Vertex Pharmaceuticals Incorporated (VRTX) Q1 2014 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and thank you for your patience. You've joined the Vertex Pharmaceuticals Incorporated First Quarter 2014 Financial Results Conference Call. [Operator Instructions] As a reminder, this conference may be recorded. I would now like to turn the call over to your host, the Vice President of Investor Relations, Mr. Michael Partridge. Sir, you may begin.

Thank you, operator, and good evening, everyone. Joining me on tonight's call are Dr. Jeff Leiden, Chairman and CEO; Dr. Jeff Chodakewitz, Chief Medical Officer; Stuart Arbuckle, Chief Commercial Officer; and Ian Smith, Chief Financial Officer. Our agenda tonight is as follows, Jeff Leiden will begin by reviewing Vertex' strategic business priorities for 2014. Then, Jeff Chodakewitz will review our clinical progress on cystic fibrosis, including the results of the study of VX-661 in combination with KALYDECO in G551D, F508del patients. Dr. Chodakewitz joined Vertex at the start of 2014 and we are happy to have him join us on this and future calls. Next, Stuart will discuss KALYDECO product revenues and provide some commentary on the outlook for KALYDECO growth in 2014. To close, Ian will review the financial results and discuss our updated financial guidance. Joining us for Q&A are Dr. Bob Kauffman and Dr. Peter Mueller. We plan for the call to run for a total of 1 hour. [Operator Instructions] This conference call will include forward-looking statements, which are subject to the risks and uncertainties, including those discussed in detail in our reports filed with the Securities and Exchange Commission, including our 10-K and 10-Q. These statements including, without limitation, those regarding the performance of KALYDECO, our development plans and expectations and our guidance, are based on management's current assumptions and are subject to risks and uncertainties that could cause actual outcomes and events to differ materially. Information regarding our use of GAAP and non-GAAP financial measures and a reconciliation of GAAP to non-GAAP is available in our first quarter 2014 financial results press release. This press release is on our website, and I would also refer you to Slide 4 of tonight's webcast. Thank you. I will now turn the call over to Jeff Leiden.

Thanks, Michael, and good evening, everyone. Vertex is, in many ways, a company in transition. We have made cystic fibrosis a clear focus of our business and we've made a decision to no longer invest in hepatitis C. At the same time, we're continuing to invest in early-stage research to identify compounds in CF and other areas that could become transformative medicines for patients with serious diseases, as well as opportunities for Vertex growth. This transition in Vertex business represents our continued execution against 3 strategic priorities for 2014. I talked about these priorities earlier this year, and I'd like to review them again now. First, focus on driving forward our cystic fibrosis development programs to enable the launch of multiple transformative CF medicines globally. Our clinical and commercial teams have made significant progress in CF already this year. For example, we have expanded the U.S. KALYDECO label to include additional patients. We have now completed dosing in the Phase III studies of VX-809. We have recently received orphan status for VX-661, and we are today reporting positive clinical results for the study of VX-661 in patients heterozygous for G551D and F508del. Second, continue to invest in research. Internal research is a critical growth engine for Vertex, and we expect that it will produce more transformative medicines to provide sustainable growth in the future. And third, maintain financial strength so that we can continue to invest in the discovery and development of our new medicines. We ended the first quarter with more than $1.3 billion in cash and with KALYDECO revenues of approximately $100 million that we expect to grow through the year and in future years. This financial position supports our investments for future growth. Our goal in CF is to treat as many people as possible and to enhance the benefit for those that we treat. Jeff Chodakewitz will give you more detail on our recent CF clinical progress. But before I turn the call over to him, I would like to acknowledge that today marks the start of CF Awareness Month and we recognize all of the patients, families, doctors, nurses and caregivers who are involved in the treatment of CF. Thank you for your continued support as we seek to improve the lives of more people with this devastating disease. Our continued progress in CF clinical development is a strong sign that our commitment to people with CF is unwavering. With that, I'll turn it over to Jeff.

Thanks, Jeff, and good evening. It's a pleasure to speak with you all today. I'll provide an overview of some important progress we've made with our 3 CF development programs, specifically, ivacaftor monotherapy, lumacaftor and ivacaftor combination therapy and VX-661 and ivacaftor combination therapy. Let me begin with ivacaftor monotherapy. Our clinical and regulatory efforts are focused on demonstrating clinical activity in additional patients that may benefit from this medicine. In February, following approval from the U.S. FDA, we began treating patients with 8 additional mutations with KALYDECO. These are all mutations for which in vitro data predicted critical responsiveness to KALYDECO. We've submitted an MAA variation to the European Medicines Agency and expect to receive an opinion from the CHMP for these additional gating mutations around midyear. An additional mutation for which we are seeking ivacaftor label expansion is R117H. As you may recall, we announced results in December from a Phase III trial in this population. We did not achieve statistical significance in the primary endpoint for the overall population of this trial, but ivacaftor did produce a statistically and clinically significant improvement in FEV1 and other measures in a prespecified subpopulation of participants 18 years of age and older. We plan to submit an sNDA and MAA variation for people with CF, ages 18 and older, with at least 1 copy of the R117H mutation. The sNDA submission is planned for midyear, with the MAA variation to follow in the second half of 2014. We expect to continue our discussions with the FDA following our sNDA submission. We estimate that there are more than 700 patients, aged 18 and older, worldwide with the R117H mutation. We also believe that ivacaftor may be able to help other patients who have residual CFTR function. We are conducting a proof-of-concept study in these patients, and we are on track to report results later this quarter. In our view, a successful result in this study of approximately 20 participants would lay the groundwork for pivotal studies of ivacaftor in patients with residual CFTR function. Finally, we expect to report results from our pediatric safety study of ivacaftor use in 2- to 5-year olds with gating mutations in the third quarter, while patients in the pediatric study who have completed 24 weeks of treatment have now progressed into the rollover study. Moving to our lumacaftor-ivacaftor combination program. The 6-month dosing period is complete for the Phase III TRAFFIC and TRANSPORT studies in patients, aged 12 and older, who are homozygous for the F508del mutation. We are on track to report 24-week results from these studies around midyear. If these studies are successful, we expect to file an NDA and MAA later this year. In TRAFFIC and TRANSPORT, we are seeking to demonstrate a statistically and clinically significant improvement in absolute FEV1, the primary endpoint of the study, and also benefit in important secondary endpoints, including weight gain, pulmonary exacerbations and CFQ-R. We estimate that our studies are more than 90% powered for absolute improvement in FEV1 of 3% from baseline compared to placebo at week 24. In addition to lumacaftor, we are developing another first-generation corrector, VX-661. Our principal strategy with VX-661 is to develop it as part of a future triple combination therapy with a next-generation corrector and a potentiator. Dosing has begun in our 12-week study of 661 plus ivacaftor in patients 18 and older who are homozygous for the F508del mutation. We expect to enroll approximately 40 patients who will be treated with 661 plus ivacaftor or placebo. Our goal of this study is to assess the safety and efficacy of 661 plus ivacaftor over a longer duration of treatment. If successful, this study would be enabling for future development of VX-661 from use in a triple combination or potentially to progress it as part of a dual combination. Today, Vertex announced results from a proof-of-concept study of the corrector 661 in combination with ivacaftor in CF patients who have achieved 551D mutation on 1 allele and the F508del mutation on the other allele. The basis for exploring the VX-661 plus ivacaftor combination in this population is based on in vitro observations from HBE cells, in which adding VX-661 to ivacaftor improved chloride transport. The purpose of the trial was to see if the improvement in CFTR function that we observed in the lab could translate into a signal of clinical efficacy in the G551D, F508del heterozygous population. We dosed VX-661 as 100 milligrams daily in patients who were already taking 150 milligrams of KALYDECO twice daily for an average of 1 year. We were testing this clinical hypothesis for the first time and we're not sure if we'd be able to detect the signal. The patients that we enrolled were G551D patients who had already been receiving commercial KALYDECO and, therefore, were likely to have already seen a significant clinical benefit from taking this medication. Also, the study was small. We enrolled just 18 patients. And the duration of therapy was short, just 28 days, for which VX-661 would be added on top of ivacaftor. The results do, in fact, show a clear signal of efficacy. There was a mean 4.6 percentage point absolute improvement within group in ppFEV1 through 28 days. The relative FEV1 improvement was 7.3%. These improvements were both statistic -- were both statistically significant. We also saw a small reduction in sweat chloride on treatment within group, with a p value of 0.053. FEV1 and sweat chloride returned toward baseline in the 4 weeks following completion of the 661 treatment period. These off-treatment effects were statistically significant within group as well. In this study, 4 patients already receiving KALYDECO were randomized to placebo to maintain study blinding. From a safety perspective, the regimen was generally well tolerated over 28 days and all 18 patients completed the treatment period. The most common adverse events in the treatment group were pulmonary exacerbation, headache and upper respiratory tract infection. And we expect to present more details on these results at a medical meeting in 2014. For us, there are really 3 important takeaways: first, once again, in vitro data with our CFTR modulators has been predictive of a clinical benefit in the specific patient population; second, correctors like 661 appear to be clinically active on a single F508del allele; and lastly, these results suggest to us that treatment for heterozygous patients may be further enhanced with CFTR modulators that are tailored toward each allele. Based on the data announced today from this study and pending the results of the 12-week study, we could decide to develop VX-661 as part of a dual combination for certain heterozygous patient groups. Also based on these data, we plan to discuss with global regulatory authorities the potential approval pathway for VX-661 in combination with KALYDECO for people with CF who have the F508del mutation and another mutation known to respond to KALYDECO alone. In summary, I'm pleased with our progress with our CF development programs and I look forward to updating you throughout the year. With that, I'll turn it over to Stuart.

Good evening, everybody. Tonight, I'll report our first quarter product revenues and outline our expectations for KALYDECO revenue growth in 2014. KALYDECO generated $100 million in worldwide net revenues for the first quarter, including U.S. sales of approximately $56 million and international sales of approximately $44 million. This represents an increase of 61% from revenues of $62 million that we reported for the first quarter of 2013. The reason for the decline from the $109 million that we reported in the fourth quarter of 2013 was the favorable impact of stocking and other nonrecurring business adjustments in the fourth quarter that we discussed on our quarterly call in January. We continue to see a very similar level of underlying demand and strong patient adherence to treatment. We expect KALYDECO revenues to grow through 2014, beginning in the second quarter, based on the following anticipated key factors: first, label expansion. Since receiving approval from the FDA in late February for 8 additional mutations, we are seeing strong uptake similar to what we saw in patients with the G551D mutation. We expect uptake to continue as we seek to reach the vast majority of the approximately 115 newly eligible patients in the U.S. by year-end. We have filed an MAA variation with the EMA for a similar label expansion and currently anticipate approval in the second half of 2014. Following EMA approval, we will seek reimbursement for these new patients, both in countries where KALYDECO is already formally reimbursed, such as France, and in new countries like Italy where, because there are so few G551D patients, the medicine has not yet been formally reviewed. There are approximately 250 eligible people with these additional mutations living in the EU. Our second key factor of KALYDECO growth is geographic expansion from securing public reimbursements in Canada and Australia. The reimbursement process is unfortunately taking longer than we and, more importantly, people with CF wants it. Reimbursement discussions in both countries are active and ongoing and our goal remains to get this transformational medicine to all eligible children and adults as quickly as possible. There are approximately 300 people with the G551D mutation in Canada and Australia, and we know they and their families are anxiously awaiting a positive outcome from these discussions. To summarize, we expect label and geographic expansion to enable us to grow our KALYDECO revenues beginning in the second quarter and continuing throughout the year. This growth is reflected in our KALYDECO net revenue guidance, which we are maintaining. I look forward to updating you on our continued progress throughout the year, and I'll now turn it over to Ian.

Thanks, Stuart, and good evening to everyone. Tonight, I would like to discuss our current financial position, our first quarter 2014 results and our updated 2014 financial guidance. During the first quarter, we've further prioritized our development activities toward cystic fibrosis. Our financial results and guidance reflect this prioritization. As we entered the second quarter, we are a business primarily funded by CF revenues and a strong cash position. We are investing in the discovery and development of medicines for CF and the creation of new medicines in other areas. Now to our cash position. We finished the quarter with over $1.3 billion in cash. This balance sheet strength, combined with the expected growth in our KALYDECO revenues, supports our business investment and still allows us to exit 2014 with a strong balance sheet. We continue to manage this cash position as the priority, as it enables the transition of our business to growth and profitability driven by our CF medicines. An example of our prioritization to protect our financial strength is shown by our decision not to proceed with VX-135 in hepatitis C, amending our arrangement with Alios and our decision to out-license the compound. This is all reflected in our updated 2014 financial guidance that I'll discuss in a moment. Now to the first quarter 2014 results. We generated $108 million in total non-GAAP revenues in the first quarter. $100 million of this was KALYDECO. Based on our decision to end further investments in hepatitis C I've announced this evening, the HCV-related revenues, totaling approximately $10 million, are excluded from non-GAAP revenues. Now to our operating expense. Our first quarter total non-GAAP operating expense was $234 million, down $41 million or 15% from Q1 2013. The first quarter non-GAAP operating expense includes $182 million of R&D expenses compared to $195 million in the first quarter of 2013, and $52 million of SG&A expenses compared to $80 million in the first quarter of 2013. These reductions of 7% and 35%, respectively, reflect our commitment to prioritization towards the CF medicines and the protection of our financial position. Our GAAP net loss for the first quarter was $232 million or $1 per diluted share. Our non-GAAP net loss was $151 million or $0.65 per diluted share. Turning now to our financial guidance and, firstly, our revenue guidance for the full year 2014. We are maintaining our KALYDECO guidance of $470 million to $500 million that we provided earlier this year. Achieving this guidance depends on 3 key drivers of growth: the continued uptake for the 8 additional mutations in the U.S. and achieving reimbursement for the G551D patients in Australia and Canada, and the approval for other additional gating-type mutations in the EU. With the exclusion of our HCV revenues, we are revising our total non-GAAP revenue guidance to be in the range of $520 million to $550 million. This range includes some revenues for the potential R&D collaborations. Based on our move away from HCV and further focusing our investment towards CF, we are reducing our operating expense guidance. We now anticipate 2014 non-GAAP operating expenses to be $890 million to $930 million. Additionally, I would just note that Q1 non-GAAP OpEx for R&D and SG&A provide a reasonable basis for modeling our expenses during the rest of the year. Total spend may decline slightly towards the end of the year based on clinical development activities. In summary, we are committed to completing 2014 with a strong balance sheet and managing our operating expense at a decreased level compared to prior years, while we grow our CF revenues. With that, I'll ask the operator to please open the lines to questions.

[Operator Instructions] Our first question comes from Geoff Meacham of JPMorgan. Geoffrey C. Meacham - JP Morgan Chase & Co, Research Division: Got a couple on the 661 KALYDECO combo study. I'm assuming that the patients on the placebo arm on just KALYDECO monotherapy were stable. Are you -- that data isn't in the slide, but I'm assuming that, that is largely unchanged over the evaluation period.

Geoff, so it's Jeff Chodakewitz. I think it's important to step back and say why those patients were in there. It's really -- it was only 4 patients, as we mentioned, and they were really there for blinding. So we really focused on the within-group analysis that you saw. It wasn't really intended as a controlled trial with the placebo group. They were just there sort of maintaining the integrity of the blind. Geoffrey C. Meacham - JP Morgan Chase & Co, Research Division: I see. Okay, that makes sense. And then does this data change your view ultimately about the need for a second corrector? And I joined the call a little bit late, so I wasn't sure if you guys gave an update on the next-gen corrector behind 661?

Geoff, it's Jeff Leiden. I think our view of 661, first of all, has been enhanced by this data. We're learning about as we progress 661. And I think from this trial, we learned a couple of things: one, that it's possible to correct a single Delta 508 allele, at least, to these patients with a G551D allele; and secondly, you could do that on top of patients who have been stabling on KALYDECO for quite a bit of time. Most of the patients have been on for a year or so, and that was an important finding. I think we're still thinking about 661 as part of a potential triple regimen to get enhanced benefit for everybody that's homozygous and heterozygous. But I think what this data tells us is that 661 may also have a role in certain of these heterozygous patients, and that's worth exploring. And if you'll remember, we talked about this being an interesting trial from that perspective a while ago.

Our next question comes from Geoff Porges of Bernstein. Geoffrey C. Porges - Sanford C. Bernstein & Co., LLC., Research Division: I just like to follow up on the 661 combination. So first, Jeff Chodakewitz, you mentioned the control group. But could you tell us whether there was a difference in the cough exacerbations and other adverse events between the control group and the treatment group? And then secondly, the G551 patients, are they typical G551D patients? They seem to have a pretty low FEV1 as a starting point in this trial. And then just to push a little bit more, why wouldn't you be contemplating now taking 661 with a standard dose of KALYDECO into a larger study in the homozygous Delta F508s?

Great. Thanks, Geoff. So in terms of the placebo patients, again, it was really there for blinding. That's really what our intent was. And in terms of the severity, I think that it was a little bit lower as you said, than perhaps what we've seen in some other studies. But I think it is well within the range and the groups were -- it was pretty balanced, and it was really just a -- we think it was really pretty representative of what we've seen in our 551D studies. I'm sorry, your last question was around... Geoffrey C. Porges - Sanford C. Bernstein & Co., LLC., Research Division: Yes. Why wouldn't you...

Make sense of 661, Jeff, and whether we'd consider going into a larger study for the Delta F508 homozygous?

Yes. So I think for us, really, we think we're just getting this data. We really have to understand it a little bit better, it's early days. We also have the ongoing 12-week study for 661 and ivacaftor in the F508del homozygous population. And what we're going to do is really step back when we get those results and really say what's the right path forward for 661.

Our next question comes from Robyn Karnauskas of Deutsche Bank.

This is Alethia for Robyn. And just another 661 question, but just curious if this information that you have about 661, how does this change your perspective on acquiring or looking at complementary assets for your CF program?

Jeff is going to take that question in terms of our portfolio approach.

Yes. I think we've said all along that we want to continue our leadership position in CF, and that's going to come in 2 forms. One is to continue to develop our expanding portfolio of medicines, so not only 809 and 661 but next-gen correctors, which we think are going to be very important in both expanding the region, expanding the benefit. But the other thing is we have a very active, as you can imagine, ongoing surveillance program of all of the potential, other assets out there in CF. And we'll continue to watch that and make sure that we're either partnering, acquiring or doing whatever we can to get the best regimens we can in combination with our drugs.

Our next question comes from Mark Schoenebaum of ISI Group.

I just wanted to -- on the placebo issue, would you guys be willing to characterize behavior? I know it's only 4, but would you guys be willing to characterize the behavior of the 4 placebo patients? Maybe give us some reassurance that they didn't behave as well as the treated patients or something like that? And then my second question was -- in the data where you removed the drug, were both drugs removed, or was that just 661?

Great. Thanks, Mark. So sure, let me try to give you a little more color. Again, it's 4 patients. I'd say that what we observed in those patients was really that there were modest shifting around. It was variable, as you'd expect, in this disease with 4 patients getting placebo. And it was very typical of what we've seen in general with smaller cohorts of patients receiving placebo. So hopefully, that gives you a little better flavor of that.

So modest would not be 4.6% on an absolute basis, I would imagine?

Correct, it's modest changes. So I think that the -- the other piece, in terms of the study design, was really that we started and then stopped the VX-661. KALYDECO, which our patients were receiving in a marketed setting, was continued throughout the study. And in the follow-up period, they just stayed on their KALYDECO.

Mark, it's Jeff Leiden. We've obviously had a number of questions on this and on the placebo group. And maybe let me take a step back, if you will, as I often do on this Phase II data, and tell you how we look at those data, which is we always look at the entirety of the results, and we try to look at it from every direction to see if we're seeing a consistency of the CF, we're seeing statistical significance. I guess when I look at this data, the thing that is very convincing to me is that when you start by looking at what we saw in the lab themselves, which was this additive effect of 661 on KALYDECO in exactly these cells, G551D, Delta 508, that again has translated into what we're seeing in the clinic. And obviously, now with 5, 6, 7 trials, relate that with the in vitro results, translates into in vivo results. Then we saw the on-effect when we added 661 to patients who were stable on KALYDECO and, honestly, with only 14 patients, surprisingly, we saw a statistically significant on-effect. And then, of course, the off-effect is important. So we removed the 661, left the patients on KALYDECO, and we saw a statistically significant decrease. And that pattern, I think in a Phase II trial, is always very important. And then, finally, correlated with sweat chloride, which always gives us confidence that what we're seeing is an on-mechanism effect. So when I look -- when I walk around the trial and look at it from all perspectives, I see a consistency of statistical significance and this nice correlation from in vitro to in vivo. I think that's what's really gives us the high level of confidence.

Our next question comes from Michael Yee of RBC Capital.

A quick question. First, I will ask a question not related to placebo, which is in your ongoing TRAFFIC and TRANSPORT study, I know they're still thinking about secondary endpoints and Wall Street's looking at how to evaluate that. Have you done any work around correlating waking and exacerbations and how that correlates to FEV1 effects? Obviously, we know what the results were in the KALYDECO monotherapy studies and the effects there and what that translated to for waking and exacerbations. But given potentially less effects in the Del 508 homozygous population, how do you predict what would happen there, or we're flying a bit blind? Then I have a follow-up.

So it's Jeff Chodakewitz, let me take a start at your question. I think -- I don't think we looked recently, at least, in a formal way, correlations between these different measures. But I do think you raised a couple of important points. First of all that we have to really understand the totality of the disease and, really, we're assessing the benefit for patients on all of those endpoints. It's really not just about any one of them. But as you imply, it's really about understanding all of them and understanding the pattern and seeing what benefit the drug brings. We do have a set of secondary endpoints that we have preplanned and prespecified, and that's going to include a number of important outcomes like BMI, as you're talking about, in terms of pulmonary exacerbations, as well as patient-reported outcomes like CFQ-R. So we're really going to take a good look at all of that and take the time to really understand all the results.

Okay. And then a follow-up is -- I'm sorry, I have to sneak in one more on the placebo, but I mean these patients in the study, they knew it was a randomized study versus placebo. They didn't know the exact randomization. I'm just trying to think about how much placebo effect there really would be given the randomization of 7:2?

Right. So people did -- were informed, of course, that it was a placebo-controlled study. Patients do, then they form a consent. But as is the intent, we were very careful with the placebo so they would not be able to detect which one they were on.

Our next question comes from Terence Flynn of Goldman Sachs.

First, just was wondering on the -- again, I'm sorry to go back to the placebo group in the 661 combo trial. Can you give us any sense of the sweat chloride data in those patients? I know you said it was -- that it sounds like the data bounced around a little bit. But just wondering if you could characterize that on a basis versus what you saw in the treatment group. And then maybe just remind us, I know you guys changed the endpoint of the Phase III combo study for KALYDECO-809. Can you give us maybe give us some background there? What drove that decision?

Sure. So again, in terms of the sweat chloride and the placebo patients, it was quite -- it was variable. There were modest changes. Again, very typical of what you'd expect to see with a small number of patients receiving placebo. And that was true across all the metrics.

Receiving KALYDECO as placebo.

Sorry, yes. Well, they received placebo on top of KALYDECO. So in terms of the Phase III endpoint change, just so I -- just maybe so I'm clear, which -- can you say a little bit more about which aspect of that, that you wanted me to comment on?

Sure, just -- I think you guys had switched from a relative to an absolute improvement on SAE and was just wondering again what drove that decision.

Right. So we think, in fact, both measures are important. We -- specifically shifting to absolute was based on a request that we got from the FDA, but we are going to have relative as a key secondary measure. So that's really the basis for our change.

The next question comes from Brian Abrahams from Wells Fargo.

The first question, on biology, do you believe that the increase in group FEV1, the 661-KALYDECO study, is due to enhancement of the G551D-CFTR function or an action of the combo on the F508-CFTR? And then just another question on the placebo. I'm sorry about that. But I realized the study wasn't powered or designed this way, but if you were to do a statistical test on the 661 group versus placebo, just given the variability you described for the placebo and the statistical significance on the in-group different for the drug arm, would it be safe to say that placebo-adjusted mean FEV1 improvements would've also been statistically significant?

It's Peter speaking. I answer the biology question. And so the belief that we have in there is some in vitro trial that justifies that this -- that adding a corrector to the mix enhances trafficking of the CFTR protein to the surface. So part of the additional increase in activity that you see here is basically enhancing the amount of CFTR protein to the surface. Now when it is on the surface, it also gets potentiated by KALYDECO. So therefore, you get basically a double effect, sort of enhancing, trafficking and...

Well, I think, Peter, the question was do we believe the effect was predominantly on the G551D allele or the 508 allele? And we believe the effect is predominantly on the 508del allele from everything we've seen in vitro.

That's correct.

And Jeff, do you want to take that second question?

Yes. So I actually think that probably we've covered this quite completely. I think that the -- we think that the important analysis that we've talked about and that Jeff Leiden referred to is really the within-group analysis and both in terms of the sweat chloride and FEV1, the on and the off.

Our next question comes from Ying Huang of Barclays.

So first of all, can you tell us -- you plan to file the R117H indications here. Is that based on the recent feedback from FDA and also EMA? And then secondly, also, I noticed that in this trial, you only tested 100 milligram dose for VX-661. Have you determined that that's going to be the dose you will take VX-661 forward, going to a potential pivotal trial? And then lastly, KALYDECO, can you talk to us about the underlying demand by, for example, bottles decent [ph] in 1Q over 4Q last year? And also, maybe patients on therapy even though we know there's inventory drawdown out here.

So it's Jeff Chodakewitz, let me take the first couple of those and I'll pass it on to Stuart. In terms of the R117H, we can't comment on our discussions with the FDA, or speak on behalf of the FDA, as we presented, we think that the data coming out of that study in -- for the patients who are greater than 18 was both statistically and clinically significant, recognizing that we did fail on the primary endpoint. And we plan to file based on that analysis and we'll have ongoing dialogue with regulatory agencies. In terms of the dose of 661, it was very early in our program. That was the dose that we studied. But I don't think we're in a position yet to say that we've decided on what VX-661 dose would be.

And on the revenue. The fourth quarter of 2013 was inflated by some one-off business adjustments. The first quarter was negatively impacted by some of those same impacts, such as increased stocking in the fourth quarter. The underlying patient demand is absolutely rock solid and was very similar in terms of total patients, in terms of compliance, persistence and all those sort of things between the 2 quarters.

We'll go to the next question from Howard Liang of Leerink.

I have a couple of questions, one science question, the other one, commercial. In the lab, can you remind us, did VX-661 have a bigger effect in the G551D, 508del heterozygous cells or F508 homozygous cells? I guess, how do you think about the difference between G551D, 508 heterozygous versus homozygous? Why a positive signal that you saw here may or may not have read through for the upcoming TRAFFIC/TRANSPORT studies?

So Howard, it's Jeff. And we don't mean to imply that this signal has implications directly that we can read out in the upcoming TRAFFIC and TRANSPORT. So if we -- if you heard that, that's certainly not our implication. We believe that there's a lot of data that supports the upcoming TRAFFIC and TRANSPORT results from the in vitro data to the in vivo data, and that's really where our confidence comes from for TRAFFIC and TRANSPORT. And what we did learn here, and I think was very, very important, is the patients who were stably treated on KALYDECO, with all of the benefits they get on the G551D allele, will also have 508 on the other allele, can see enhanced clinical benefit by treating -- by adding in a corrector. And that was obviously the big question on our minds, could you translate that out into an improvement in FEV1, once you've already fully treated the 551D allele? And I think the answers from the study is clearly yes. And that does have implications for how we think about developing 661, but not the TRAFFIC and TRANSPORT.

Can I just follow up? I guess why wouldn't you have a readthrough to TRAFFIC and TRANSPORT?

The readthrough that we have -- we had before, right, in other words, we know that corrector plus potentiator is effective in patients who have Delta 508 alleles, and we've seen that in 809 in multiple studies, I think 5 Phase II studies, and we've seen it now with 661. So I guess, if you said we have yet 1 more study that shows a corrector plus a potentiator can enhance 508 function, yes, it does. But I think we had a lot of evidence there. Where we didn't have a lot of evidence is, what happens when you add a corrector on top of KALYDECO in a 551D with 1 508 allele? We just didn't know what was going to happen, and that's really what we learned from this trial.

Okay. It makes sense. Can I just follow up on -- commercially, I think most G551D patients have delta 508 on the other allele. So when you add 661 to KALYDECO, do you expect to realize more revenue per patient? And do you think the market can bear a higher cost than KALYDECO?

So you're correct about 70% of them have 508s on the other allele, the G551D patients. In terms of pricing, much too early to talk about that. We literally just got this data in the last few days. The thing we're really excited about is the fact that we appear to be able to be bringing additional benefit to the patients who are already doing well on KALYDECO, but that's about the maximum take-home, I think, you can take right now. It's really much too early to comment on what pricing implications that may have.

Our next question comes from Matt Roden of UBS.

So if you integrate everything you know about the correctors, all the preclinical and clinical data, do you have any basis for thinking about how the benefit, assuming there is one, changes over time once you get out past 4 weeks? So we have 2 4-week observations of combination therapy. So I'd like to get your thoughts as to whether or not you think there's a basis for believing the benefit can increase, stay the same or maybe wane as you go out to 24 weeks past the full week readout? I know we have to do the experiment, but I'm just wondering if you have a basis for a view on that?

Yes. I wish I could give you a real data-based answer, I just can't. At this point, what we have is 4-week data. And that the only relevant piece of data I can give you is what we know about KALYDECO, right, which is now we have multiple years, up to 4 or 5 years with some patients, and some of which has been recently published, in which we're seeing a tremendous stability of the response starting at 2 weeks and going up multiple years. But obviously, that's KALYDECO, that's a potentiator. And unfortunately, we just don't have longer data than that. But we'll see the first evidence when we file this year our TRAFFIC and TRANSPORT. And then as Jeff said, I think the 661 plus KALYDECO 3 months' trial is going to be very, very important. First, obviously, for safety, but also because it's the first time we're going to see that 3-month data with 661. And rather than speculate, I'd rather just show you the data when we have it.

Okay. And then on the multiple sclerosis, remyelinating program, I think you previously described this as further along in your other pre-clinic programs. Just wondering if this is something we could see in the clinic this year?

So Matt, I think you asked the same question on the last call as well. But yes, we do have a program in this area. It is early and there will be a time when we'd give you more information on that.

Our next question comes from Liisa Bayko of JMP Securities.

You talked about now exploring regulatory pathways for the combination of 661 and KALYDECO in sort of that ideal situation where you have someone that responds to KALYDECO on one -- an allele that responds to KALYDECO and another allele that's 508 -- I'm sorry, another allele that's -- yes, 508. Can you maybe talk about what the patient population size is for that, just so we can have a better sense?

So there's 2 ways to think about that, right? One is in the G551D patients, where I think Stuart mentioned before somewhere between 65% and 70% or so of patients who are G551D on one allele or 508 on the other allele. So that's directly applicable to what we showed today. The other one though would be KALYDECO -- other KALYDECO responsive alleles, such as gating and residual function. And so, I think the math is a reasonable thing to do is to assume around 65% to 70% of all of those patients with the KALYDECO responsive allele on one side, we'll have 508 on the other side. The thing we don't yet know is what is really the numerator there, how many of all those residual functions are going to be responsive to KALYDECO. And that's what we're going to figure out as we begin to get the results from these NF1 trials. Did that answer your question okay?

It does.

Operator, I think we'll take 2 more questions.

Our next question comes from Brian Skorney of Robert W. Baird. Brian P. Skorney - Robert W. Baird & Co. Incorporated, Research Division: I guess the only thing I'm questioning, I sort of thought this was going to be a high hurdle because the patients were already on base one KALYDECO. I would've expected them to be relatively healthy. But just on the FEV1 measurement, I mean, actually it looks like they're sicker than patients who enrolled on the original KALYDECO Phase III study, who were on nothing. So can you just help us put this in context, how sick these patients really were even though they were on KALYDECO and how meaningful the improvement winds up being from the addition of 661?

So it's Jeff, Brian. I think that the -- first of all, I think it is a high hurdle, as you said, we were not sure given that. And the fact that's very high proportion of patients who get KALYDECO response, we were not sure that we'd able to show an improvement. I think that in terms of the baseline, when we look back, we do see some variability. Again, in terms of where patients end up after a period of time. I think this -- the patients' baseline characteristics here were a little lower, but really not out of the range that we might expect to see. And again, I think the fact that they were on stable KALYDECO for average of about 1 year, as we talked about, that's really, I think, just emphasizes the high hurdle that the small study had for showing a benefit. So -- and that's why we're enthused about the results.

Our next question comes from Ravi Mehrotra of Crédit Suisse.

This is Koon actually asking a question on behalf of Ravi. I just wanted to know if you'll comment on the range in the changes you saw in the absolute FEV1 and sweat chloride?

I think so. No, I really think -- as we said, this is really very early results. It's a small study. We're really going to have to continue evaluating it. We will be presenting the data in a scientific form in the coming year.

Thank you. Thanks for joining us this evening, and we look forward to catching up with most of you soon.

Thank you, sir. Ladies and gentlemen, that does conclude Vertex Pharmaceuticals Incorporated First Quarter 2014 Financial Results Conference Call. You may disconnect your lines at this time. Have a great day.