Gilead Sciences, Inc. (GILD) Q1 2012 Earnings Report, Transcript and Summary

Susan Hubbard - Vice President of Investor Relations Robin L. Washington - Chief Financial Officer, Principal Accounting Officer and Senior Vice President Kevin B. Young - Executive Vice President of Commercial Operations Norbert W. Bischofberger - Chief Scientific Officer and Executive Vice President of Research & Development John F. Milligan - President and Chief Operating Officer

Geoffrey C. Meacham - JP Morgan Chase & Co, Research Division Mark J. Schoenebaum - ISI Group Inc., Research Division Rachel L. McMinn - BofA Merrill Lynch, Research Division Matthew Roden - UBS Investment Bank, Research Division Yaron Werber - Citigroup Inc, Research Division Sapna Srivastava - Goldman Sachs Group Inc., Research Division Robyn Karnauskas - Deutsche Bank AG, Research Division Thomas Wei - Jefferies & Company, Inc., Research Division Michael J. Yee - RBC Capital Markets, LLC, Research Division Lee Kalowski - Crédit Suisse AG, Research Division Christine Chiou - Barclays Capital, Research Division Brian Skorney - Brean Murray, Carret & Co., LLC, Research Division Philip Nadeau - Cowen and Company, LLC, Research Division

Ladies and gentlemen, thank you for standing by, and welcome to Gilead Sciences First Quarter 2012 Earnings Conference Call. My name is Stacy, and I will be your conference operator for today. [Operator Instructions] As a reminder, this conference call is being recorded today, April 26, 2012. I would now like to turn the call over to Susan Hubbard, Vice President of Investor Relations. Please go ahead.

Thank you, Stacy, and good afternoon, everyone. Welcome to Gilead's First Quarter 2012 Earnings Conference Call. We issued a press release this afternoon providing earnings results for the quarter. This press release is available on our website as are the slides to provide more detail around some of the topics we will cover on today's call. The speakers for today will be John Milligan, President and Chief Operating Officer; Norbert Bischofberger, Executive Vice President of R&D and Chief Scientific Officer; Kevin Young, Executive Vice President of Commercial Operations; and Robin Washington, Senior Vice President and Chief Financial Officer. Robin will first discuss our financial results. Kevin will review our commercial performance, and Norbert will then provide an update on our research and development progress. John Milligan will wrap up our prepared remarks by discussing our Quad drug and our future outlook and opportunities. Our Chairman and CEO, John Martin will not be participating on today's call as he is currently out of the office. As a reminder, during today's call we will be making forward-looking statements regarding our financial outlook in addition to regulatory and product development plans. These risks include the possibility of unfavorable results from our clinical studies, including those evaluating GS-7977, GS 5885 and GS-9669 alone or in combination with other antivirals. The possibility that we may be unable to meet the anticipated timing for receiving clinical data and making regulatory filings, and the possibly that we unable to develop an all-oral antiviral regimen for HAV, genotype 1 patients or a pan-genotypic regimen for all HCV patients. These statements are subject to risks and uncertainties that may cause actual results to differ from those expressed in any forward-looking statements. A discussion of these risks can be found on our latest SEC disclosure documents and recent press releases. In addition, please note we undertake no duty to update or revise them. We also will be using non-GAAP financial measures to help you understand our underlying business performance. The GAAP reconciliations are provided in our press release, as well as on our corporate website. I will now turn the call over to Robin Washington.

Thank you, Susan, and thank you, all for joining us today. 2012 began with solid first quarter financial results. Our commercial execution continues, delivering product sales of $2.2 billion, an increase of 19% year-over-year and 4% sequentially despite a challenging macroeconomic environment. The U.S. contributed $1.3 billion to product sales, up 25% compared to $1 billion in the first quarter of last year. Europe contributed $764 million to product sales, up 7% year-over-year, driven primarily by growing underlying product demand. Non-GAAP product gross margin decreased to 74.5% from 75.7% in the same period of last year, primarily due to the higher cost of efavirenz and changes in product mix, as Atripla and Complera increased as a percentage of revenue while Truvada decreased. Non-GAAP R&D expenses for the quarter were $331 million compared to $237 million in the same period of last year, due to increased clinical activities, particularly in liver disease and expenses associated with the continued advancement of our product pipeline. Non-GAAP SG&A expenses for the quarter were $308 million, up from $263 million in the same period of last year, primarily due to an increase in expenses to support the ongoing growth of our business and an increase in the U.S. pharmaceutical excise tax. During the first quarter, we completed our $11.1 billion acquisition of Pharmasset, and have allocated the purchase price as follows: The balance sheet includes in-process research and development of $10.7 billion, goodwill of $75 million and net assets and liabilities assumed of $64 million. The GAAP income statement includes a $194 million compensation charge expense for acceleration of unvested stock options. This compensation expense also negatively impacted operating cash flow for the quarter. Turning to interest and other income and expense. During the quarter, we recorded a net loss of $40 million or $0.03…

Thank you, Robin. I'm very pleased with the solid start we have made to 2012. Our antiviral franchise generated product sales of $1.9 billion for the first quarter, up 18% over first quarter 2011 and up 4% sequentially. Performance in the U.S. was particularly robust, with Atripla and Truvada delivering year-on-year growth of 21% and 17%, respectively. We also had a very healthy U.S. contribution from Complera of nearly $50 million. I'll return to Complera in a moment. During this most recent quarter, there was no change in major U.S. wholesale inventory levels. They remained at the lower end of our contractual range. However, it was an exceedingly strong quarter for non-retail purchasing, the biggest quarter in absolute terms we have seen. There was clearly a demand-based component to our large ADAP purchases recorded in the first quarter. As of last week, ADAP wait lists are down by 67% from their peak in 2011, the current number being slightly in excess of 3,000 patients. Growth in total prescriptions, especially those for STRs, single tablet regimens, also remained very positive. All major ADAP states now have both Atripla and Complera on their drug formularies. Finally, the March 2012 update in DHHS guidelines further reinforced earlier treatment of HIV-infected individuals. However, over and above solid market growth characteristics, it is important to highlight that ADAP purchasing likely carried an inventory component, with states like Florida, Illinois and Virginia adopting a use-it-or-lose-it approach, recognizing that fiscal year 2011 federal budgets were only finalized in September of last year. Whilst it is very difficult to predict quarter-by-quarter ADAP purchasing, the strong first quarter performance may have a knock-on effect to our second quarter of 2012. The ADAP Federal budget for 2012 has now been confirmed at $955 million, an uplift of $50 million over…

Thank you, Kevin. Just last week, we had a significant scientific presence at the European Liver Disease Conference EASL. This meeting has grown considerably over the years. And this year, there were more than 9,000 health care professionals and scientists in attendance. More than 30 presentations covered Gilead programs and products. And new data were presented on GS-7977, our nucleotide analog for HCV infection; GS 5885, our NS5A inhibitor; and GS-9669, our novel NS5B inhibitor. Also presented were very important data on Viread for the treatment of hepatitis B. The very first presentation at the EASL conference reported on an arm in the ATOMIC study, which evaluated a regimen of GS-7977 with pegylated interferon ribavirin for 12 weeks in genotype 1 infected patients. This data indicated that this regimen resulted in 90% to 94% cure rates without the need for response-guided therapy algorithms. The addition of GS-7977 to pegylated interferon ribavirin did not result in additional side effects. And thus, this 12-week regimen could be an important advancement for the field. Data from the ELECTRON study were also presented, and in addition, we chose to include data from the QUANTUM study of last week's press release. Both data sets were from study arms evaluating GS-7977 with ribavirin for 12 weeks in genotype 1 treatment-naïve HCV-infected patients. The studies are ongoing and only SVR4 data were available at the time of the EASL Conference. The SVR4 rates in ELECTRON were 88% and in QUANTUM 59%. These are exciting results and indicate that it is possible the genotype 1 patient to be cured with a 12-week course of GS-7977 and ribavirin. Bristol-Myers Squibb presented data from their study, utilizing GS-7977 in combination with their NS5A inhibitor, daclatasvir, with or without ribavirin in genotype 1 infected patients. After 24 weeks of treatment, 100%…

Thank you, Norbert. I feel very pleased with the progress of Gilead's extensive portfolio of investigational agents, but I'd like to add a few additional comments regarding our HIV products. In early March, the Conference on Retroviruses and Opportunistic Infections, also known as CROI, took place in Seattle. Gilead had a significant presence with 45 abstracts focused on Gilead's marketed and pipeline products. Key oral presentations and posters included data on Quad, GS-7340, Viread for HPV and GS-7977 for HCV, as well as Truvada for the prevention of HIV or PrEP. Importantly, for the first time, we were able to present the data from studies 102 and 103, the 2 registrational trials of Quad. Data from Study 102 indicates that Quad was noninferior to Atripla based on the primary endpoint of percentage of patients below 50 copies per milliliter at week 48. Quad demonstrated advantages in neuropsychiatric side effects and in cholesterol levels. Likewise, Study 103 demonstrated that Quad was noninferior to a regimen containing ritonavir-boosted atazanavir and Truvada. In the second study, Quad had advantages over boosted -- over the boosted atazanavir regimen with regard to elevated bilirubin levels and triglyceride levels. In both studies, discontinuation rates due to adverse events were similar. The full results of these 2 studies have been submitted for publication in a peer-reviewed journal and the applications for marketing approval are under review in the United States, Europe, Australia, Canada and Switzerland. In the U.S., FDA will convene an antiviral division advisory committee panel to discuss the Quad application on May 11. As many of you know, the PDUFA date for Quad is August 27. Over the past few years, we collaborated with various organizations to explore the use of Truvada for PrEP. Several studies have shown that the use of Truvada in uninfected…

[Operator Instructions] Your first question comes from the line of Geoff Meacham with JPMorgan. Geoffrey C. Meacham - JP Morgan Chase & Co, Research Division: A couple for you on hep C. So the first one is, is there a new data set that you're waiting for as kind of a gating factor to start an all-oral Phase III for genotype 1? And then the second part of that is, what would a Phase II look like when you combine, say, 7977 and 5885? Would the bias here be for the fast simple study or more of a comprehensive study that looks at things like duration, different doses, population, things like that?

Geoff, I'll answer these 2 questions. So what would a phase -- what are we waiting for? We are not necessarily waiting for any particular data set, but we will have discussions very soon with regulatory authorities to make proposals to them. And at this point, we really don't know what regulatory authorities think. We need their feedback and input. But the 2 upcoming data sets that are both very interesting that will determine or could determine both the design and the duration of a Phase III study is number one, the 24-week data in QUANTUM with GS-7977 and ribavirin; and secondly, the 12-week data from the Bristol-Myers study with 7977 and daclatasvir. So both of those things really address one, if you extend the duration of the 7977 ribavirin, can you get higher response rate? Or secondly, if you shorten the Bristol-Myers daclatasvir, 7977, can you maintain the high response rate. Those are the 2 questions that are being answered. Then regarding Phase II, we are really thinking about something short, because we feel there are no concerns -- we don't have any outstanding concerns about safety of any of our products, 5885 or 7977, so we feel comfortable moving into a large Phase III study. And with regards to efficacy, all those questions really have been, for the most part, answered with the exception of the 2 that I mentioned before. And let me just say one last thing because it may come up. We also feel very comfortable extrapolating from daclatasvir to 5885 because these 2 compounds, by all criteria, are very similar at least in genotype 1. Geoffrey C. Meacham - JP Morgan Chase & Co, Research Division: Just as a quick follow-up, Norbert, would you be in a position to start several or multiple different strategies for Phase III for an all-oral? Or do you want to have a single strategy?

No. It would be -- absolutely, we would favor multiple because as you know, there is the issue of treatment duration. There's the issue of complexity. For instance, ATOMIC design, that's still on the table. That's actually a huge step forward from, if you think about where we are today in terms of standard-of-care to be able to shorten that to something, 12-week no response guided therapy with the well-tolerated 7977 added to peg/riba. That's certainly an option as well.

Your next question comes from the line of Mark Schoenebaum with ISI Group.

I was just wondering -- I have a question on the genotype 2/3 Phase III study versus standard-of-care. Is that a noninferiority trial or a superiority trial? And if it's noninferiority and -- sorry, and if possible, could you give us the criteria for noninferiority and superiority? And the second question is today, Bristol-Myers expressed in earnings call a desire to partner with Gilead. I was wondering if you could walk us through your updated thoughts on partnering with Bristol.

Mark, the first question is very simple. It's designed as a noninferiority study with a delta of 15%. But I would like to tell you that FDA has communicated to Pharmasset previously, so none of us were at those meetings, that even if they miss noninferiority, it would still be approvable because it's a huge advantage and a huge step forward to treat somebody for 3 months with the 2 oral drugs rather than 6 months with peg-interferon. But the answer is it's noninferiority with a delta of 15%.

Mark, it's John Milligan. I'm going to take the second part of that question. As Norbert mentioned, there are data sets that we're waiting for and that there's lots of options for us to move forward in multiple directions. But with regard to how or where or when we might work with Bristol-Myers, we're just not going to be able to comment on that strategy. I think it's very -- as we've learned in the past, it's very difficult to negotiate these things in public. And so I'm basically going to be held to no comment for today on this subject.

Your next question comes from the line of Rachel McMinn with Bank of America Merrill Lynch.

I have a question on the HPV programs, surprise, surprise. I guess my big question for you is why are you expanding ELECTRON specifically in no responders. You've got a 90% relapse rate with the 2 drugs. Now you're looking at this 2 different triple options with 5885 or 9669. Didn't we learn from EASL that the null population is particularly difficult and really setting a totally different bar relative to naïve? Why not go in and just do the naïve 12-week data, and then even throw in 9669 on top of 5885 to get the triple, more potent regimen instead of keeping riba in there?

So Rachel you have a very good point and the answer to that is very simple. We're doing both. We actually have a drug interaction study that is about to complete with 9669 and 7977. Once we get the data from that study, we will initiate a Phase II study in treatment-naïve patients genotype 1 to test that hypothesis. But Rachel, so the second point I would like to make, this is a small pilot experiment with 10 patients. So it's really not a big study, but it's an experiment. The third comment is we debated this internally. If you look at 7977 ribavirin in genotype 1 nonresponders, it gives you high relapse rates. But we felt we wanted to change only 1 variable and not 2. If you do a 2-variable experiment, both extend the treatment duration and add another drug, then in the end, you don't know. So we made the decision to add one more drug and see whether the results will be different. And lastly, let me also tell you from the large amount of data that we're getting, I think it's clear now, at least with 7977, which has a very high barrier to resistance, you can have a longer treatment duration and good responses or a shorter treatment duration, but then you have to add other drugs. So we think 12 weeks of 7977 with NS5A, with the 5885 and ribavirin, I am convinced we'll give you good treatment responses, but again we will find out once we get the data.

But just to clarify it looks like you're not actually doing that in naïve. You're not going to replicate the daclatasvir study, you're going to let Bristol's 12-week data in the fourth quarter kind of provide the proof of principle in naïve and you're going to explore the rest of this in nulls. So do I have that right?

That's right. Rachel, we -- as I said, we feel very comfortable scientifically to extrapolating from daclatasvir to 5885, at least in genotype 1 patients, not in 2 and 3 but in 1s. And we think there is no reason to believe that we wouldn't get the same results with 5885 as Bristol-Myers has achieved with daclatasvir.

Your next question comes from the line of Matt Roden with UBS.

It's a little bit of a broader question on hepatitis C. I understand that you can't talk about the specific path towards advancement until you speak with the FDA, but can you talk a little bit about, as many options as you can that could potentially take it forward and what the rationale would be for taking 7977, 5885 combo directly into Phase III without a Phase II? And do you have any indication for whether or not the FDA has a position on that at this point?

Yes. Matt, no. So I do not -- we don't have any indication what the FDA is thinking about that. So I'm, of course, speculating. But if we did this, it wouldn't be jumping with both feet into a large Phase III study but rather something like you enroll a cohort, you do a certain interim analysis after a certain time point and we still have to figure out how that -- what that time point is, convince yourself that the data looks good, and then you involve the rest of the cohort. So it would be like a Phase II, III approach, and that's something, which we'd have to think through carefully and propose to the FDA.

Your next question comes from the line of Yaron Werber with Citi.

I have 2 questions. First one, just maybe for Norbert just a little bit. Do you have any sense what happened to QUANTUM? Why is the data so bizarre and different than ELECTRON? And then secondly, with respect to -- when we spoke at EASL, it sounds like one of the things you were thinking about in Phase III is potentially doing a doubling of 7977, 5885 arm for 24 weeks. So my question is do you need to do some kind of a small Phase II, where you actually test them together for 24 weeks? Or is the FDA going to be comfortable extrapolating that data separately, from each drug separately?

So the answer to the second question, I said before we really have to have a debate with the FDA, and we haven't done that yet. I do not think -- I do not know what they think about it. And as I said, we wouldn't -- certainly the proposal would not be jumping to a 500-patient Phase III study with both feet but rather do something like a cohort first, and then you do an interim analysis and then you enroll the rest of the patients. So it would be something like more a Phase II/III study. And regarding your second question, the difference between QUANTUM and ELECTRON, the short answer is I don't know. But the 2 comments I would like to make, there was a difference in baseline CC genotype between the 2 study. As you may have seen in ELECTRON, there were 44% of the patients were CCs. And in QUANTUM, if I remember correctly, 16%. So the QUANTUM kind of underrepresented statistically what you would expect in the general population and ELECTRON overrepresented. That may be one reason why the ELECTRON patient population responded better. The other thing is, keep in mind these are still small numbers. It was 17 in QUANTUM and 25 in ELECTRON -- or the other way, yes. And the other thing, please, the CC genotype, I think, it was clear at EASL, there were 2 presentations, one by BI and one by BMS, where it was clear that CC IL28 genotype does matter even with all oral antiviral. So it's entirely possible that, that explains the discrepancy between these 2 ELECTRON and PROTON results.

But you would think that the CC wouldn't matter depending on the potency, right? I mean with 7977, we've seen with daclatasvir, it doesn't really matter. So when you get to pretty high potency, at that point, it doesn't matter.

That's true. You can then argue, if you have a high enough potent regimen, you don't -- influence of IL28 baseline genotype disappears. And again we will find that out along the way, if we add either 5885 or 9669 to 7977. Then if that influence -- IL28 disappears, then you're right.

Your next question comes from the line of Sapna Srivastava with Goldman Sachs.

I have 2 quick questions. One is just how do you get the comfort with your NS5A on the efficacy side? And I understand your rationale on the safety side. There's just a 800-patient dataset, but please help us understand on the efficacy side. And second just a quick clarification on the potential ADAP inventory levels. How should we be thinking how much of the strength this quarter could potentially be knocked off, as you said, in 2Q? Just a forward thinking for 2Q estimates.

Yes. Sapna, I will take the first question. So the efficacy side on the NS5A, we look at both modeling, so you look at in vitro antiviral potency against genotypes and within genotype and polymorphs, and our compound is very, very similar to daclatasvir in genotype 1. Then you can do the -- at the reduced dose, the 60 milligram daclatasvir versus the 90 milligram of 5885, you do the protein-adjusted IC50 or the IQ, and those IQs are very similar. And lastly, you look at the existing data that we have from 5885 both on monotherapy, and when you add it to peg-interferon ribavirin, you look at the steepness of the viral load decline, and it's again very similar. So we feel very comfortable that 5885, from an efficacy point of view, will mirror daclatasvir. It will be a good compound.

Sapna, this is Kevin. Just let me quickly talk about ADAP funding. So it was a big quarter for the end of the 2011 financial year for ADAP. That was particularly Florida and Illinois, less so places like Texas and Puerto Rico. The main products that they have purchased in is Atripla and Truvada because, of course, they are the workhorses of their budgets. Difficult to estimate what that will do to quarter 2. We're very pleased that the 2012 budgets have gone out already. You can find those on the HRSA websites. So basically everything bar the $35 million that was announced by President Obama on World AIDS Day have gone out, so that's a great reassurance. Remember last year, they didn't know the actual numbers until September, so that did lead to the big fourth quarter and quarter 1 of this year. So I think that provides a lot of stability. Clearly, Complera is starting to pick up, which is great for us, and I think experience has been very good in the field with Complera. So it was one of the biggest that we have seen in absolute terms, not in percentage terms but in absolute terms, and I think we'll just have to see how that plays out for Q2, Q3.

Your next question comes from the line of Robyn Karnauskas with Deutsche Bank.

May I just ask an HIV question, if I don't bore anyone? But your European market looked a little weak, a little flat quarter-over-quarter as far as patient adds. I was wondering if can give some color on the trends you're seeing in Europe and whether or not you think that you could see growth reinvigorate in the latter half of the year.

Robyn, it's Kevin. Yes, I think I'd agree with you. For the past sort of quarter, it has been flat, but I actually took that in a very positive sense. Europe is not easy, particularly Southern Europe, and I think we continue to do very well. If you look at quarter-on-quarter, there was about a 2% growth in demand. We got a positive 1% in FX and about a minus 1% in price. So it wasn't about quarter. We're starting to roll out Eviplera now to the big markets. The U.K. and Germany have got the product. And the initial response, I think, has been pretty positive. So I think on the back of Eviplera, we hope that we're going to have another burst and another push around single-tablet regimens. I should also add, Robin, guidelines continue to start to highlight single-tablet regimens and fixed-dose combinations. In our slide on guidelines that's in the earnings deck, you may have noticed the new draft BHIVA guidelines. I'll point out 2 things on those guidelines. Number one, the preferred regimens are all Truvada, so the alternative regimen is now abacavir, so it's been downgraded. There's a call out there highlighting the preference for fixed-dose combinations, avoiding selectivity of the individual agents. So again, the preference and the benefit of single tablets is coming through more and more. So I think you can call it a solid quarter in Europe, and we're certainly looking forward to Eviplera being a driver of growth in the second half of the year.

But if the patient numbers are increasing and market share has gone down, I know it's a lag, why is that? Do you have a sense of why that is?

Well, again, I wouldn't look too much into sort of 1% here, 1% there, Robyn. I'm literally looking at my market share slide for the big 5 markets, and it hasn't gone down, but it's basically flat when you look at total Truvada split up by Atripla and Truvada itself. So again, extrapolating from some of the numbers is quite hard sometimes. Please bear in mind that the European databases are just not quite as good as the U.S.

Your next question comes from the line of Thomas Wei with Jefferies. Thomas Wei - Jefferies & Company, Inc., Research Division: I had to have 3 questions. I didn't get a chance to talk to you all after the poster on 5885, and the resistance profile was presented over the weekend. How should we interpret the data in genotype 1b where it looked like there was about a thousand-fold shift with 5885 with the Y93H mutation, but only a 24-fold shift with the Bristol drug? Your comments on the ability to extrapolate the Bristol combination data to 5885 suggests that you don't think that, that's actually that big of a deal, and we'd love to understand what other factors we should keep in mind in interpreting that resistance data.

Yes. Thomas, that was the 1 polymorph that showed actually the biggest difference, and all I can say that polymorph is extremely rare and has not been described in the literature as reoccurring. It's one that we generated ourselves in the lab to just see how far can we push the system. So whether that polymorph is at -- in any way exists or is relevant clinically is an open question. The other thing you also should keep in mind that even with that shift at the 90 milligram dose, we're still way above a multiple of the IQ. So if you at least believe what we learned from HIV, this would add -- even against that polymorph would be very inhibitory compound at the 90 milligram dose.

Your next question comes from the line of Michael Yee with RBC Capital Markets.

I just wanted a clarification on this Phase IIb combination study that you're going to run as a extension arm of ELECTRON. Can you again clarify why you're running at nonresponders? Isn't that risky? Can you just run on multiple arms? And then as a follow-up, can you co-formulate 7977 and 5885 or anything about that? I assume that's an easy hurdle, but just asking.

Michael, the answer to the second question, that's an effort that's ongoing, and we have prototype pills internally. Of course they have to go through the usual stability, analytical method and then PK evaluation at least in humans. But that seems to be fairly straightforward. These compounds, from a physicochemical point of view, are both well behaved. So the reason why we're doing the null responders is simply -- so it's not that we're only doing the null responders, we're thinking about treatment-naïve patients more in a Phase II, III sense. We are more certain that the combination data works because of the experience with the Bristol-Myers data, and what we don't know and why we waited, we would like to find out in the genotype, in the null responders. Remember, in the 12-week study, 7977 plus ribavirin null responders, we had 9 out of 10 rebounders but there were no breakthroughs. So virologically, this is a very, very potent regimen, but maybe we just didn't have enough pressure on the system to cure a significant number of patients in 12 weeks. So that's why we decided to add another drug -- repeat the same experiment that we have done before, just adding another drug. So it's a one-variable experiment that would give us some information at the end. That was the decision.

Your next question comes from the line of Ravi Mehrotra with Crédit Suisse. Lee Kalowski - Crédit Suisse AG, Research Division: Actually this is Lee Kalowski filling in for Ravi. One question I have is do you have the breakout for Complera in terms of naïve versus switching? And then in terms of Quad, how important are those switching studies going to be in terms of the uptake? And until that data is ready, I guess, where would you expect the initial uptake to go?

It's Kevin here. Yes. So in terms of the breakout, yes, I've got that data for you. So 35% of the Complera patients are coming from naïve, so basically new patients. And by the way, that's our label today in the U.S. that we stay on with our field force. 40% are switches from products other than Atripla, and that's largely the protease inhibitor combinations and 25% are switches from Atripla itself. Just to paint a little bit of color for you in terms of patients, the vast majority of patients, more than 90% of the patients are going on Complera, are those patients below 100,000 copies, again, as per our label. And there's actually quite a large percentage as we expected with our female patients, approximately 30% of the starts on Complera are female. That compares with only 13% as a comparative for Atripla. So a high, high proportion relative to Atripla for the female patient because of the better pregnancy category. In terms of Quad, just like Complera, we will have -- we expect to have a naïve label that won't be switching there. So the studies that were highlighted today are very important to us. We'd like to get that data into the label. Switching patients is still an important bucket, as you can see from Complera. So yes, we would want that data but we'll only initially be able to go out seeking the naïve patients. But I do want to emphasize that the nice thing about Quad, and it will be our first detail with our expanded field force, is that we've got that larger slice of the naïve pie because we've got the data 102 versus Atripla and 103 versus the boosted proteases. So as I've said in the past, we have the opportunity to address the 9 out of 10 patients that are beginning in the naïve setting. So it gives us a much, much broader expanded horizon than we've got today with Complera.

Your next question comes from the line of Tony Butler with Barclays.

This is Christine Chiou in for Tony. The first is your QUANTUM study still has 24-week treatment arm stuff to read out, and let's suppose that 7977 and ribavirin for 24 weeks do lead to a competitive STR, do you think this combo is still relevant given what took place at EASL? And would you really still develop a 24-week combo for GT1s? And then second, we're beginning to see the growth momentum for Incivek slowing down, and what do you think this says about the HCV market?

Yes. So regarding 7977 ribavirin for 24 weeks, yes, I do believe that's a valid regimen because it's one drug and cheap generic ribavirin that many physicians tell me they are comfortable with and not unhappy with. And then, of course, you have to balance, is that a better regimen than something more drugs, more complicated, maybe more expensive for 12 weeks. And obviously we're going to do both and see where the clinical data lead us. And what was the second question?

I'll take that one, Norbert. The question was about Incivek sales going down and what does that say about the market. I think it says more about the profile of the protease inhibitor containing regimens of peg/riba where it's very clear that while the overall response rate has gone up, the overall complexity of the regimen and the side effect profile was much more severe especially in severely ill patients than people had previously thought. And there was numerous data sets at EASL that outlined the cost and complexity of keeping people on these protease inhibitor-containing regimens, including the need for transfusions and the very poor side effect profile in certain patients. And so what it suggests to us is that now the patients -- doctors and patients know that easier regimens are on the horizon, and they're going to continue to warehouse those patients until those first regimens are on the market. And they also know that they're coming very soon, so that's exciting for them as well.

Your next question comes from the line of Brian Skorney with Brean Murray. Brian Skorney - Brean Murray, Carret & Co., LLC, Research Division: I just wanted -- maybe I can get some top line kind of commentary on the hep C landscape from you guys and where you might see yourselves in the competitive space. I mean, obviously, I think we can all agree it's going to be a very, very meaningful driver to be kind of the first mover to have an all-oral. I'm just wondering what you think of if a competitor were able to navigate a pan-genotypic oral regimen to market between the approval of genotype-specific regimens anchored by 7977 but before you could get your own Gilead, own pan-genotypic regimen? What sort of impact do you think this could have commercially? Do think this is a big deal to be the first mover here? Or do you think this is less of a competitive advantage?

Brian, it's Kevin. I'll kick off. I think it's a very big deal to be a first mover, particularly with 7977. I think the results that we saw in Barcelona, and everybody around this table were there in Barcelona, says that we have a very, very potent agent. So a good drug, being first is to me a double whammy, and so we look forward to hopefully that opportunity. We have a lot of expertise and extremely good relationships in this setting. We have field forces here in the U.S. and in Europe who visit the liver disease specialists. We'll have to expand those activities, and we'll have to balance our HPV responsibility versus HCV, but we really do have very, very good relationships. And that overlap is even bigger in Europe. So we are already starting to do the work on field force sizing. I can certainly reassure you that we'll have a incredibly committed presence in hepatitis C. And, of course, as the other dimension for Gilead is, are there additional opportunities with the infectious disease doctor? So we're planning to be ready along Norbert's timelines for the genotype 2 and 3, and then hopefully not too further down the road, we can get the genotype 1. So yes, first-move advantage is important, and we look forward to that opportunity. Brian Skorney - Brean Murray, Carret & Co., LLC, Research Division: I guess, maybe be a little more specific. I'm talking what you would think about the impact of being first mover with a pan-genotypic regimen. I guess, I'm kind of looking at Bristol and saying if you're extrapolating the daclatasvir data to 5885, maybe it's also reasonable to extrapolate some of the 7977 data to at least the 200 meg dose of INX-189. It would seem that Bristol has a viable pan-genotypic combo, at least moving into clinical development very soon. I'm just wondering if they would sort of take the lead in a pan-genotypic. Do you see that as a major competitive risk?

Well, I think the 2 and 3 will give us a very, very nice entry. I'm hoping that Norbert and the team here will get us into 1 very, very quickly, and ultimately with our own pan-genotypic regimen. So time will tell. I think it's very difficult to make predictions on that sort of race, if you like, between Bristol-Myers and Gilead. We think we've got the best backbone here, and we think we've got alternative approaches, as you've heard from Norbert. So I think we're going to be highly competitive in that race to pan-genotypic.

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

A question on the upcoming Quad panel. In the past you suggested that the reason why the FDA probably wanted to review the Quad was because of the cobicistat kind of component. In other words, just a couple of weeks away from the panel, I'm guessing Gilead has reviewed the briefing documents that the FDA has prepared. Could you give us some more information on what issues, if any, the FDA has with the Quad and whether we should expect any surprises when we see those briefing documents in a couple of weeks?

Well, Phil, obviously those documents are not public and won't be released to the public until a couple of days ahead of time. So we just can't comment on those.

Phil, I think I said last time that the one issue that we always felt was something that warrants discussion has to do with that cobicistat increases creatinines, and it generates some question to the practicing physician when is the creatinine elevation just simply a laboratory thing that's caused by cobicistat, and when is it something that you have to take serious as an issue of nephrotoxicity.

So what could come out of the panel on that topic? Would it be a discussion of labeling? Or I guess what kind of conclusion could the panel draw?

Yes, that's what I think the discussion would be about, what should the physician do in terms of -- normally you use serum creatinine for measure of kidney function. Now that we have cobicistat, that increases creatinine, how do you interpret serum creatinine levels? That's what -- I think -- well I don't know what it's going to be but that's what I guess is going to be the nature of the discussion. And the potential outcome could be a labeling or monitoring recommendations or something like that.

And at this point, Ms. Hubbard, we have run out of time for additional question.

Thank you, Stacy, and thank you everybody very much for joining us today. We appreciate your interest in Gilead as always, and we'll be here and ready to take your calls after we conclude. Thanks, everybody. Goodbye.

We thank you for your participation in today's conference. This does conclude your presentation. You may now disconnect, and have a great day.