regimens, which is chronic improvement, seeing improvement in things like the slope of decline of the pulmonary function curve, the pulmonary exacerbations, the hospitalizations, the use of antibiotics, mortality, transplantation, et cetera. And one of the things we've learned from our trials, as you know, is that there's somewhat of a disconnect between the acute improvement that you see and the chronic improvement that you see. So as you know, in KALYDECO, we can see -- we see double-digit acute improvements in FEV1. With ORKAMBI, we see about a 2.8% to 3% improvement acutely. But when we look chronically, what we see is actually quite similar between the 2 medicines in terms of decreases in slope of decline and pulmonary exacerbations, et cetera.

And that is really, really important for these patients over the long run. So what we want to do now with the triple -- well, we know we can drive a lot more activity off the single or double 508 alleles -- is to test the acute hypothesis, can we drive more activity even acutely by having a lot more 508 correction? And chronically, we want to ask, will these patients do better as we get them closer and closer to essentially carrier levels? We believe that, but we have to prove it in clinical trials. It's a really important thing because this may be some of the last times we can do these kinds of trials in these patients and compare them to placebo or simpler treatment. So we will want to improve those in the Phase III program.

Our next question comes from Phil Nadeau of Cowen & Company.

First, just a housekeeping one for Ian. Ian, could you break down the revenue for KALYDECO and ORKAMBI by geographic region?

Actually, I'll pass it on to Stuart. I think he has the numbers at his fingertips.

I do, Phil. So for KALYDECO, we had $133 million in the quarter in the U.S., $80 million ex U.S. And for ORKAMBI, it was $293 million in the U.S., and $43 million ex U.S. in Q3.

That's very helpful. And then, a second question on the triple development. In the past, you had suggested that the Concert molecule maybe wouldn't be ready to move forward into Phase III in the first half of next year, but that was going to be something that you looked at subsequently. So I'm kind of curious, what has changed over the last few months that gives you more confidence in moving Concert's molecule forward, right in the first half of the year in the initial Phase III program?

Yes, Phil. Nothing has changed, except we've been working really hard on that program to accumulate the data we would need to include it in these Phase II programs. And basically, what we're announcing today is that we feel we have an update and we feel comfortable in including it in the programs. We've done so. And so we do think it will allow us in the first half of next year to make that kind of decision about whether to include it or not, and with which regimens.

Phil, I would just add that our product communications around the inclusion in the Phase II and, therefore, in the Phase III potential was a lot to do with the speed of closure. We were going through Hart-Scott-Rodino review with that transaction, and we were unsure of when that would close and when we'd get a positive opinion. We did get that and it's allowed us to move quickly internally and, therefore, incorporate VX-561 into the Phase II programs to gather data to make the decision in early '18.

Great. And then, one last question on the Phase III programs. You've differentiated between the acute effects of the drugs and the chronic effects, and it seems like maybe you're going to try to make the case to the FDA that the acute effects are beneficial enough to patients to support an approval on their own. I'm kind of curious, though, on the safety side, how much safe -- what's the minimum amount of safety data that could potentially be supplied for a drug that's going to used chronically? Do you need a significant number of patients that are out to 12 months versus some way, some precedent for including shorter course of treatments in the safety package for chronic therapy?

So Phil, you've identified all of the key questions, but that's exactly what we're discussing with the FDA, all of those that you talked about. As I said, we think we've learned a lot from our trials. It will allow us to streamline these trials, particularly for those patients who don't have anything today. But until we have those discussions finalized and we've seen all the data from the Phase IIs, it's just too early for me to give you any specific answers because we just don't have agreement yet from the regulators on each of those questions.

Our next question comes from Robyn Karnauskas of Citi.

Two questions. First one may be really stupid. So besides a risk-mitigating strategy taking up to 2 products forward, is there anything else, any other reasons why you'd take 2 products forward? And if they both look good in their final analysis, would you launch both products? How are you thinking about that? And then, the second question is more of a big-picture question on your pipeline. I know you're all -- all focus is on getting these triple combos and moving them forward throughout '18, but when do you think you'll start to sort of talk to The Street about your pipeline, all the stuff you've been working on for a while, and your big-picture strategy for diversifying the company? I know you're -- Ian, you'd mentioned about doing deals and things like that, but when do you think that will become a big focus for you to communicate that to us?

Yes, Robyn, it's Jeff. So you're absolutely right, our #1 focus, for obvious reasons and for patient reasons, is to move the triples forward as quickly as we can. That doesn't mean that we're not focusing on the non-CF pipeline. We are, we're just not talking about it as much. Your question was, why take 2 forward? And it really is just what you said; it's a portfolio risk-mitigation strategy. As you know, Phase III is about both efficacy, multiple endpoints, as well as safety. And we would just -- we have multiple regimens. We would hate to take one regimen forward, get any kind of surprise in Phase III and be set back by 1 year or 2 for patients because these patients are waiting for the therapy. So by taking 2 forward, we just feel we're mitigating the risk, significantly, actually.

And then, with respect -- and yes, would we launch 2? Everything is going to depend on the data, but if you ask me today, our intention would be to pick the best regimen and launch 1 regimen, assuming there were no major differences in different patient populations. Because at the end of the day, it's just less confusing for patients, I think, if there's 1 regimen that treats 90% of patients. With respect to the pipeline, as I said, the programs are moving forward very nicely. We're very pleased. David Altshuler is really driving those programs forward at multiple sites. And we expect to bring some of them into the clinic next year. And I think, certainly, when we begin to bring them into the clinic, we'll begin to talk more about them and show you some of that data, so you should expect from us to hear more about several of these in 2018.

Our next question comes from Ying Huang of Bank of America Merrill Lynch.

This is (inaudible) on for Ying. So a couple. First is that, what is the compliance and persistence rate for ORKAMBI between 6 and 11? Do you think that you have any more room to grow? And secondly, for the 3Q numbers, should we see any inventory or seasonality impact from ORKAMBI? And lastly, the question is, given that the Phase II data of the triple is pretty strong, do you guys have a minimum number of patients required to enroll into the Phase III trial?

So I just -- you didn't come through very clear, so if we could just make sure we've got the question, so the first question was does ORKAMBI has room to grow? Maybe Stuart.

It's compliance and persistence for patients aged from 6 to 11. I'm just wondering, for that particular subgroup of patients, do you still have more room to grow? Or do you think they could potentially just wait until newer therapies comes to the market?

So yes, okay, Stuart will take that.

So yes, in the 6 to 11 age group here in the U.S., we've seen very robust uptake in that patient group. And we've seen, as we anticipated, persistence levels which are quite high and compliance levels which are quite high, both of them actually higher than we've seen in the older patients. And that was as we anticipated. It was based on the clinical profile that we saw in the studies that we did in this patient population where the respiratory adverse events and dropout rates were very low. And we know in these younger patients, because they're largely under the supervision of their parents, their compliance rates tend to be higher. In terms of further growth, yes, I anticipate we are going to see further growth in that 6 to 11 population, both here in the U.S., because I don't think we've yet reached peak penetration in that patient group, and then, obviously, we're still awaiting the approval of that indication outside of the U.S. And that will also give us an opportunity to further grow ORKAMBI in that population.

I think your second question was around seasonality. Did we see any seasonality in Q3, or inventory impacts for Q3? We really didn't see anything out of the ordinary in terms of inventory impacts in Q3. One thing we didn't experience this year that we had experienced last year with ORKAMBI was a compliance dip during the summer months. I think we learned the lessons of why that happened. The team put a number of programs in place, and we didn't see any compliance dip with ORKAMBI in Q3, and that was one of the reasons why we saw such strong demand during the quarter.

I think your second question was are there enough patients for the clinical trials of triple. Is that your question or...

Yes. I'm speaking about (inaudible), like how many do you need?

Yes. So in the homozygous and het/min patients, where we've had a lot of experience now enrolling our Phase II trials with triples, we're very, very confident that not only are there enough patients, but the demand to get into these clinical trials and try these regimens, given the strength of the results that we're seeing, is very, very high. I think there's enough patients in the other population as well. The only thing I would note to you is in the KALYDECO-treated gating patients, for example, it has been a little more difficult to recruit those trials, as you know, because the patients are so well treated. And so to convince them to come back in takes a little more convincing. We'll see how that goes when we get to those patients, but that's certainly not going to slow down any sort of regulatory filings.

Our next question comes from Adam Walsh of Stifel.

This is for Jeff C. On the ENaC inhibitor, I think your earlier assays had suggested that the combination therapy with ORKAMBI might work. I'm just wondering if there's any read-through to the primary ciliary dyskinesia study that's ongoing, how we should think about the assay performance there and whether or not we would think of any readthrough from the ORKAMBI combo study?

Yes, maybe I'll take that, this is Jeff Leiden. So first, the question just in CF, just to remind you that the ENaC inhibitor, obviously, is a complementary mechanism. It doesn't work through chloride transport. It works through sodium transport. And therefore, we can't do the kinds of HB assays on chloride transport that we typically do with any of our CFTR modulators. But the best that we can do and did do in the HB assays was to do the height of the hydration layer and the frequency of ciliary beating. And in those assays, with ORKAMBI, it looked like adding ENaC enhanced the activity of ORKAMBI to increase the fluid layer and the ciliary beat frequency, which is why we went ahead to do the Phase III trial. Which, I think, taught us that there isn't a clinical benefit that we're seeing in that population after addition of ORKAMBI. How does that read through to primary ciliary dyskinesia? I'm not sure you can make a direct comparison. The question is, again, does increasing the fluid layer improve pulmonary function in primary ciliary dyskinesia, and the reason we're doing the trial is there's theoretical reasons to think so, but the only way to really answer the question is in patients, and that trial is ongoing. And when we finish, I think we'll have a pretty definitive answer about the role of ENaC inhibitors in that disease as well.

When will that finish?

We haven't -- it depends a little bit on how it enrolls. It's enrolling now well, but we'd like to sort of finish and see the enrollment complete and understand how that plays out in the timeline. We'll let you know when we get there.

And our final question comes from Tony Butler of Guggenheim Securities.

Briefly, Jeff, you alluded to the notion of once the triple data come in and -- a single medicine would be selected to move forward and file, and that makes total sense for all populations. But if I think back to the July time period when you did reveal data on -- at least the early data on the triples, there did seem to be some distinguishing characteristics among some medicines, some triples, which looked better in 508 homozygous patients, yet some others that look better in het/mins. And that may not necessarily bear out longer term for longer-term studies, so I'm just curious if, in fact, that is not only correct, but more importantly, is there a hierarchical decision based upon populations that may make more sense?

Yes. Great question. Let me take it in 2 parts. First of all, just to give you our view of the early Phase II data, we don't believe that you can actually distinguish yet between the different triple combinations with respect to FEV1 or population. They were all in the close to double-digit range, and given the size of those trials, distinguishing between 9% and 10% in a trial like that is probably not a good idea. We may or may not have enough data once we get through all the Phase II trials to make those decisions and distinguish that way, and we'll see.

And your second part of the question, which is absolutely right, is, of course, the final decision will depend upon the data. And the Phase III data will be, hopefully, quite robust and give us the ability to really distinguish both on the efficacy and the safety side. Having said all that, my hope, and it's a hope, is that what we'll see is a pretty clear best regimen, if you will, which would allow us to launch 1 regimen. As you say, in the chance or the possibility that we saw significant and reproducible differences between the different regimens in different populations, of course, we would always launch the best population -- the best regimen into each population. So it will depend on what we see, but I'm just trying to give you a little bit of crystal ball-gazing on what we'd like to do.

Okay. Thanks, everybody, for joining us this evening. The Investor Relations team is going to be in the office if you have additional questions. This concludes the call, and you may now disconnect.