Oct 18, 2006
Executives
John Milligan - Executive Vice President, Chief Financial Officer John Martin - President, Chief Executive Officer, Director Norbert Bischofberger - Executive Vice President of Research and Development Kevin Young - Executive Vice President, Commercial Operations
Analysts
Meg Malloy - Goldman Sachs Geoffrey Porges - Sanford Bernstein Thomas Wei - Piper Jaffray Yaron Werber – Citigroup Bret Holley - CIBC World Markets Geoff Meacham – JPMorgan Ian Somaiya - Thomas Weisel Partners Mike King - Rodman & Renshaw Michael Aberman - Credit Suisse Sapna Srivastava - Morgan Stanley George Farmer - Wachovia Securities David Amsellem - Friedman, Billings, Ramsey Phil Nadeau - Cowen & Co
Operator
Ladies and gentlemen, thank you for standing by. Welcome to the Gilead Sciences Third Quarter 2006 Earnings Conference Call.
At this time, all participants are in listen-only mode. Later we will conduct a question and answer session.
[Operator Instructions]. As a reminder, this conference call is being recorded today, Wednesday, October 18, 2006.
Your speakers for the day are John Milligan, Executive Vice President and Chief Financial Officer; John Martin, President and Chief Executive Officer; Norbert Bischofberger, Executive Vice President of Research and Development, and Kevin Young, Executive Vice President of Commercial Operations. I would now like to turn the call over to Dr.
Milligan. Please go ahead.
John Milligan - Executive Vice President, Chief Financial Officer
Good afternoon and welcome to Gilead’s Third Quarter 2006 Earnings Conference Call. We issued a press release this afternoon providing results for the third quarter ended September 30th, 2006, describing the Company’s quarterly highlights.
The press release is also available on our website. Also joining us on today’s call are Matt Loar, Vice President of Finance; Mark Perry, Senior Business Advisor; and Susan Hubbard, Vice President of Investor Relations.
I will begin the call by reviewing the third quarter financial results and then I will provide updated financial guidance for 2006. John Martin, Norbert Bischofberger and Kevin Young will take you through the corporate and product-related highlights for the quarter.
We will allow time at the end of this call to answer your questions. First, let me start with the standard safe harbor statement.
I would like to remind you that we will be making statements relating to future events, expectations, trends, objectives, financial results that constitute forward-looking statements within the meaning of the Private Securities Act of 1995. These statements are based on certain assumptions and are subject to a number of risks and uncertainties that could cause our actual results to differ materially from those expressed in any forward-looking statement.
I refer you to our Form 10-K and Form 10-Q for the first and second quarters of 2006, subsequent press releases and other publicly filed SEC disclosure documents for a detailed description of the risk factors affecting our business. In addition, please note that we undertake no obligation to update or revise these forward-looking statements.
In addition, to the extent we discuss our proposed acquisition of Myogen during this call, such discussions are neither an offer to purchase nor an offer to sell any securities of Myogen. A tender offer for Myogen’s outstanding shares is being made only by the offer to purchase filed with the Securities and Exchange Commission on October 16th, 2006.
Now, turning to the specific results for the third quarter, in short, the third quarter of 2006 was a very solid quarter for Gilead, with record-setting net product sales and total revenue. Total revenues were up 52% compared to the same quarter last year, driven primarily by the strong growth of our HIV product sales.
Gilead reported a net loss of $52 million or $0.11 per share on a fully diluted basis for the three months ended September 30th, 2006. This includes $26 million of after-tax stock-based compensation expense from our adoption of FAS 123(R) on January 1st of 2006 as well as the $356 million in-process R&D charge related to our acquisition of Corus Pharma.
Excluding the impact of the in-process R&D charge, but including stock-based compensation expenses our earnings per share would have been $0.64 per diluted share, representing a 68% increase when compared to $0.38 per diluted share recorded in the same period last year. A reconciliation of these non-GAAP measures to applicable GAAP measures is provided in our press release issued earlier today.
Now turning to revenues. Total revenues for the third quarter of 2006 were $749 million, an increase of 52% from total revenues of $493 million in the third quarter of 2005.
This growth was primarily driven by higher HIV product sales and higher royalty revenues from Roche’s second quarter Tamiflu sales. Revenue from Gilead’s product sales were $670 million, an increase of 43% from the third quarter of this year compared to the same period of last year, driven primarily by our strong HIV product sales.
Product sales increased sequentially by 13% compared to the second quarter of this year. HIV product sales grew to $557 million for the third quarter of 2006, up 53% compared to $364 million in the third quarter of 2005 and up 17% sequentially from the second quarter of 2006.
Truvada sales were $309 million for the third quarter of 2006, almost double Truvada sales for the same period last year and an increase of 3% sequentially. Truvada sales accounted for more than half of our total HIV franchise sales in the third quarter of 2006.
In the United States, Truvada sales were 201 million in the third quarter of 2006, an increase of 44% compared to the third quarter of 2005, but a decrease of 3% sequentially, likely due to patients switching from a Truvada containing regimen to one containing Atripla. Viread sales in the United States for the third quarter of 2006 decreased by 4% compared both to the same period last year and sequentially.
In its first quarter on the market, Atripla contributed $68 million to our HIV franchise sales. As a reminder, Gilead records 100% of the Atripla revenue, since Gilead consolidates the financial results of our joint venture with BMS.
The economic value of Sustiva, which is distributed back to BMS, is captured in our cost of goods sold line. The Sustiva component of ATRIPLA sales comprised approximately $25 million of the $68 million, with Truvada comprising the remainder.
In Europe, Truvada sales in the third quarter of 2006 increased sequentially by 60%, while Viread sales decreased sequentially by 10%. We continue to see the impact of the European launch of Truvada on Viread and expect, as we saw in the United States, declining sales of Viread as patients switch from a Viread containing regimen to one containing Truvada.
However, in countries where Truvada is either early in its launch or has not yet launched Viread sales remain strong. As for the United States, Viread should stabilize because it will continue to be a critical component of HIV therapy in Europe.
In the U.S., we also see the impact of Atripla launch on Truvada and expect Truvada sales to begin to level off but remain strong because Truvada is now established as the NRTI backbone of choice with protease inhibitors. Hepsera for the treatment of chronic hepatitis B had sales of $55 million in the third quarter of 2006, an 18% increase compared to the third quarter of 2005, driven primarily by sales volume growth in our European markets.
Hepsera sales decreased slightly by 3% from the second quarter of 2006 due primarily to a reclassification of a pharmaceutical tax imposed by the French government from expenses to contra revenue. Finally, sales of AmBisome were $55 million for the third quarter of 2006, an increase of 1% over the same period of 2005.
Compared to the same period last year, revenue from Gilead’s royalty and contract revenues for the third quarter of 2006 nearly tripled. This was due primarily to increased Tamiflu royalty revenues recognized from higher Tamiflu sales made by Roche, as well as the elimination of the cost of goods adjustment, which had historically reduced the amount of Tamiflu royalties paid to Gilead.
The royalty payment received from Roche was $63 million. Total royalty and contract revenues decreased sequentially by $16 million due primarily to the seasonality of the flu and the resulting impact on Roche’s Tamiflu sales.
As you may know, Roche released our third quarter 2006 earnings yesterday and reported quarterly Tamiflu sales of CHF669 million or approximately $529 million based on current currency exchange rates. Gilead will record royalty revenues from these sales after certain adjustments at 22% during the fourth quarter of this year, since royalties are paid one quarter in arrears.
In Roche’s earnings call, their management increased their (indiscernible) guidance from a range of CHF1.2 to CHF1.3 billion to a range of CHF1.6 to CHF1.7 billion and reconfirmed that by year-end, they will have capacity to provide 400 million treatment courses annually. Finally, we received a $5 million milestone payment from GSK, marking four consecutive quarters where Hepsera sales exceeded $75 million from the GSK territory.
This $5 million is recorded as deferred revenue and will be amortized into contract revenues over the life of our manufacturing supply agreement with GSK. Turning to gross margins.
Product gross margins for the third quarter of 2006 was approximately 84% compared to product gross margin of approximately 86% for the same quarter of 2005. The lower gross margin is due primarily to the launch of Atripla, which carries the Sustiva related portion of revenue at a zero gross margin, as well as product mix exchanges and the inclusion of stock-based compensation expense in 2006, partially offset by the positive impact of the royalty buyout from Emory related to Emtricitabine sales.
Now turning to expenses, research and development expenses were $93 million for the third quarter of 2006, which included stock-based compensation expense of $13 million. This is an increase of 18% from $79 million in the same period of last year and a slight increase of 3% from $91 million sequentially.
In addition to stock-based compensation expense, other factors which led to an increase in R&D expenses for the third quarter of 2006 were increased headcount, increased clinical and product development expenses and research and development activities, and the inclusion of operating expenses from our new Seattle operation. SG&A expenses in the third quarter of 2006 were $133 million, which included stock-based compensation expense of $16 million.
This is an increase of 31% from $101 million in the same quarter of 2005 and a decrease of 30% from $152 million sequentially. Factors impacting the higher SG&A expenses in the third quarter of 2006 compared to the third quarter of 2005 were primarily increased headcount and expenses driven by our significant business growth and business development activities.
The lower SG&A expenses in the third quarter of 2006 compared to the second quarter of 2006 were primarily due to timing of various expenses and initiatives such as stock-based compensation expense, the Atripla launch and the reclassification of the French tax associated with Hepsera mentioned earlier. Our tax rate for the third quarter of 2006 before the purchased in-process R&D expense, was 31.8%, a slight change from 32% tax rate for 2005, excluding the benefit of the foreign earnings repatriation in the fourth quarter of 2005.
No tax benefit was recorded for the purchased in-process R&D expense. Finally, I would like to turn to the cash-flow statement and balance sheet to highlight our cash-flow performance for the quarter.
The balance sheet at September 30th, 2006 shows cash, cash equivalents and marketable securities of $3.2 billion. This is a decrease of $95 million when compared to the balance sheet of $3.3 billion at June 30th, 2006.
The decrease during the quarter is primarily due to $331 million of cash paid on our acquisition of the net assets of Corus and $50 million paid towards our principal on our term loan, partially offset by $251 million of operating cash flow generated during the quarter. As announced on October 2006, we have signed a definitive agreement to acquire Myogen, Inc., subject to certain closing conditions.
The cash value of this transaction is approximately $2.5 billion. We intend to use our existing cash to complete this transaction, which we expect to close in the fourth quarter of 2006.
Now I’d like to turn to our financial guidance for the full year 2006. I will only be providing updates on specific guidance components that have changed since we first released 2006 guidance in January of this year and our subsequent updates of April and July of this year.
You can locate all of our guidance for the year 2006 on Gilead’s corporate website. For our entire HIV franchise, which includes Viread, Emtriva, Truvada and now Atripla, we’re slightly raising our full-year 2006 guidance for net product sales for the HIV franchise from a range of $1.95 billion to $2 billion to a range of $2 billion to $2.05 billion.
Turning to expense guidance, as a reminder the remaining guidance on expenses includes the impact of stock-based compensation expensing resulting for our adoption of FAS 123(R). We’re raising our R&D guidance from a range of $345 million to $370 million to a range of $365 million to $385 million.
We are also tightening our SG&A guidance from a range of $550 million to $580 million to a range of $555 million to $575 million. Our updated R&D and SG&A expense guidance reflecting anticipated operating expenses associated with both the Corus and Raylo acquisitions but do not reflect expenses associated with potential acquisition of Myogen.
We will provide updated guidance on our fourth-quarter and year-end conference call in January. At this point, I would like to turn the call over to John Martin, who will review our corporate highlights for the third quarter of 2006.
John?
John Martin - President, Chief Executive Officer, Director
Thank you, John. Good afternoon everyone, and thank you for joining us today.
We’re pleased to summarize for you Gilead’s many accomplishments here in the third quarter of 2006. I will start by providing an update on our business, then Norbert Bischofberger will summarize our pipeline efforts and research programs and Kevin Young will review our commercial efforts.
To begin, I’d like to comment on our ongoing efforts to expand our research, development and commercial focus, particularly beyond antivirals and into specialty fields of pulmonology and respiratory disease. The first of them was the acquisition of Seattle-based Corus Pharmaceuticals, which we completed in August.
This acquisition brings to Gilead a team of dedicated individuals with significant expertise in the research and development of products for respiratory diseases and a pipeline of important drug candidates, including Aztreonam lysine for cystic fibrosis related lung infections. Integration of Corus’s Seattle-based employees is progressing as planned.
As you know, Aztreonam lysine is currently being evaluated in two Phase III trials with the goal of defining the safety and efficacy of the product as an inhaled treatment for cystic fibrosis patients with pseudomonas infection. We expect data from both of the Phase III trials in the first quarter of next year, which if favorable, would allow a U.S.
NDA filing sometime in the second half of 2007. To further expand our pulmonology franchise, we recently announced plans to acquire Colorado-based Myogen.
Based on the data from the two Phase III trials, we believe that Emtricitabine, Myogen’s lead product for pulmonary arterial hypertension, has the potential to be a best-in-class endothelin receptor antagonist. We look forward to welcoming the employees of Myogen and supporting their efforts to file an NDA in the U.S.
before the end of this year. As you may have seen, we launched the tender offer earlier this week and believe that we will be able to close prior to the end of 2006.
This planned acquisition marks the third this year and we’re already working hard to ensure that the companies integrate smoothly and efficiently. We’re confident of our ability to swiftly combine operations and stay on track for NDA filings and subsequent launches of ambrisentan in 2007 and Aztreonam lysine in 2008.
These new products will provide us with a solid foundation for both near-term and long-term revenue growth. We will provide more detailed information related to the impact of the Myogen transaction on our future financial results once we’ve completed the transaction.
Kevin will discuss in greater detail our commercial successes with Atripla later in the call and I wanted to highlight a few key points regarding the progress we have made so far. The FDA approval of Atripla in July of this year marks the introduction of the first and only complete, highly active antiretroviral therapy combination now available in the U.S.
as a once daily single tablet. The U.S.
launch of Atripla only two days after approval was a critical first step in making it available to patients globally. While we have not yet finalized an agreement with Merck and Bristol-Myers for Europe, sufficient progress has been made on that front to allow for the filing of the marketing authorization application for Atripla in the European Union last week.
A review of the MAA will be conducted by the EMEA under the centralized licensing procedure, which, when finalized, will provide one marketing operation authorization in all member states of the European Union. In August, we finalized an agreement with Merck to distribute Atripla to resource-limited countries, where the epidemic has hit hardest.
As you will recall, the U.S. FDA approval of Atripla included the approval of two trade dresses, one for the U.S., and an alternative trade dress, which will help facilitate product distribution and tracking in the developing countries around the world.
Gilead has begun manufacturing Atripla using a fabric supplied by Merck and will initiate shipments of the drug to Merck shortly. Merck in turn will handle distribution of the product in the countries covered by the agreement, as well as regulatory submissions to achieve registrations.
Also last month, we announced we have completed an agreement with Bristol-Myers for Canada and we have now completed the Atripla regulatory filing in that country. And finally moving on to our access program, we continue to work vigilantly to help provide access to HIV medications in resource-limited parts of the world.
Gilead recently signed non-exclusive license agreements with the 11 Indian generic manufactures. These license agreements bring these companies their rights to produce and distribute generic versions of Viread to 95 low-income countries around the world, including India, which has the largest number of people in the world living with HIV.
I will now turn the call over to Norbert to review our pipeline and research efforts. Norbert?
Norbert Bischofberger - Executive Vice President of Research and Development
Thank you, John and good afternoon, everyone. To begin with, I would like to give an update on the Phase II dose ranging study of GS9137, which is our lead Integrase Inhibitor candidate for the treatment of HIV.
This study was designed to evaluate three doses of 9137 and those are 20, 50 and 125 milligrams, each boosted with 100 milligrams of Retinovir, along with a background regimen which could not include protease inhibitors. Recently, Gilead’s independent data safety monitoring board concluded a review of the eight-week data from the study and determined that the current data suggests suboptimal efficacy in the 20 milligram arm compared to the other arms, and recommended that patients under 20 milligram arm be enrolled to the high-dose group, which is the 125 milligram dose.
In addition, the DSMB recommended that all patients be given the option of adding one of the two protease inhibitors, either Tipranovir or Darunavir to their regimen. This recommendation is based on recently obtained data indicating that there is no drug interaction between 9137 and Tipranovir or Daronavir.
When this proposed change, the design of our Phase II study is similar to the Merck 518 Integrase Inhibitor Phase II study, where the concurrent use of protease inhibitors was allowed. Our Phase II study will continue as planned and all patients will have crossed the 24-week time point in late December.
Subsequently, Gilead will meet with the FDA to discuss the data from this study and to help determine the trial design for our Phase III clinical program. We have also identified a second Integrase Inhibitor, GS-9224 for which an IND will be filed in the next couple of days.
We expect the Phase I study to begin before the end of this year. Based on preclinical data, GS-9225 should not require Ritonavir boosting, and it also appears to have the potential to be dosed once daily.
For Tenofovir DF for chronic hepatitis B, we have completed enrollment in our two pivotal Phase III studies with a 48-week end point, we anticipate data from this study some time late next year, with the opportunity for regulatory filings, both in U.S. and the European Union occurring in the first half of 2008.
For the treatment of hepatitis C, Achillion expects to begin dosing patients in a Phase I/II viral dynamic study of GS-9132 prior to the end of the year. The dose escalation trial will serve as a proof of concept study, determining the anti-viral potency and safety of GS-9132 in HCV infected patients.
And finally, we have identified a novel non-nucleoside HCV Polymerase Inhibitor, GS-9190, for which we have recently filed an IND. This compound has completed one month animal toxicology studies, it has low nanomolar potency for both HCV genotypes 1a and 1b and the replicon assay system, and it also has a good pharmacokinetic profile.
We have received FDA concurrence, and we anticipate beginning the Phase I study in HCV infected individuals this quarter and expect to have preliminary safety and efficacy data from this study some time in the early part of 2007. In summary, I’m very pleased with the research and development advancements we’ve achieved this quarter.
And with that, I will now turn the call over to Kevin Young to discuss our commercial efforts. Kevin?
Kevin Young - Executive Vice President,Commercial Operations
Thank you, Norbert. Good afternoon everyone.
To begin, I’d like to discuss the recent activities around Atripla launch. In July of this year Gilead and Bristol-Myers received approval from the U.S.
Food and Drug Administration for Atripla dose one pill once daily regimen for the treatment of HIV. I’m pleased to report that we have seen very rapid uptake of Atripla since the launch only three months ago.
Launch trajectory of Atripla, when looking at prescription data, has surpassed that of Truvada, which was the number one antiretroviral launch in HIV history in terms of market share gain. In its thirteenth week on the market, Atripla’s share of the new prescription and total prescription market was 10.6% and 6.4%, respectively, as compared to 8.5% and 4.8% for Truvada at the same time point.
The Atripla launch has been supported by a successful series of activities focused on former acceptance with private payors, Medicaid, VA and AIDS drug assistance programs. In record time since its launch, Atripla has now been added to all state Medicaid formularies and ADAP formularies, including Washington, D.C.
and Puerto Rico. And with most payors, patients have the additional benefit of only one copay for the three medications in Atripla.
Since we receive third-party patient data on a one-quarter lag, it is too early for us to pinpoint where the Atripla uptake is coming from. But we can say that based on weekly prescription data received from Wolters Kluwer Health, Combivir’s new prescription market share has dropped two market share points since the launch.
In addition, the ratio of total Truvada, which is both Truvada alone, combined with the Truvada components of Atripla, has grown when compared to Epzicom, from 3.4 to 1 to 4.2 to 1. We will be able to provide more visibility on these market dynamics during the fourth-quarter earnings call in January, after we have received third-party patient data.
For Truvada, we’re also very pleased to see that despite some conversions to Atripla, presumably in patients currently receiving Truvada plus Sustiva, Truvada remains the most prescribed brand in HIV, with total prescriptions 15% higher than the second-most prescribed brand, Sustiva. Total Truvada prescriptions in the third quarter grew 2% over the prior quarter, with only a decrease of 2 market share points since the launch of Atripla.
We can attribute this continued strength to the following -- Truvada continues to be the leading NRTI backbone for all patients started on or switched to a protease inhibitor. Patients on Truvada plus a protease inhibitor currently represent 57% of all Truvada treated patients compared to 43% that are taking a non-nucleoside reversed transcriptase containing regimen.
Very importantly, just as of last week, the newly published DHSS Guidelines state that Truvada is a preferred backbone of choice with both non-nucleoside reversed transcriptase inhibitors and protease inhibitors. These guidelines now reflect what we know is already happening in the marketplace.
In the second quarter of this year, Truvada continued to hold its position as the number one prescribed NRTI backbone with the top three third agents -- the NNRTI Sustiva and the two protease inhibitors, Reotas (ph) and Kaletra. These revised guidelines will be important in expanding our reach into settings that tend to be slower adopters of newer therapies, particularly the mid to lower decile positions and the correctional setting.
We will also benefit from key data presented in two recent congresses -- the International AIDS Conference in Toronto in August and the Lipodystrophy Conference in San Francisco in September, from studies 934 and 903, respectively, that continue to demonstrate the favorable long-term safety profile of a Tenofovir containing regimen. There are several other important third-party market research figures worth mentioning that illustrates Truvada’s growth.
As of the end of the second quarter, 42% of all patients treated in the U.S. are on a Truvada backbone, up from 36% the prior quarter.
Truvada also captured more than 66% of new patient starts of the treatment of naive patients during the quarter. Truvada remains the number one backbone of choice in all patient demographics, including African-Americans and women.
Reflecting our efforts to take market share from our competition, as of the second quarter, the number of patients receiving Combivir has decreased by 37% since the launch of Truvada. Strikingly, the number of patients starting therapy with Combivir has fallen off from a peak of well over 60% in 1999 to 11% currently.
As we stated previously, the choice measure of market share for any HIV product is the molecule share. I am pleased to say that Tenofovir, the molecule in Viread, Truvada and now Atripla, continues to be the most prescribed molecule in HIV, with total prescriptions for the Tenofovir molecule exceeding total prescriptions for 3TC by 10% in August.
With the launch of Atripla, Gilead is on the verge of having the two most prescribed molecules in HIV. In fact, as of October 6th, 2006, the new prescription market share for emtricitabine, the molecule both in Emtriva and Truvada is now just 0.8 market share points behind that of 3TC molecule.
We see continued opportunities to grow our HIV franchise. With the introduction of Atripla, we believe we have a product that will become the antiretroviral regimen of choice for treatment-naive patients and patients appropriate for NNRTI therapy.
With Truvada, we have the backbone of choice for patients started on or switched to a protease inhibitor. And with Viread, we have a powerful single agent that we believe will continue to play a prominent role in treatment-experienced patients.
Turning to our HIV franchise performance in Europe, Truvada has been launched in 15 European countries, the latest launch being in the Netherlands. We have seen strong uptake of Truvada, particularly in France, the most recent of the big five countries to launch, and we have also seen continued growth in the UK, Germany, Spain and Italy.
Since the beginning of this year, which is when Truvada became fully available in the five major countries of Europe, Truvada use, as defined in Europe as days on therapy, has grown 75%. This growth indicates that Truvada is becoming the leading backbone in HIV therapy in Europe.
As of August, Truvada is now the leading branded NRTI and backbone of choice in the UK and Germany. While in France, the largest HIV market in Europe, and a country that has traditionally been a strong Combivir market, Truvada is closing the gap.
Truvada also continues to outperform Kivexa, known as Epzicom in the U.S., with sales 2.2 times that of Kivexa in six quarters since its launch. We’re seeing similar trends in Europe occur as in the United States with regards to patients new to therapy.
Truvada has now surpassed Combivir in all new starts and has grown to more than 40% patient share in the second quarter of 2006 compared to Combivir, which has declined to 24%. We’ve also seen Truvada plus Sustiva become the top regimen for patients initiating therapy, which is very important as we anticipate the approval of Atripla in Europe later next year.
On a molecule basis, Tenofovir DF, which is the active molecule in both Viread and Truvada, increased to a strong 39% patient share. In France, the Tenofovir molecule in either of its available forms, Viread or Truvada, has recently surpassed AZT and is closing the gap on the number one prescribed molecule in that market, 3TC.
In summary, the uptake dynamics of Truvada in Europe mirrors that of the U.S., albeit 12 to 18 months behind in time lines because of slower pricing and reimbursement. We look forward to furthering momentum with Truvada and will continue to target both naive and switched patients who can benefit from a Truvada-based regimen.
Turning briefly to Hepsera, in the United States, Hepsera continued to be the leading antiviral agent for treatment of chronic hepatitis B. Furthermore, more Hepsera was prescribed in the third quarter of 2006 than in any previous quarter.
Furthermore, more Hepsera was prescribed in the third quarter of 2006 than in any previous quarter. One and half years after the launch of a competitor, entecavir from Bristol-Myers, Hepsera prescription volume continue to increase with the growth of 3% from the second quarter of this year.
Importantly, Hepsera has maintained new prescription and total prescription market share above 50%. As we had pointed out previously, the introduction of Hepsera as well as entecavir has actually grown the market for patients receiving oral antiviral therapy.
And as the market leader, Hepsera has benefited more from this market growth with the pool of hepatitis B treated patients in the U.S. increasing by 30% over the last 12 months, from 33,000 to 43,000.
The majority of entecavir’s share has to date been at the expense of Lamivudine. Internationally, Hepsera is making steady gains in market share against Lamivudine, exiting the second quarter of this year at 37%, up from the first quarter share of 35%.
Our antifungal, AmBisome, reported another solid sales quarter of nearly $56 million. This product continues to maintain its position, thanks to strong brand reputation as a proven treatment for confirmed invasive fungal infections.
In summary, I am pleased with our performance during the third quarter, especially the early uptake of Atripla in the U.S. and the continued growth of Truvada in Europe.
The Gilead commercial operations area looks forward to next year and exciting challenge of preparing for the anticipated 2007 and 2008 launches of ambrisentan for pulmonary arterial hypertension and Aztreonam lysine for cystic fibrosis. I will now turn the call back over to the operator so that we can take your questions.
Operator?
Operator
[Operator Instructions]. Our first question is from the line of Meg Malloy, with Goldman Sachs.
Meg Malloy - Goldman Sachs
Thanks very much. Two quick questions.
First, for Norbert -- could you clarify a little bit -- I am not clear on the Integrase Inhibitor study. If you add PI’s, how would the impact of a boosted PI be differentiated in the study?
And then secondly, would you walk through the assumptions on timing for both candidates? Thanks.
Norbert Bischofberger
So if you know in the Merck study, the concurrent use of protease inhibitors was allowed, and these patients are highly treatment experienced at baseline. We currently don’t have all the phenotyping and genotyping data in, but my expectation is they are highly resistant to all classes of other agents.
We could not allow the use of protease inhibitors at the beginning of the study simply because we did not have any information on drug interactions. Now that we have that information, we are allowing the use of Tipranovir and Darunavir, two protease inhibitors, where we know that they do not have a drug interaction with boosted 9137.
Now, you asked about what comparison would we make. The comparison would simply be if the comparator arm is the best choice optimized background regimen versus the optimized background regimen that now includes protease inhibitor in that they added on the 9137.
So, that would be the comparison we’re making.
Meg Malloy - Goldman Sachs
Okay, so should we expect that the data from this study might be later than Q1?
Norbert Bischofberger
One thing I can tell you is the last patient crosses the 24-week time point sometime middle to late December. And I would anticipate it is going to take us two, three, four months to really analyze the data, clean them up and get them ready for public presentation.
Meg Malloy - Goldman Sachs
Okay, thanks and if you with just review -- I think you mentioned something about a 2008 submission, but could you walk through that for both of the candidates, please?
Norbert Bischofberger
Yes, so, first of all, 9137 -- so we will analyze the data sometime early next year, have a conversation with FDA also hopefully in the first quarter, and then determine, based on the data and on what FDA says, to initiate the Phase III program. The Phase III program initially will certainly be a study in the treatment experienced patients, probably with a 24-week end point.
And depending on when we initiate them, how long enrollment is going to be, I would anticipate sometime early 2008, probably last patient crossing the 24-week time point, but then add to that another six to eight months for the NDA filing.
Meg Malloy - Goldman Sachs
Okay and that study underway would be one of the two pivotal studies, right?
Norbert Bischofberger
Say that again? I am sorry.
Meg Malloy - Goldman Sachs
I don’t want to take too many questions, but that would presuppose that the current study underway would count as a sort of Phase II/III study?
Norbert Bischofberger
Yes. Or we would simply initiate two Phase III studies.
That does not change the timelines.
Meg Malloy - Goldman Sachs
And your secondary (multiple speakers)
Norbert Bischofberger
With regards to the other integrase inhibitor, it’s really too early. I am not prepared at this conference call to speculate about timelines.
But you can just apply generic timelines to it. I think your guess will be as accurate as mine.
Meg Malloy - Goldman Sachs
Okay, thanks very much.
Operator
And sir, our next question is from the line of Geoffrey Porges, with Sanford Bernstein.
Geoffrey Porges - Sanford Bernstein
Congratulations on a great quarter. Just wanted to follow-up quickly on the integrase inhibitor.
You mentioned that -- I think you said on the call that the lowest dose arm, the 25 mg arm, is switching to the high dose and then they are having the option of including a protease inhibitor. What is happening with the middle dose?
Is that also switching to the higher dose? And then is the decision to add the PI at the physician’s discretion or is that a change in the protocol?
Thanks.
Norbert Bischofberger
Let me maybe back up and make another comment. We picked the three doses with a sixfold dose range from 120 mg based on pharmacokinetic, pharmacodynamic modeling.
What we found out awhile ago is that unfortunately, and unexpectedly, the 20 mg dose gave us about a 30 to fortyfold lower -- 40% lower exposure than we had predicted. And it was based on that that actually we at Gilead asked the DSMB to review the emerging data, and really not unexpectedly, the conclusion was that 20 mg dose is not – it’s suboptimal in terms of efficacy.
The DSMB also made the right commendation to keep the other two doses onboard and only the 20 mg dose will be moved to the higher dose. So from that, I conclude that the other two doses looked okay.
And then finally, your other question was the addition of protease inhibitors. Yes, it is entirely up to physician and patient’s discretion.
Geoffrey Porges - Sanford Bernstein
Thanks very much.
Operator
And sir, our next question is from the line of Thomas Wei, with Piper Jaffray.
Thomas Wei - Piper Jaffray
Alright, thanks very much. I had a question on the guidance for the HIV sales.
So if we just look at what you did for the third quarter and the guidance for the rest of the year at the lower end, it would have seemed like the HIV sales are going to go down in the fourth quarter and at the higher end, it assumes that the sales are flat from 3Q to 4Q. Can you help us understand why you’re predicting those dynamics on a sequential basis and if there is anything more onetime-ish in the third quarter that you did not call out in the prepared comments?
John Milligan
Hi Tom, this is John Milligan. Yes, if you look at that, clearly on the low end, it would predict a down quarter, and we see tremendous strength in the franchise going forward.
There are a lot of dynamics that are always uncertain in the future, and I think this is the appropriate guidance for going forward. I can say that in terms of onetime events, the only really onetime event we had this quarter was we had two specific orders from Brazil this quarter.
As you may recall, we missed an order in the second quarter or they missed an order, and so we had two things slow into the quarter. We are anticipating another order from Brazil in the fourth quarter, but certainly not two events.
I would say in terms of the rest of the franchise, actually, we are very comfortable with where we are in terms of the process. I will mention that on the inventory programs that we have been running have been quite successful, but an interesting trend occurred this quarter.
Inventories actually came down slightly across all products by one or two days, which -- I don’t know what that was a result of. It was just an observation that we made in the United States.
So all our inventories are within the range that we mandate by contract, but they did come down. As you may recall, they were right about the midrange.
So they are in fact a little bit on the low end. We do not know what this means.
We do not expect it to go down anymore. We do not know if it will come back up either.
And so I think, Tom, that’s just the guidance that we felt was appropriate given where we are at this point in time.
Thomas Wei - Piper Jaffray
And can you help us quantify that Brazil effect in 3Q and what you are expecting in 4Q?
John Milligan
Well, the total orders for Brazil in Q3 were just under $15 million. So we would expect an order to be approximately half that.
Operator
And sir, your next question is from the line of Yaron Werber, with Citigroup.
Yaron Werber – Citigroup
Congratulations from me on a nice quarter as well. Can you in any way quantify for us what do you believe the inventory right now in the channel is for Atripla and what would be appropriate for this kind of compound right now?
John Milligan
It is a good question. We have been reluctant to put a dollar on it.
We did put an inventory in there and we have stipulated in our inventory agreement that we would have the same range for this product as we do for others. So we can report that it is within that range, again, much like the other products, slightly on what we consider to be the lower end of that range.
I know what we’re trying to get at is what is the run rate going forward, and I think at this point, we’re not sure how the inventory will build up or not and we’re a little reluctant to give a dollar amount to that.
Kevin Young
Obviously, inventory will tick up as demand ticks up. Just to add to John’s comment, in terms of opening inventory, we saw reorders for Atripla in the third week after launch, so that gives a good feeling that obviously product was moving through the channel in line with the prescriptions.
And we think we pitched it right. Always difficult to judge your opening inventory, but we think we pitched it right and we have I think a very normal pattern of ordering going on, which is -- essentially parallels Truvada and Viread now.
Yaron Werber – Citigroup
And can you just remind us -- your contracted range is between three and four weeks?
John Milligan
Yes.
Yaron Werber – Citigroup
And let me ask you one question -- you mentioned that right now, your new patient starts that Truvada is capturing is 66% of all patients. Would there be any reason when you look out to the out years, just given Truvada’s safety profile and efficacy and convenience, especially as now you’re shifting to Atripla a once daily pill, would there be any reason why you would not be able to sustain that kind of penetration over the long term, at least for the baseline, in the overall market?
Kevin Young
I think it is reasonable to presume that that 60% is at least a baseline for us. I think our mantra now here at Gilead is total Truvada, and we can essentially be successful on both fronts with Truvada being combined with the protease inhibitors and, of course, having the single tablets with Sustiva.
So we feel very positive about the picture going forward. Obviously, there is competition still there.
We have to ensure that without sales organizations we get the right reach right down the spectrum of prescribing physicians. And so we think we’re in a very good position.
Yaron Werber – Citigroup
And maybe final question for me, can you share with us if you can what is your internal goal for what would be an appropriate penetration rate for your product into the market over the long term?
John Milligan
Well, Yaron I think it is fair to say we’re not happy with where we are. I think the team has done a terrific job, but given, as you pointed out, the profile of the product and the difference it is making for patients, we clearly have a goal to do much better.
Kevin Young
Maybe to answer that another way, there is 89,000 patients still on Combivir. Of that 89,000, 30,000 of those are Combivir plus Sustiva.
So there has got to be a patient group that we can go after, and there are still, believe it or not, 36,000 patients on Trizivir. And again, we believe those are patients that really could benefit from treatment with Atripla.
So we believe there are still buckets of patients who could switch as well as obviously performing very well with the naive patient group.
Operator
[Operator Instructions]. Sir our next question is from the line of Bret Holley, with CIBC World Markets.
Bret Holley - CIBC World Markets
Hi, thanks for taking my question. I’ve got a question on Merck’s progress relative to your progress for the GS-9137.
What are you hearing as far as their progress in Phase III? And how far do you think that they are ahead or how far do you figure they are ahead with their integrase inhibitor?
Norbert Bischofberger
I am sure you -- this is all -- I do not know anymore than what is publicly available, but they have initiated and completed enrollment in two experienced patient studies and they have initiated a naive study as well. They have presented data on both Phase II studies in the experienced and naive populations and the data look very exciting.
And that is what makes us excited about our integrase inhibitor, because the Merck compound is given BID and we have QD administration. But overall, I would certainly say that the Merck program is ahead of ours.
There’s no question, maybe by a year to 18 months, something like that.
Bret Holley - CIBC World Markets
And Norbert, how would you characterize the efficacy? Based on the data that we have seen, would you characterize the efficacy being essentially equivalent to these two agents and hence the reason why the BID dosing versus the QD dosing is going to be such a commercial differentiator?
Norbert Bischofberger
The efficacy, I think -- the only thing I can compare is the -- what Merck published about their Phase I/II study in naive patients’ monotherapy compared to our naive study monotherapy and the potency is very similar between the two protease inhibitors. But that also makes sense from a pharmacokinetic, pharmacodynamic and in vitro potency point of view.
Yaron Werber – Citigroup
Thanks, Norbert.
Operator
Sir, our next question is from the line of Geoff Meacham, with JPMorgan.
Geoff Meacham – JPMorgan
Hi guys and congratulations on a good quarter. I know you gave some market share numbers.
I was wondering if you could just give us actual patient numbers this early in the Atripla launch, the patients switching from Truvada to Atripla and then Combivir to Atripla.
Kevin Young
Unfortunately, I have nothing to give you because our (indiscernible) database will not report until later in the fourth quarter for the third quarter. So we just don’t have any transparency into patient numbers and where those patients are being transferred from, what are the patient regimens they are coming from.
So right now, we’re operating on prescription volume and we are operating on obviously prescription share. So we just don’t have that right now.
Geoff Meacham – JPMorgan
Okay and just with respect to the European filing, obviously, you guys just filed. But any indication from your early discussions that we can get a faster than, say, a 10 to 12-month review in Europe?
John Milligan
We have no indication that will happen any faster than the traditional process.
Operator
And sir, our next question is from the line of Ian Somaiya, with Thomas Weisel Partners.
Ian Somaiya - Thomas Weisel Partners
Thanks for taking my questions. I have two -- the first one is for Norbert, just on 9137.
Does this impact the planned, at least initial design for the Phase III trials in terms of combining 9137 with Truvada?
Norbert Bischofberger
No, it will not. I mean, we have always intended to do the study as a combination study with Truvada in the naive patient population.
The only thing we can do now is use concurrent protease inhibitors. That is the big change from what we had previously in the Phase II study.
Ian Somaiya - Thomas Weisel Partners
And I do not know it Kevin can comment on that -- what would be the impact of requiring or needing concurrent PI use?
Kevin Young
Basically, from I think the data that we see, in the United States, approximately 70% of the patients have a boosted protease inhibitor regimen. So we feel that is okay from a commercial point of view.
And to have that once daily advantages and then have the boosting, we certainly think it is a very commercializable regimen.
Norbert Bischofberger
Just one comment that you do not misunderstand us -- it is not that we require patients to use concurrent protease inhibitors. We give patients that are highly treatment-experienced a system to most classes of antiretrovirals, we give them the option to, in addition to nucleosides and P20, they can also use protease inhibitors as part of their background regimen.
Ian Somaiya - Thomas Weisel Partners
And just the second question for Kevin -- can you comment on what the duration of scripts is for Atripla at this point, and how that compares to Truvada?
Kevin Young
Unfortunately, I have got no insight certainly at my fingertips into that. We can certainly try to follow up later.
Obviously, the majority is based on -- it is kind of NRX because that is the lead indicator with only 13 weeks of prescribing. So our typical -- and our published prescription is about 1.2 months.
So we would expect it to get right at 1.2 months worth of therapy. So that is the combination of individual bottles as well as the longer-term prescription of mail order.
So I think Atripla will settle down quite quickly to match what we have with Truvada or Viread.
Ian Somaiya - Thomas Weisel Partners
Okay, thank you.
Operator
And sir, we have a question from the line of Mike King, with Rodman & Renshaw.
Mike King - Rodman & Renshaw
Thanks for taking my question and let me add my congratulations on a fine quarter. I wonder, Kevin, can you give us a breakdown of the use of Truvada with -- the proportion of Truvada that is used with PI’s versus with Sustiva or other NNRTI’s and what you think the impact and the change in the guidelines might be going forward?
Kevin Young
The split for Truvada is 57% used with all PI’s and 43% with all NNRTI’s. So that is the split.
We think that having the DHSS guideline preference endorsed is good. Whether we are going to see a shift in that, I think that remains to be seen.
Clearly, the single tablet regimen obviously supports the combined use of Truvada plus the NNRTI. And then again, there are obviously loyalists to the combination with a protease inhibitor.
So I think, obviously, we can be successful on both fronts. How that dimension is going to play out, I think remains to be seen, actually.
Mike King - Rodman & Renshaw
Okay, fair enough. And then if I can ask my requisite second question, for either you or John Milligan, in terms of the guidance to the HIV franchise, I am wondering what impact do you think the change in the guidelines will have in your forward guidance?
Or is it just too early for the change in the guidelines to have any effect in ‘06 and more of a matter of what impact it will have on ‘07 just in terms of the guidelines of when to treat and the new CDC testing requirements, etc.?
Kevin Young
I will let John add to my comments, but obviously, it is all very positive news. The one thing about guidelines is they actually tend to come out after clinical practices change.
And certainly opinion leaders and our high decile doctors are very much -- have already led the DHSS guideline. It is just with the lower decile doctors who look towards those to change their habits in a setting like the corrections area.
I think in terms of CDC, we see that as very, very significant to increase the pool of patients right to the top. But it is going to take some time for those to be instituted at the state level, and in some cases, it will require legislative change.
We’re going to work very hard. We have dedicated teams that help with government affairs at a local and state level.
So we’re going to be working with committee people and government people to try and affect that change, but I think it is going to be progressive.
John Milligan
I would agree with that. I mean it’s very clear that the things that are being put into place this year will start to have an impact as we go into ‘07, ‘08 and beyond.
We’re really not factoring for 2006 much of an additional impact because these things are relatively new. It takes some time to filter down, and clearly we’ll have to help to get into those lower-decile doctors with both our sales force and the Bristol-Myers sales force who are copromoting Atripla.
And that will be our opportunity to make that impact felt next year and beyond.
Operator
And sir, our next question is from the line of Mark Aberman with Credit Suisse.
Michael Aberman - Credit Suisse
Thank you, it’s actually Michael. I am going back to integrase inhibitor.
I am wondering, can you discuss your thoughts on the treatment naives for the boosted, and whether or not you plan to use that agent or your follow-on agent in the treatment-naive? Is there anything about PI boosting that prevents you from going with a triple therapy, let’s say a Truvada plus your GS-9137 in the treatment-naive population?
Norbert Bischofberger
Mike, I would say there is nothing that prevents you, but there is a challenge that we have to overcome. The challenge is, there is some concern that certain people have about the development of protease inhibitor resistance mutation in the treatment of naive individuals because they are only on 100 mg of Retonavir.
I mean our counterpoint of that is if you have a completely suppressive regimen and there is no viral replication going on, then by definition you cannot develop resistance. But we just have to generate some data to make everybody feel comfortable with that, which I think will happen.
Michael Aberman - Credit Suisse
So how does that effect your timeline for treatment naive with this agent? In other words you think your other compounds might leapfrog for the treatment-naive?
Norbert Bischofberger
I think -- I would not want to speculate on that now, but the first study is certainly going to be the 9137 boosted with Retonavir in treatment-experienced patients. And subsequent to that, we will consider going into naive patients.
Michael Aberman - Credit Suisse
And then if I can again follow up on this issue of the DSMB. I suppose the question is, did the DSMB come to you and say hey, we want to use protease inhibitors?
Or did you provide them with data and say hey, by the way, you can now use protease inhibitors?
Norbert Bischofberger
I want to point out this is a blinded -- that the doses that we’re using, 20 to 120 mg, we are blinded to. So we submitted to the DSMB unblinded dose, unblinded data and also informed them about our recently completed drug interaction studies.
And then it is the DSMB that came to the conclusion that the 20 mg dose should be discontinued and that we should offer protease inhibitor use to patients.
Michael Aberman - Credit Suisse
That is very helpful. Thanks.
Operator
And sir, our next question is from the line of Sapna Srivastava with Morgan Stanley.
Sapna Srivastava - Morgan Stanley
Just could you give some more color on your efforts to gain market share from well-controlled patients on Combivir for Truvada and Atripla? And if the recent guideline change of now prescribing -- now making Truvada also the backbone for PI’s could impact that?
Kevin Young
Well, obviously as I said earlier, there are 89,000 patients in total, 30,000 Combivir plus Sustiva, which is obviously a primary area for us to concentrate on from a Gilead point of view. We are, obviously, using the 934 data.
We did not get that into our label until the second quarter of this year, and that is very, very important to us. So that is something that we totally leverage in terms of our sales aid.
It is in the Atripla sales aid, and we very much highlight that when we talk to our physicians. So I think we have made great progress with nearly 40% of Combivir patients reduced in number, and we believe we can continue that decline in Combivir patients with a great deal of focus.
Sapna Srivastava - Morgan Stanley
And just two quick questions -- one just on the tax rate, what we can expect for the future, for the next quarter, at least. And also just on Tamiflu copromote, what do you plan to do with that as it comes up in 2007?
John Milligan
So just to be clear, our full-year guidance for tax rate remains unchanged. I do not know that I can accurately predict a quarterly tax rate based on the variations in the business, so I would just assume our guidance holds.
And then secondly, the Tamiflu copromote -- we at this time have no plans to promote Tamiflu for the season. We will of course evaluate the effort of Roche as they go through the season and we are starting to look now at 2007.
And we have not yet made a determination whether we would have the opportunity to copromote. Or whether we should take up on this opportunity to copromote.
I will say that for example, as we’re launching ambrisentan next year, I don’t know that it would be the best use of time to have that sales force distracted with two new products, but it is something we will certainly consider.
Kevin Young
We have also been very pleased, looking at the brand plan for Tamiflu for the ‘06, ‘07 seasonal flu to see the increased effort that Roche is going to put behind the brand, including direct-to-the-consumer advertising. So I think we feel very good that they have really stepped things up.
Sapna Srivastava - Morgan Stanley
Thank you.
Operator
And sir, our next question is from the line of George Farmer with Wachovia Securities.
George Farmer - Wachovia Securities
Hi, thanks for taking my question. Switching gears for a moment to the Myogen pipeline, could you review with us your plans for a Phase III trial with darusentan, if that’s still in the cards, and when you think you would get that going?
John Milligan
So very clearly, the tender offer has not been completed, and Myogen is still an independent company. They have started one Phase III study and darusentan.
They are looking at refractory hypertension. And they do anticipate starting a second Phase III study of darusentan before year-end.
So at this point, I think it is fair to say that we’ll be going forward with two Phase III studies of approximately 1000 patients in total for darusentan. We’ll begin by looking at refractory hypertension, but as we move these two companies closer together, we’re going to start looking very carefully at the different kinds of populations that we think could benefit most from this kind of product, which may not be a broad refractory hypertension, but more specific subsegments of refractory hypertension, and that is what we will be exploring with the team there.
But really, it is very early on and we have not yet had a chance to get the teams together to discuss strategy.
George Farmer - Wachovia Securities
Okay great, and Norbert, could you share with us just a few details about the HCV drug, 9132, and how that actually differentiates from the other protease inhibitors that are in development?
Norbert Bischofberger
As you may remember, 9132 was discovered by Achillion, and it is a compound that is an HCV replicase inhibitor. It involves HCV protease, but it is not an active site inhibitor like the other -- all the other HCV protease inhibitors that are currently out there.
So the expectation would fully be -- and in vitro data confirmed this -- that resistant mutations to the existent protease inhibitors would not be cross-resistant to this one. And so this one should be able to be used in combination with existing -- like the Vertex protease inhibitor, because it is really a different mechanism.
Operator
And our next question is from the line of Jim Reddoch with Friedman, Billings, Ramsey.
David Amsellem - Friedman, Billings, Ramsey
Hi, David Amsellem for Jim Reddoch. A couple questions.
Just on, turning back to inventory, can you sit with a number of weeks on hand were for Truvada and Viread at the end of the quarter and did that decline during the quarter? And then secondly, this is just more of a viewpoint question, maybe for Kevin -- of patients on Truvada, Combivir or new patient starts, which of these do you believe is the lowest hanging fruit in terms of Atripla adoption?
Thanks.
John Milligan
Just quickly on that last question, was that new starts versus switches?
David Amsellem - Friedman, Billings, Ramsey
Well, basically switches versus new starts, switches being Truvada switches, Combivir switches and then new starts, of those three.
Kevin Young
Let me take the second one first. I think there’s obviously three buckets of patients, as I call them, that we can target.
Our own patients, and there are 71,000 Truvada plus Sustiva patients who obviously would benefit from conversion to one tablet once a day. There are 30,000 Combivir plus Sustiva.
And as I said also earlier, there are 36,000 patients still currently on Trizivir. So I think they are the buckets of patients that we will be targeting to convert.
John Milligan
And then with regard to inventory, you mentioned change in weeks on hand. I think our targets vary across the major wholesalers where we have inventory management agreements.
But our target is between three to four weeks. So actually changes that we measure are measured in days, not weeks.
At the beginning of this quarter, so the end of last quarter, as we reported, our inventory was above the midpoint, and it is now below the midpoint in that range. And so we have had a change across all of these things, about two days of inventory.
Given the amount of inventory we put through there, that is actually millions of dollars of differences, and that can make a difference in yearly. So I do not know what the fluctuation will be for the future, but we are very happy with the range that it is in right now.
David Amsellem - Friedman, Billings, Ramsey
Okay, that is great. That’s helpful.
Thanks.
Operator
Phil Nadeau - Cowen & Co
Good afternoon, congratulations on the good quarter. Two questions on the market.
Can you tell us what you’re seeing today in market growth, how many patients are starting therapy today versus say 12 months ago? And you mentioned the CDC guidelines could greatly increase the pool of new patient starts.
Could you put some numbers around that? Is that a 25% increase or a doubling of new patient starts?
Kevin Young
Overall, the total number of patients currently on antiretroviral therapies, 464,000 in the U.S. On an annual basis, that seems to be increasing at just under 8%.
Actually, that has gone up versus previous years, which again I think demonstrates that patients are -- there is initiative behind getting patients screened, and more importantly, on therapy with a drug like Truvada and now Atripla, which is well-tolerated and patients can stay on therapy for long periods of time. So I think there is a very positive dynamic in patients going onto therapy.
And of course, if the CDC guidelines start to really stick at a state level, then that dynamic of the market changes. If you take a place like Washington, D.C., who have really taken up the charge, they are doing an immense amount of work in terms of screening.
And if other states start to follow that cue, clearly, we are going to see more diagnosed patients.
John Milligan
And I guess your other question was about the CDC and what those guideline changes will mean for us. It is estimated that there is at least 250,000 people undiagnosed in the United States, and so that increased testing can only enhance that.
And we have got some visibility that the diagnosis rate is coming up. We had previously estimated it was about 750.
The most recent data we have suggests about 825,000 people have now been diagnosed, so that is an increase of about 750,000 -- I am sorry, 75,000 people over the course of the last -- I think this is over the last year or so. And that will lead to greater patient starts, we believe.
Phil Nadeau - Cowen & Co
Thanks. That is very helpful.
Operator
And Dr. Milligan, there appear to be no questions at this time.
John Milligan
Thank you, operator, and thank you all for joining us today. We appreciate your continued interest in Gilead and look forward to providing you with updates on our future progress.
Operator
Ladies and gentlemen, we thank you for your participation in today’s conference. This concludes your presentation, and you may now disconnect.