福泰製藥 (VRTX) 2013 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Vertex Pharmaceuticals Incorporated second-quarter 2013 financial results conference call. At this time, all participants are in a listen-only mode. Later we will conduct a question and answer session, and instructions will be given at that time.

(Operator Instructions)

As a reminder, today's call is being recorded. I would now like to turn the call over to Michael Partridge. Please go ahead, sir.

Thank you, operator, and good evening to everyone. Joining me on tonight's call are Dr. Jeff Leiden, Vertex's Chairman and CEO, Bob Kauffman, Chief Medical Officer, Stuart Arbuckle, Chief Commercial Officer, and Ian Smith, Chief Financial Officer. Jeff will begin with a few comments on the execution of our strategy to date in 2013. Then Bob will provide an update on our pipeline progress, including our mid and late stage programs in cystic fibrosis and hepatitis C.

Stuart will follow with comments on trends for our key marketed medicines, KALYDECO and INCIVEK. And to close, Ian will review the second-quarter financial results and update our 2013 guidance. Following prepared remarks, Peter Mueller, Chief Scientific Officer, will join us for a question and answer period. After the call, we will be available in our offices for follow-up.

Before we begin, I will note that information discussed on this conference call includes forward-looking statements, which are subject to risks and uncertainties. These risks and uncertainties are discussed in detail in our 10-K and 10-Q reports filed with the Securities and Exchange Commission, and also in the press release announcing our financial results tonight. These statements, including without limitation, those regarding the commercial performance of INCIVEK and KALYDECO, our development plans and expectations, and our guidance are based on management's current assumptions. These statements are subject to risks and uncertainties that could cause actual outcomes and events to differ materially.

GAAP and non-GAAP financial measures will be discussed on this call. The information regarding our use of these measures and a reconciliation of GAAP to non-GAAP is available in our second-quarter 2013 financial press release, and is described on slide 4 of tonight's webcast. I will now turn it over to Jeff.

Thanks, Michael. Good evening, everyone. Tonight, I am pleased to discuss the progress that Vertex has made in the first half of 2013. At the start of this year, we outlined three strategic imperatives for our business that would position us for long-term growth.

First, prioritizing our development investment to advance our late-stage medicines for cystic fibrosis, hepatitis C and autoimmune diseases. Second, focus on continuing to discover innovative medicines for serious diseases from our research efforts. And third, maintaining our financial strength to support the investment required to realize our significant growth opportunities in the near future.

Halfway through 2013, we have accomplished a great deal to advance our late-stage medicines, and we continue to reinvest in research to bring further ground-breaking therapies to patients, all while enhancing our financial strength. In cystic fibrosis, KALYDECO has been successfully launched for G551D patients in Europe this year. We have multiple studies under way that could potentially expand the KALYDECO label. Today, we are announcing the first clinical data for Ivacaftor monotherapy outside of G551D. Ivacaftor monotherapy has received breakthrough designation from the FDA, and we plan to submit regulatory filings in the US and the EU in the second half of 2013 to expand the use of Ivacaftor.

Our Phase 3 program for VX-809 or Lumacaftor in combination with Ivacaftor is under way to evaluate six months of treatment in approximately 1,000 patients with homozygous for the F delta 508 mutation. We are on track to submit an NDA in the US with this combination in 2014. The combination of Lumacaftor and Ivacaftor has also received breakthrough designation from the FDA. In the second-quarter, we reported Phase 2 data with another corrector, VX-661 that further validated the strategy of combining a corrector and a potentiator to treat CF patients with two copies of the Delta 508 mutation.

In the second half of 2013, we plan to conduct a study of VX-661 in combination with Ivacaftor in G511D delta 508 heterozygous patients. Our in vitro data supports the potential for further benefit for these patients. In hepatitis C, we have generated nearly $450 million in hepatitis C revenues and royalties in the first half of this year. We have initiated multiple studies of VX-135 in combination with ribavirin, and separately in combination with daclatasvir. Despite the partial clinical hold placed on the US study, our strategy remains clear, to identify all oral and high viral cure regimens containing VX-135 with durations of 12 weeks or less in multiple genotypes.

Commercially, we continue to generate significant revenue in the hepatitis C market with INCIVO-based therapy, although the number of new patients initiating treatment continues to decline. In CF, KALYDECO reimbursement and uptake in Europe has been very strong, resulting in significant growth in our KALYDECO revenues.

Financially, we enhanced our strength by seeking to maximize revenues while making prudent investments focused on R&D and controlling expenses. These actions are reflected in our Q2 results. Importantly, we ended the quarter with over $1.4 billion of cash, cash equivalents, and marketable securities, with no outstanding convertible debt. In summary, we are creating hope for people with CF and HCV, value for shareholders, and a platform for long-term growth by executing on our strategy. Now, I will turn it over to Bob.

Thank you, Jeff, and good evening. Tonight, I will review the status and timing of our three ongoing studies with VX-135. Then, I will also review our CF development pipeline and provide a few updates based on our progress to date.

First, VX-135 for hepatitis C. We have initiated a 12-week study of 100-milligrams and 200-milligrams of VX-135, plus daclatasvir in 20 genotype 1 patients in New Zealand. Pending data from this initial cohort of patients, we plan to expand the study to enroll additional patients with both genotypes 1 and 3, and to evaluate an eight-week combination regimen in genotype 1. Safety and efficacy results from the first part of the study are expected to be available in early 2014.

In our European study of VX-135 plus ribavirin, 20 patients, 10 in each of the 100-milligram and 200-milligram arms have now completed 12 weeks of dosing. Both the 100-milligram and 200-milligram doses were well tolerated, with no discontinuations and no serious adverse events reported. 100% of patients achieved undetectable HCV RNA during the 12-week dosing period, and 70% and 80% of patients in the 100- and 200-milligram dosing arms, respectively, had undetectable virus within four weeks of initiating treatment. We expect to provide complete safety and efficacy results from the 100- and 200-milligram arms in the second half of 2013.

In the US, dosing and evaluation is ongoing in the study of 100-milligrams of VX-135 plus ribavirin. 10 patients with genotype 1 hepatitis C are enrolled in this dose group, and all patients have now completed at least 10 weeks of treatment. We are working to provide to the FDA the requested clinical, preclinical and pharmacokinetic data from ongoing VX-135 studies, with the goal of resolving the partial clinical hold, and supporting the evaluation of a 200-milligram dose of VX-135 in the US. We expect to complete submission of the requested data to the FDA in the fourth quarter. Safety and efficacy results for the 100-milligram arm of the study are expected to be available later this year.

In addition, our partner J&J has completed a drug-drug interaction study of VX-135 in combination with Simeprevir in healthy volunteers, and we expect dosing to commence in genotype 1 patients in the second half of 2013, pending data analysis and dose selection. Our strategy continues to be one of identifying high viral cure regimens of 12 weeks or less, effective in multiple genotypes, and we have multiple trials planned or underway to fulfill this strategy.

Now, moving to cystic fibrosis. Our most advanced compound is KALYDECO, which is indicated for the treatment of G551D patients over the age of six. We are studying this compound known generically as Ivacaftor in a number of additional patient populations that we believe have the potential to benefit from KALYDECO based on in vitro data along with our understanding of the disease biology and KALYDECO mechanism of action.

We have three Phase 3 label expansion studies underway, and we reported top line results in the first of these studies, the non-G551D gating study today. Ivacaftor produced statistically significant lung function improvements in patients with the nine characterized non-G551D gating mutations. This was a blinded cross-over study design.

A schematic of the study is shown on slide 11 of the webcast. Patients were randomly assigned to receive either Ivacaftor or placebo for eight weeks, then go through a full week washout. And then receive the alternate treatment for eight weeks, that is through weeks 12 to 20. This study included 39 participants ages six and over, who have at least one of the nine characterized non-G551D gating mutations.

The primary end point was absolute change from baseline to percent predicted FEV1. In the study, the mean absolute treatment difference between Ivacaftor and placebo was 10.7%, with a p value of less than 0.0001. And the mean relative difference was 14.2%, again, with a p value of less than 0.0001, as measured by FEV-1 through the eight-week treatment period. The mean, absolute and relative FEV1 improvements during Ivacaftor treatment, that is the within group changes, were 7.5% and 10.8%, respectively. Both were highly significant, with p values of less than 0.0001.

We also achieved statistical significance for secondary end points of weight gain, and improvement in the respiratory domain of the CFQ-R, the patient-reported quality of life measure. Safety was similar to the experience to date with KALYDECO and G551D. Ivacaftor was well-tolerated. And in the study, the most common adverse events were pulmonary exacerbation, cough, headache, and abdominal pain, each occurring more frequently while patients received placebo.

These results are another example of the correlation of clinical data with in vitro chloride transport data that we have seen across the CF program. In the second half of 2013, we expect to present full data from this study at a medical meeting. Also in the second half of 2013, we intend to file both an sNDA in the US and an MAA variation in Europe, for use of Ivacaftor in patients with other gating mutations.

We have two additional Phase 3 label-expansion studies ongoing in CF patients, in patients with the R117H mutation, and in children ages two through five with the gating mutation, including G551D. Slide 12 shows the design of each of these studies. We expect the results from the R117H study in the second half of this year, and pending study results, we expect to submit an sNDA in early 2014.

In the study in children ages two through five, we have now completed the PK portion of the study and we have selected a dose. Enrollment in the 24-week dosing portion of this study is now ongoing, and we expect results in mid-2014.

We are also conducting a proof-of-concept study in patients with residual CFT-R function, also known as the n-of-1 study. The design of this study is shown on slide 13. Like the gating study, the n-of-1 study has a cross-over design. Cross-over designs have increased power to show a treatment effect, by focusing on the drug-on and drug-off effects in individual patients, which may reduce variability. Data from this study are expected in the first half of 2014.

Our CFT-R corrector, Lumacaftor or VX-809, is in two global Phase 3 studies, traffic and transport, to determine its efficacy and safety in people with two copies of the F508del mutation. Site initiation and recruitment is going well, and we expect to complete enrollment in both studies in the second half of 2013, and we are on track for reporting data and submitting an NDA and MAA in 2014.

As part of the Lumacaftor registration program, we are planning to start an 8-week study of Lumacaftor in combination with Ivacaftor in F508del heterozygous patients in the US in the second half of 2013. This placebo-controlled Phase 2 study is part of our registration package agreed with the FDA earlier this year, and is expected to enroll about 80 patients randomized between active and placebo arms. Based on our prior data in this population, we do not anticipate that the Lumacaftor and Ivacaftor combination will have a clinically meaningful benefit for heterozygous patients, but the study has been requested by the FDA for labeling purposes.

Tonight, we are also announcing a new study for VX-661, another corrector we have in clinical development. We plan to evaluate VX-661 plus Ivacaftor in a proof-of-concept clinical study for the mechanism of using a corrector and potentiator in heterozygous patients who have the G551D mutation, a gating mutation on one allele, and F508del on the other allele. The patients who enroll in this study will already be receiving KALYDECO, as labeled in the US and EU. We expect to start this four week Phase 2 study shortly in approximately 20 patients, and report data in late 2013, or early 2014.

In addition, we are planning to start a Phase 2 proof-of-concept study with the corrector VX-983, pending the outcome of ongoing studies and discussions with regulators. The results from these Phase 2 studies with VX-661 and VX-983 will be important factors, as we consider which molecule to prioritize for further development as part of dual or triple combination regimens.

We are also actively working to advance second-generation correctors that act on a different part of the CFT-R folding process and could be used as part of a dual corrector regimen, along with Ivacaftor to further enhance benefit for patients who are homozygous or heterozygous for F508del. Our goal is to bring a second-generation corrector into clinical development by the end of 2014.

At the recent ECSF meeting, Dr. Fred Van Goor of Vertex presented data that showed a combination of two correctors plus Ivacaftor nearly doubled chloride transport in human bronchial epithelial cells with two copies of the F508del mutation, compared to the use of a single-corrector plus Ivacaftor. In summary, we are pleased to have a full pipeline of medicines that may provide life-changing treatment for more people with CF. And now, I will turn it over to Stuart.

Thank you, Bob, and good evening. I will provide some background and commentary on our improvements in INCIVEK for genotype 1 chronic hepatitis C infection, and KALYDECO for CF patients who have the G511D mutation. Starting with INCIVEK. Since launching INCIVEK in May 2011, we have treated more than 65,000 patients in the US and Canada, and recorded approximately $2.5 billion in cumulative revenues.

While we see ongoing demand for hepatitis C treatment for motivated patients and those who need to be treated now, we are seeing a decline in the number of patients treated in anticipation of the next generation of medicines to treat hepatitis C. We expect this decline to continue for the remainder of the year. However, INCIVEK continues to generate significant revenues, and so is an important component of building the financial strength we need to fund our future opportunities. Our net revenues from INCIVEK in North America were $156 million in the second-quarter, as we maintained our market-leading position even in this declining market.

Outside of the US, telaprevir performed well in the first half of 2013. The availability of INCIVO marketed by Janssen had expanded to additional countries, and based on Janssen's sales in the second-quarter, Vertex earned $44 million in royalties in Q2. This was largely driven by performance in Latin America. However, the outlook for INCIVO for the remainder of 2013 is difficult to forecast, due to seasonal prescribing patterns, and the potential reduction in patients initiating therapy in advance of new regimens entering the market.

Moving to KALYDECO, worldwide revenues were $99 million for the second-quarter of 2013, including US sales of $55 million based on the rapid adoption of KALYDECO in the US, following our launch in early 2012. Over the first half of this year, we have successfully completed reimbursement discussions in most of the major EU markets, including England, Scotland, Ireland, Northern Ireland, and Wales, and made KALYDECO available for the many patients who need it. Since then, we have seen a similar pattern of rapid uptake of KALYDECO, as we saw in the US last year. I am proud of the incredibly effective efforts of our field organization, which together with the dedicated healthcare providers who care for CF patients, have enabled us to begin treating nearly all the eligible G551D patients in the EU.

We achieved approximately $44 million in KALYDECO sales outside of the US in the second-quarter, which represents significant growth over the first quarter. This figure does include approximately $2.5 million of inventory build in Europe.

We do not see further meaningful growth for KALYDECO in 2013. However, we do see three potential drivers for revenue growth beginning in 2014, subject to the successful completion of reimbursement and regulatory discussions, and likely to occur at different times throughout the year.

First, achieving public reimbursement in Australia and Canada. KALYDECO was approved in Australia earlier this month, and is the third largest potential market for KALYDECO after the US and England, with approximately 200 eligible G551D patients over the age of six. We estimate that there are approximately 100 eligible G551D patients in Canada.

Second, regulatory approvals for patients with non-G551D gating mutations, based on the data that Bob described earlier. This could allow for an expanded label as early as the first half of 2014, when we could begin to add as many as 400 additional cystic fibrosis patients globally, based on a label for all gating mutations.

Third, potential for future growth in non-gating mutations, such as residual function patients, including R117H. This will be based on successful outcomes from the ongoing clinical trials that Bob described earlier. In summary, I am pleased with the progress we have made in the first half of 2013, and would like to thank all of those who have contributed to these results. And now I will turn it over to Ian.

Thank you, Stuart. I will start by reaffirming our business and financial strategy. We are investing and advancing important medicines to drive sustained, long-term revenue and earnings growth. Our goal during this period is to maintain financial strength by prioritizing our spend towards R&D, and balancing our expenses with revenues while protecting our balance sheet. I am happy to report that we remain nicely on track.

Now to the second-quarter of 2013 results. Total revenues were $311 million. Our GAAP -- total non-GAAP operating expenses, excluding cost of revenues and other charges, were $281 million for a non-GAAP loss of $6.2 million or $0.03 per share.

From a balance sheet perspective, we completed the quarter with over $1.4 billion of cash, cash equivalents and marketable securities, and no convertible debt. This was the result of the June debt to equity conversion. These results, and specifically our balance sheet strength, will enable us to continue to maintain our investment in our priority medicines.

Now for the more detailed results. In the second-quarter, total KALYDECO revenues were approximately $99 million. EU uptake in the second-quarter exceeded our expectations, and KALYDECO is now treating nearly all of the eligible G551D patients in the US and EU. As such, we are raising our guidance for 2013 KALYDECO revenues to $345 million to $360 million.

INCIVEK revenues in the second-quarter were $156 million, reflecting ongoing demand for hepatitis C treatment even as the market anticipates the arrival of other regimens in 2014. INCIVO royalties were $44 million, and our combined hepatitis C revenues contributed approximately $200 million to our top line in the second-quarter. Finally, collaborative and other revenues in the second-quarter were $12 million.

Because of the increased visibility into revenues for 2013, we are tightening our 2013 total revenue guidance range to $1.1 billion to $1.2 billion, which includes the increased KALYDECO guidance of $345 million to $360 million.

Now to our non-GAAP operating expense of $281 million, which includes cost of revenues and other charges. This is similar to the first quarter, made up of $191 million in R&D investments, and $90 million in SG&A expenses. Consistent with our financial strategy, we are prioritizing investment in R&D, and being disciplined with our SG&A spending. We are reiterating our non-GAAP operating expense of $1.09 billion to $1.15 billion for 2013.

In summary, financially we are executing on our strategy, and prioritizing investment within our business to drive long-term growth. Our cash position and our financial strength will enable us to continue to maintain our investment to progress our medicines in development. I would now like to open the line to questions.

(Operator Instructions)

The first question comes from Geoff Meacham from JPMorgan.

Hi. Thanks for the question, and congrats on the CF data. One on CF and one on hep C. So for the 809 KALYDECO trial on heterozygous, are there any limitations on enrollment for genotype beyond G551D, that is on a non delta F allele? I just want to make sure there are maybe -- you are excluding patients that may have some residual KALYDECO function. And I guess the question there is, if there is a negative control for labeling, but you see a signal, then what is the path beyond that?

This is Bob. It is hard to speculate on the path, until we actually see what the data show. But we choose the patients for this study who have mutations on the other allele, that would not, not respond to KALYDECO. So we obviously, try to keep our baseline at an absolute minimum. And that is what we have done throughout the heterozygous studies and we will do the same in this one.

Got you. And just a real quick one on VX-135. I know you have been asked a lot about spend in Hep B. Is there a time to get off the partial hold from the FDA, or some data set that you view as a trigger for making future investments in 135? Thanks.

This is Bob. I will answer sort of the first part of that. And that is, based on the particulars that the FDA has asked us for, we expect to be able to have a complete data set in the early part of the fourth quarter. And we then expect the FDA to review those data and then respond back to us. As to the other part, I will pass it on to Ian.

Yes, hi, Geoff. Thanks for the question. The question has been asked a few times since Thursday night actually. And the way we -- we are very aware of the competitive nature of this market, this disease area. And we are always looking and making a decision to see whether we can compete in what is a significant market, and a global market.

At this point in time, we have the assets, and we have the internal capability that we want to progress that opportunity. As we go through, there will be a very important Phase 2 period of establishing a potential regimen to compete in this market. We will be making those kind of decisions. But at this point in time, we plan to keep committed to HCV and drive forward for an all-oral regimen.

Okay, thanks.

The next question comes from Geoffrey Porges from Sanford Bernstein.

Thanks very much for taking the question, and also, congrats on the additional CF Phase 3 data. Perhaps just to drill in a little bit more on 135, Bob, could you let us know, first, what the primary metabolic pathway if for VX-135? Secondly, the thinking behind taking it up to the 400-milligram dose? And then what the outcome was for those patients with the elevated liver functions? And lastly, can you just tell us whether you have seen any transaminase elevation, not enough to get to a [SAU] or discontinuation? But any at all at the 200-milligram dose?

So I'll take -- if I can remember all four of those questions, I will take them in order. The pathway of metabolism is just primarily renal excreted as most [nucs] are.

Okay.

In terms of why we went to the 400-milligram dose, I would say that part of the Phase 2 evaluation of any compound really is to explore the dose range. When we had completed the 100- and 200-milligram doses in the European study we have very good tolerability, and therefore we decided to dose escalate. And we wanted to sort of maximize the viral kinetics in order to get the best response possible.

The outcome of the elevated transaminase patients, and then any at 200?

Yes. There were none in 200. More at 100. And they all resolved really quickly, within a week or so.

Okay. Thank you.

And in terms of the other transaminase elevations, I guess I just answered that by saying, we really haven't seen anything in the 100- and 200-milligram groups.

Right. That's very helpful. Appreciate it.

The next question comes from Michael Yee from RBC Capital Markets.

Yes, hi, thanks. Question on Hep C and then on CF. On the 135, can you walk through what type of PK exposure you are seeing at 100, 200 and 400? And maybe whether there is any sort of overlap there? Just trying to think about exposure, 400 is only twice 200, so think about that.

And then on CF, can you explain the confidence in the other non-gating mutations. Excuse me, the, the residual function studies, where the confidence comes from then? And what I would say is obviously, that is a huge population. So just trying to understand whether there's any read-through in today's data into that study? Thanks.

Okay. So in terms of 135, I will say the following. Firstly, in the three subjects who had LFT abnormalities in the 400-milligram group, the exposures in those three patients were essentially the highest of any of the 10 subjects that were enrolled in that dose group. And in addition, the exposures in the 200-milligram group differed on average by about twofold, with those in the 400-milligram group.

So there really is a fair spread between those subjects who had the LFT abnormalities in the 400-milligram group and the 200-milligram group. On the other hand, I will just say that this is a very small data set. And clearly, we have to establish this in larger numbers of patients to really understand this. But that is really where we are headed in the next set of evaluations.

Okay. And then on CF?

Yes. So you asked about our confidence in the patients with residual function. And I think that has largely to do with, first of all, both in vitro data that show that we can enhance the chloride transport in patients, in cells who carry the R117H mutation, and other mutations that result in residual function. And we have had a very good in vitro/in vivo correlation, correlating chloride transport in vitro and clinical results so far in the program, and we would expect that to continue in this group as well.

Next question, please?

The next question comes from Mark Schoenebaum from ISI Group.

Hi. Thanks for taking the question. Maybe the first one for Bob. Bob, I think the concern, I guess, for 135 is the potential for mitochondrial toxicity. Can you walk us through evidence we have thus far that this drug is not -- does not have mitochondrial toxicity? Maybe make an argument to try to convince us all?

And then number two is, the second question is an expense question. Maybe the first part for Jeff. And that is $1.4 billion of cash on the balance sheet is quite a bit, very formidable. Be curious to get your updated thoughts on what you are going to do with that cash? What kind of stuff conceptually is on the table versus off the table? And maybe I will just stop there. Thank you very much.

So this is Bob. I will answer the first part. We have done an extensive evaluation of VX-135 for mitochondrial toxicity, using all state-of-the-art techniques. And in fact, we have seen no evidence for mitochondrial toxicity for this compound. And that is really about all I can say.

And Mark, as to your second question, I will just comment on the cash, and turn it over to Jeff about more strategically where we may go. But from a cash position, as you can see, we are in this position where we are accumulating. We want to maintain our strength from the balance sheet. That includes managing the liabilities on the balance sheet. So we are happy we have got over $1.4 billion at June 30, and we've been able to convert our debt into equity, so we don't have any significant liability or debt on the balance sheet.

So at this point, we are still in the phase of building. I think you will see that continuing for at least a short period of time. We are very aware as we go into 2014 that as -- sorry -- INCIVEK revenues decline, that we may need to apply some of the cash from the balance sheet to continue to invest in our medicines. As we come through that period, I think that is when your question becomes most pertinent, in terms of how do we put the capital to use? And we are starting to think about that in our business at this point.

Yes, Mike, this is Jeff. A quick comment to add to what Bob said on the mitochondrial toxicity. Yes, one of the other things that we learned from the 100- and 200-milligram trial that was done in Europe, was that we were monitoring those patients by both echo and EKG. And I think one of the important findings is, now that those patients have all concluded 12 weeks of therapy, we really didn't see any echo or EKG signs of cardiac pathology, which obviously, is the mitochondrial toxicity of the most interest, and frankly, the most worrying (inaudible).

Thanks a lot.

With respect to the $1.4 billion, only thing I would add is what I have always said before. We are very pleased with our own pipeline, and we think that we have a number of interesting assets coming forward. But we are also always looking strategically for things that can complement what we have done, certainly in CF and HCV, but also in other areas that fit our specialty and transformative medicine strategy.

The next question comes from Rachel McMinn from Bank of America Merrill Lynch.

Thanks. I appreciate the question. So I guess, I wanted to ask Geoff's question a little bit differently. I mean, given the news on the VX-135 from last week, how should we think about your SG&A infrastructure going forward? I recognize that you are not ready to terminate the program. You are not saying anything of the sort. But I think you had previously alluded to keeping your HCV sales force through VX-135 development. How should we think about potential cost savings following the Gilead launch?

And then separately, I wanted to better understand the rationale for the study design for 661 in 551D patients. There is already a very strong benefit with KALYDECO alone. So I am just wondering if you think you will be hitting up against the ceiling, and why you wouldn't be looking at 809 in that setting. Thanks.

So Rachel, it is Stuart here. I will take the first question on HCV SG&A, and then hand it over to Bob. So on our HCV investment, I mean we have an absolutely world-class HCV sales force. And although it is relatively small, one of the beauties of the specialty business model, is it's a small sales force. We know it has been instrumental in our success to date. However, as with all aspects of our business, we are constantly evaluating the return on investment we get for that sales force.

And in our judgment, the sales force continues to be a really good investment, based on the revenues that they are generating today. What I would say though, is we are, as we say, looking at spend across all of our areas, and we have substantially reduced in the HCV area, our marketing expenses from 2012 into 2013. And that continues to be the case going into the second half of 2013, where we are looking very much at managing those expenses, in line with the trajectory of our revenues. And on the CF question, I will hand it over to Bob.

Right. So your question relates to the combination of 661 plus Ivacaftor. And the rationale for this is based on in vitro data, again which is generated Vertex to show that you can actually get a substantially greater effect on chloride transport in vitro in G551D F508del heterozygous cells in culture. And the rationale is whether in fact, in vivo, in humans, whether a greater clinical benefit could be obtained by this combination, again, going back to the correlation that we have seen between in vitro and in vivo.

As to the choice of 661, I would just say that VX-809 is in an extensive program, a Phase 3 program with many, many studies. This is another opportunity for us to learn more about 661, and that is why we chose it for the combination.

Okay, thanks.

The next question comes from Brian Abrahams from Wells Fargo.

Thanks very much for taking my question, and congrats on all the progress in CF. Question on 135. I think what a lot of us are trying to figure out is whether the 100 or 200 mg dose is could potentially be sufficient. I know rapidity of [RNA] reductions are often seen as a surrogate for potency. You reported 70% and 80% RVRs. Obviously, comparing across small studies has limitations.

But I was wondering if you could help us understand how the population in this European study may have compared to that of the analogous study conducted for sofosbuvir plus ribavirin, and whether you had any details of characteristics of patients who were maybe a bit slower to respond in the study. And maybe could you possibly give us RVR results for those US patients on the 100 mg? That would help as well. Thank you.

Brian, I will just jump in front of Bob. We are not going to compare studies, cost studies actually. We are not going to compare patient demographics between a study that was run by another company and ourselves. We are waiting for the full data to be read out, and then you can do the analysis at that point in time. But maybe Bob would like to comment on how we think about 100 and 200, and (inaudible) as you said Brian. But maybe that is where it is coming from.

Yes, so it is really impossible to ever match across studies when you don't have all the data. So I just -- we just can't really do that.

I would say that obviously in our, in our VK studies, we have seen very good antiviral activity, both 100 and 200. The results in our 200-milligram group compare favorably with the results that have been seen for sofosbuvir at 400 milligrams, and we consider both of those doses to be really, have very good activity.

As to the RVR rates that we saw in this trial, I would just remind you that our real primary development path for VX-135 is in combination with the other antiviral agents, not with ribavirin. And I think we should wait to see how our daclatasvir plus 135 study comes out, to really make judgments on 100- versus 200-milligrams and the efficacy we might see. I think it would be best to wait to see where there is more, in terms of more realistic studies in terms of ultimate development path, how they come out.

Got you. No I understand. It might be a moot point when you end an NS5A anyway. Thanks very much.

Yes.

The next question comes from Liisa Bayko from JMP Securities.

Hi, thanks for taking my questions. I just wanted to know if you could comment at all on some of the other observations in terms of exacerbations of sweat chloride that you saw with the current study results in the non-G511D gating.

So this is Bob. We have just really announced our top line results. We are obviously combing through the data of that study much more extensively now. And I think I would probably prefer to wait until those results are presented at a medical meeting to discuss any further details.

Okay. Fair enough. And then in the new study that you have initiated, the heterozygous study, G551D plus a background of delta 508. I am just curious of the G551D population, what percentage of them have F508 on the other allele?

About half of them do.

Okay. That's helpful. And then for the 400 patients, that R117H, can you maybe talk about the geographic distribution? I know you spoke about 400 patients.

It's roughly evenly split between the US and EU. And it is actually probably slightly more in the European Union, but it's roughly evenly split.

Okay.

Maybe we could -- I just want to clarify. I actually misspoke when I answered your question. In terms of G551D F508del -- probably close to 80% of those patients actually have F508del on the other allele. Sorry. Not 50%.

That's what I thought. Okay.

Yes. It's higher.

Okay. And then just final question, again, on cystic fibrosis, for G551D, I know you said you have treated now the majority of patients in the US and Europe. Can you maybe talk about how many other patients might be in other world areas? Just trying to understand how much more upside there is in KALYDECO for the G551D population. Thanks.

Yes, the most significant other markets where we are yet to get full reimbursement are Canada, where we think there is about 100 eligible patients. And then Australia, where there is about 200 eligible patients.

And clearly, that has all to do with the G551D mutation being called kind of the Celtic mutation. Those are areas where there was obviously a lot of immigration. So those are really the remaining areas of growth for us for the G551D population, and really their reimbursement is going to be the key that will unlock the drug to be made available for those patients.

Thanks a lot. Can I just sneak in one more question? Can you maybe talk about compliance so far on KALYDECO, kind of what -- now that it has been on the market for sometime at least in the US and some of the bigger European countries, what sort of normalized compliance rates are we observing? And that is my final question. Thank you.

Sure. So what I would say is that we are seeing rates of compliance which are as high as certainly I have ever seen in any therapy area. And I would say they are above 80% which is really pretty substantial.

Thank you very much.

The next question comes from Terence Flynn from Goldman Sachs.

Hi. Thanks for taking the question. Just two for me. So in terms of the other gating trial that you reported today, are you -- can you give us any sense of if you collected data on pancreatic enzyme usage? And then with respect to the new trial of 661 KALYDECO, was wondering if that is going to include a KALYDECO control arm? Or I guess, how should we think about benchmarking the combo data when we see those results? Thanks.

This is Bob. In terms of the other gating mutations, obviously we collect data on concomitant medications, but we are not conducting sort of what I would consider a really rigorous assessment of the use of pancreatic enzymes. We would be able to get some sense of it from the data we have, but it is really not designed as a key end point of the study.

And in terms of 661 with KALYDECO, remember that we will have historical data on those patients because they are on KALYDECO already. So when you add 661 on top, we would be able to compare their change from their prior baseline, which we will have measured, rather than having a concurrent control.

Thanks a lot.

These are all people eligible that are for KALYDECO and are taking it commercially, so you can't take them off the treatment.

The next question comes from Matthew Roden from UBS Securities.

Great. Thanks for taking the question. Bob, if I understood your comments on the VX-135 study with daclatasvir, it sounds like you are looking to expand the -- beyond the 20 patients that you're initially going to be expecting once you have those data.

The question I am trying to get to is, what you need to see in order to advance the molecule into registrational studies in terms of either SVRs or duration of therapy? And also related to that, when would that decision be taken? Sounds to me like regardless of what you see in the first 20 patients, that given the small sample size, you would be -- it would be warranted to take that into larger studies.

So yes, this is Bob. Yes, I agree. We would likely expand that study just to get a better sense of the treatment effect before launching into a pivotal program. The initial 20 patients are really primarily a safety PK and preliminary efficacy read assessment. We know the bar is very high in this therapeutic area. So obviously we would like to see very good efficacy, as well as good tolerability, but it may be that we will have to see some more patients before we really get a complete handle on that.

Okay. That's very helpful. Just real quick on CF--

This is Peter speaking. On top of that, I think it's really important for all of those studies we need in our safety database to move forward to longer-term pivotal type of studies. And I think that is another reason why we have to basically enhance the population in the regimen that you are going to consider later. And so this is outside of efficacy, an equally important component that you have the right safety description.

Yes. Thank you, Peter. And then a quick one for Ian or Stuart on CF. I just want to make sure that I understand the KALYDECO guidance. It looks like you have already done about $160 million, $161 million year-to-date. And if you just had sequentially flat sales in the second half, from 2Q run rate, you would already hit the top end of guidance. I heard what you said, in terms of inventory in Europe. Just trying to understand, under what scenarios you could possibly -- the bottom of the guidance could possibly be in play? Thanks.

Yes. Good question, Matt. So as we set the guidance, I mean, the numbers you called out are right on. And but there was a $2 million to $3 million inventory build, and therefore the demand in the second-quarter, even though we record $99 million of sales, the demand is maybe close to $95 million, $96 million.

And then if you do flat line that out, it puts you at around the $350 million or just above the $350 million mark, if you are doing the math that you are doing. That $350 million mark is, or just north of $350 million is actually right in the middle of the guidance we gave of $345 million to $360 million. So that's how we gave it.

We didn't want to put a precise number. But we do think it is appropriate and conservative to put a range around the trajectory of the number of patients which we are treating at this point in time.

I would just say that even though we're commenting on that range and saying we are treating nearly all the patients, the G551 patients in US and Europe now, the growth, as Stuart mentioned on the call, the growth we do anticipate to start again in 2014, and that does come from these other geographies that we start to sell drug into that we have been approved and we are looking for reimbursement. And that it comes with label expansion, to the study we are announcing today, and then the other studies that we are running. And those can come throughout 2014, giving us kind of a nice growth trajectory of an important revenue line item in 2014.

Great. Thanks, and congrats on the progress in CF.

The next question comes from Robin Karnauskas from Deutsche Bank.

Thanks for taking my question. So I guess if you were to see a really significant improvement by adding 661 to KALYDECO in these patients, do you have a sense of how nimble you can be to advancing the program forward for approval, now that you have had some more time to understand what breakthrough designation gives you?

And then the second question I have, how many patients do you think that you need to see with 135, say at 100 milligrams or 200 milligrams to be comfortable you will not see the liver tox you saw in the 400-milligram dose? Thanks.

I will take the second one first. In terms of how many patients, it is very hard to say. Obviously, statistically, if you study 300 patients and you don't see it, you can say that it's a rate of less than 1%. It all just depends on the end.

I think as Peter mentioned, as we begin to build our safety database towards pivotal studies, we will be in the 100s range for patient numbers. And that will give us a fair amount of confidence that we are not going to see it at any substantial rate.

And Robyn, for the first question with 661 and 770, in those heterozygous 551 508, patients if we see a positive signal, I think the first thing is we have to basically show enough safety data for 661, which we haven't at this given point in time. And what that means is you have to at least conduct a trial risk, a couple hundred patients at the end of the day, to be in a label position that the agencies might accept.

That's helpful. And just as a follow-up, do you feel that the fact that the liver signal was not predicted preclinically that the -- and the mechanism of the tox is not fully understood, that that could be a hurdle with the FDA? Or is it really just all about statistics and showing that the incidence is well below a certain number to get people comfortable at the FDA?

That is 135 now, I was talking about --

Yes, sorry, I was going back and forth.

Right.

Yes, with 135, I think obviously the clinical data will likely trump any other information. And I think that is really what we are largely focused on. As I mentioned, in terms of in vitro data, we have done a very extensive evaluation for mitochondrial effects and there just don't seem to be any. So we are pretty confident in that regard.

Thanks.

The next question comes from Yaron Werber from Citi.

This is [Kumar] in for Yaron. Thank you for taking my question. What is the strategy for VX-809 JAK inhibitor? And also, can the time line for the non-G551D label expansion be shorter given the breakthrough designation?

And so, I will ask Peter to talk about the breakthrough designation, how that affects time line for VX-809. And I think your first question, I didn't quite catch your name, but--

Kumar.

Kumar? The first question, I think you are referring to VX-509, the JAK inhibitor. And our strategy remains as we described at the beginning of this year is that we think the most value can be captured from VX-509 by -- and recruiting partner help. So it is something that we are evaluating, partnering opportunities to drive that molecule forward in multiple disease areas.

We are aware the market is competitive and the disease areas are broad and big, and it's a global market. And we believe to get the maximum benefit from this molecule, the best decision is for an out licensing and we are currently looking at and evaluating opportunities to do that. To the, to VX-809 in CF, I will ask Peter to comment on how breakthrough designation may affect it.

So Kumar, there are many different ways how breakthrough works. So the first, the first is you have basically more frequent interactions with regulatory authorities with the clear goal to come forward with an acceptable fast development program that authorities can accept.

In that context, we already got an acceleration because as you well know, we went down from 48 weeks or 52 weeks to a 24-week regimen, which basically cuts it substantially faster. And then also the clear-cut agreement about several doses. And no basically mono arm and all those types of things basically accelerated the overall program by about a year and a half. That is already good news.

Now obviously, the next piece is the registration and the approval process. And what regulatory authorities normally do is they talk with you about how you basically can file, so that this may be more accelerated. But it doesn't set a time line. It doesn't give you, it is two months or three months or five months or eight months. It is dependent on the data set that you have.

What might happen, which just might be a benefit that you do not have to do an [add port] -- which is something they basically sort of say at different meetings, which could also be beneficial and going positive to the time line. That's about what it is. It is the closing direction with the agency on a daily basis to make good progress.

That's very helpful. Thank you.

The next question comes from Ying Huang from Barclays.

Thanks for taking my questions as well. So I have a couple on VX-135 first. Were there any [Hyatt's law] cases reported or observed in that 400-milligram dose cohort in Europe? And then also, do you guys still plan to start a pivotal, given by the end of 2014, if assuming 100 milligrams just goes well?

So this is Bob, in terms of your first question, no, there were no cases of Hyatt's law in any of the subjects, who had elevated liver function tests. Can you repeat the second question?

Yes, if 100-milligram actually looks good and everything goes well, do you still plan to start a pivotal development of VX-135 by the end of next year?

Yes. I mean, our goal is to, as we have said, to get the results of these studies, to provide them to the FDA in the hopes of resolving the clinical hold, so that we can assess higher doses in the US. And at the same time, to expand the study once we are able to -- the daclatasvir study in particular. And I think our ultimate regimen will be guided by the data. If 100 milligrams looks good, we will consider taking that into Phase 3. It really is just very data-driven.

Thanks, Bob. And then I have one more on VX-661 here. Anything you have seen in the preclinical model that leads you to believe this compound should behave differently from VX-809 in heterozygous patients or not?

No, I don't think there is anything particular that would make one think that one way or the other, in terms of 809 versus 661.

Okay, great. Oh, if I may ask just one more. Should we expect anything reported at AASLD for 135?

I can't really comment on where our data are going to be presented at this stage of the game. Have to just stay tuned.

We can say that when the certain study is complete, we generally provide a top line press release in terms of those study results when we have both safety and efficacy. So you can expect that and generally more complete data at the medical conferences to follow.

Thank you, Ian.

Operator, this is Michael. It's now nearly 6.00. We have time for two more questions.

Thank you. The next question comes from David Friedman from Morgan Stanley.

Hi, thanks for taking the question. It's a question on 983. I was wondering if you could just talk a little bit about what the -- what final steps are necessary until that is in humans. And then also, if you have any more detail beyond just the brief sentence in the press release about your thoughts on some initial studies with that, with that drug?

All right. This is Peter. So first is basically, as it is true for all experimental medicines, you have to do a serious pre-clinical and early clinical assessments, which means tox studies, PK studies, any of those. And we are currently conducting these, and we'll see what the outcome is, and come back to you and see what strategy we can do. It's all data driven.

Hypothetically, 983 acts very similar to 809 and 661. And you can assume a similar development path as you have seen for 809 or 661 as we go forward, in case everything goes positive. (Inaudible). It already in humans anyhow. So -- but I think it is not in our POC study, if that is what you mean.

Yes. We've completed the Phase 1 healthy volunteer study.

Yes.

And right now, we are considering what the next study would be with the compounding in combination with KALYDECO. And also having, kind of the normal regulatory discussions about what study we may proceed ahead with. And we will be back to you later this year when we have greater clarity on that.

Great. Thanks a lot.

The final question comes from Salveen Richter from Canaccord.

Thanks for squeezing us in. Just two qualitative questions really briefly. Given the -- for CF, given in vitro to in vivo correlation, how do the in vitro results for the R117H and residual mutation data look versus the non-G551D gating mutation?

Peter, do you just want to close on this?

Yes. So basically, from a sensitivity in vitro 117 is highly sensitive, and we have therefore some hopes that it might also work at the end of the day in the clinic. Now in general, the comment is we have your gating type of mutations. I might just make a general comment. It's not the same for each and every class of mutations, and I want to just put that word of precaution out, because there's a difference between [drastic] mutations and gating mutations.

Got you. And then just briefly for the gating trial, the data today. Was the eight-week data in line with the 20-week data? The pre cross-over eight-week data, that is?

You mean each, each period?

Yes.

Yes.

Yes, so yes, so we looked at that specifically to see whether there was any carry-over effect after the washout period, and in fact there was no carry-over effect. The results whether you gave it first or second, were very comparable to each other.

Great. Thanks so much.

So thanks, everybody, for tuning in tonight. If you have additional questions, we are in our offices. We'll be happy to take your call. Have a good evening.

Ladies and gentlemen, that does conclude the conference for today. Again, thank you for your participation. You may all disconnect. Have a good day.