Arvind Sood – VP, IR Kevin W. Sharer - Chairman, CEO, President Richard D. Nanula - CFO, EVP George Morrow - EVP Roger M. Perlmutter, M.D., Ph.D. - EVP

Eric Schmidt - Cowen & Company Elise Wang - Citigroup Mark Schoenebaum - Bear Stearns Gene Mack - HSBC Securities Joel Sendek - Lazard Capital Market Steven Harr - Morgan Stanley Geoff Porges - Sanford Bernstein George Farmer - Wachovia Securities Jennifer Chao - Deutsche Bank Shiv Kapoor - Montgomery & Company Geoff Meacham - JPMorgan May-Kin Ho - Goldman Sachs David Witzke - Banc of America Securities Bill Tanner - Leerink Swann Stanley Grossman - Rudman Capital

My name is Gerald and I will be your conference facilitator today for Amgen’s second quarter financial results conference call. (Operator Instructions) I would now like to introduce Arvind Sood, Vice President of Investor Relations. Mr. Sood, you may now begin.

Thank you, Gerald. Good afternoon, everybody. I’d like to welcome you to our second quarter business results conference call. I hope you’ve had a chance to download the slides that we’ll use for our presentation today. These slides have been posted on our web site and they were also sent separately by email. We have decided to use slides as part of our efforts to continue to improve our communications with the investment community. So in terms of our speakers today, Kevin Sharer, our Chairman and CEO, will start the call with an overview of our second quarter performance, followed by our Chief Financial Officer, Richard Nanula, who will discuss our second quarter performance in detail and then provide an outlook for the remainder of this year. George Morrow, who’s our EVP of Commercial Operations, will then provide a commercial update; followed by Roger Perlmutter, who is our EVP of Research and Development and Roger will provide a pipeline update. But before we start I’d like to mention that during the course of our presentation today we’ll make certain forward-looking statements and, of course, actual results can vary materially. So with that I’d like to turn the call over to Kevin. Kevin W. Sharer: Thank you, Arvind. Glad to be with you all today. What we’re going to do -- this is a first for us and we’ll try to do it in a way that helps you -- we do have slides that we’re talking from and we’ll refer to the page numbers and we’ll take it from there. My comments relate to the highlights for the quarter. It’s Page 5. You can see the results there. You probably about them in the press release. We said excellent financial performance. Our revenue and adjusted EPS grew 14% and 19%, respectively and consequently we’re raising adjusted EPS guidance $0.15, to $3.75 to $3.85. Each area of our performance is real high quality and, in our view, continuing. On the revenue side, our growth was demand driven, with no year-over-year benefits from inventory build or spill-over, and particularly notice Aranesp that grew 26% to our first $1 billion quarter after 20 quarters since its launch strikes me as really, really impressive; and, we’re continuing to gain share in the U.S. and Europe. Below the revenue line, we like it too. We’re able to grow EPS 5 points faster than revenue, despite investing in R&D, as we said, at a little more than twice the rate of revenue growth. In this quarter it was 29%. We also are investing in intellectual property defense and strengthening our infrastructure. Improvements in our cost of sales due to lower royalty expenses and manufacturing efficiencies are really significant and I want to thank the members of Amgen’s operations team on such a high level of execution. Richard will get you the details on the guidance and increase in the trends behind them. We recently returned from a multi-day, multi-site European operating review and I’m particularly proud of the performance there. The total international sales grew 18% in the quarter before foreign exchange impacts. Through the good work of a lot of people at Amgen and regulators, we believe that the EMEA decisions regarding Biosimilars are good for patients and we expect to see the Biosimilars first competitors show up in Europe in 2007 and George will talk more about that. We’re investing a lot of money in the pipeline. We’re excited about the pipeline and the progress of the pipeline. This quarter had a number of highlights. Of the 11 mega-trials that we mentioned to you all in January, seven have been initiated so far. We have a lot of confidence in our late stage pipeline and we’re really committed to long-term growth. There’s one product, the AMG706, in the cancer array. Roger will talk about it. One trial GIST, there were some issues that Roger will talk about and I’ll leave that to him. I expect and hope by the third quarter when we talk to you next we’ll be on the market with panitumumab. Our colorectal cancer drug’s now called Vectibix and that obviously is pending FDA approval of our facility in cardinals. While we’re going to launch Vectibix initially with a label that’s a bit narrow, we’re working very hard in the clinic to broaden that label and expect it to be a fully competitive product. Let me move to the next slide that’s highlighted ‘Litigation Update’ it is slide 6. One thing we’ve announced before, you know it, is an ITC judge made an initial determination in the Roche case. It’s a procedural decision to delay considering the case based on the clinical trial exemption. It has nothing to do with the merits of the case and we’ll continue to vigorously contend it there. And the Peg-EPO action against Roche is in the federal court in Boston, Judge Young’s court. We expect a procedure decision in ‘06. I’d like to make just a couple of comments; Roger will amplify. We firmly believe Roche’s product, based on what we’ve seen, what they’ve said, the evidence is simply pegylated epoetin beta and that’s based upon Roche’s own data. Their product offers no meaningful advantage for patients and we believe both the product itself and its method of manufacture infringe our patents. If Roche attempts to launch this product post-approval, we’ll ask the court to issue an injunction. I want all of Amgen’s investors to rest assured that we’re fully prepared to take all the appropriate legal and commercial actions necessary to protect the company’s rights and our business. I think there are also some significant questions about Roche’s clinical data that just haven’t been answered yet and again, Roger will talk about that. I’ll come back and summarize things at the end of the call but now I’d like to turn it over to Richard Nanula. Richard D. Nanula: Thank you Kevin and I’ll start, everybody, with my remarks on Page 8, the income statement. Because it is the income statement I’ll spend a decent bit of my time on this particular page. Total revenues grew 14% in the quarter, driven by strong demand for Aranesp, Neulasta, NEUPOGEN as well as Enbrel. Overall, as Kevin mentioned, sales growth was driven by demand. We ended Q1 with Aranesp and Neulasta wholesaler inventories at below our normal expected range and by the end of Q2, inventory levels had recovered towards the upper end of the normal range and are consistent with Q205 levels. Second quarter wholesaler inventories are still within the normal range across all of our products and again, as Kevin mentioned, year-over-year growth came all from demand. Geographically, the second quarter saw U.S. sales at $2.9 billion, that’s a 13% increase. International sales were $630 million; that’s a 17% increase. We were impacted by about $9 million negatively from changes in foreign currency. Excluding that impact, worldwide sales would have grown at 14% and international would have grown at 18%. Turning to operating expenses, which I always discuss on an adjusted basis, cost of sales decreased 7% versus Q2 of ‘05, primarily due to lower royalties expenses and to a lesser extent, production efficiencies, especially with respect to Enbrel. Royalty expenses were lower than the prior year, driven mostly by the expiration at the end of 2005 of certain contractual royalty obligations that we had on Neulasta and NEUPOGEN. We also repurchased royalties from one institution on international NEUPOGEN and Neulasta and settled with another, effectively ending a royalty obligation on Enbrel. Both of those transactions were announced previously. For those of you who follow cost of sales on a percent of sales basis, we dropped a little over 3 percentage points from Q2 of 2005. Of this particular decrease, around 2 points came from the reduced royalty obligations and the remainder is primarily driven by the production efficiencies. R&D expense grew 29% versus Q2 2005. The increase was primarily due to the planned hiring in support of our aggressive clinical program and increased funding necessary to support our late-stage programs, including clinical material and manufacturing costs. Second quarter of 2006 expenses also fully reflect the impact of the Abgenix acquisition, which closed in April 2006. SG&A expenses increased 24% in the second quarter. In the second quarter, SG&A growth was due to higher staff and additional infrastructure costs to support our growing organization, including our spending on our new ERP system. We also saw higher legal expenses associated with our ongoing litigation and we also experienced higher Wyeth profit share expenses as Enbrel profitability improves due to the royalty reductions and production efficiencies. Our tax rate was 23.1% for the quarter versus 22.1% in the second quarter of 2005. The rate was a little bit higher due to a favourable resolution of a foreign tax credit claim last year and the expiration of the R&D credit for this year. For the full year we continue to expect our adjusted tax rate to be lower than the 2005 rate, even if the R&D credit is not extended. Second quarter earnings per share grew 19%, as Kevin mentioned, versus what we had as a truly outstanding quarter last year. The growth, though, was 5 points faster than revenue growth and the leverage was driven really by the fewer shares and higher interest income due to the convertible bond deal and the accelerated share repurchase we executed in the first quarter. On a GAAP basis, EPS was $0.01 in the second quarter versus $0.82 last year. As you’ll recall with our previous acquisitions of Immunex and Tularik, we’re required by GAAP to take a one-time write off of in-process R&D in the quarter in which the transaction closes. For reference, those amounts were $3 billion and $600 million, respectively, for Immunex and Tularik, so for Abgenix in the second quarter we wrote off $1.1 billion of acquired in process R&D. As was the case in each of our prior acquisitions, the bulk of the balance of the purchase price is recorded as goodwill on our balance sheet. We began recording expense associated with employee stock options effective January 1, 2006. As a result, reported GAAP results for the second quarter of 2006 were negatively impacted by $63 million pre-tax. Stock option expense in the second quarter was $0.04 versus $0.05 in the second quarter of 2005. Adjusted EPS including stock option expense was $1.01 for the second quarter of 2006, an increase of 22% compared to $0.83 in the second quarter of 2005. Moving on to slide 9, I’m trying to show you here the quarter-over-quarter or sequential growth of revenue, product sales here since 2004. We’ve noticed as our major products, particularly those that we’ve launched in the last several years, move through their life cycle, we’ve noticed some seasonality trends that we wanted to point out to you. Since 2004, as you can see from the slide, there's a pattern of higher sequential growth in the second and fourth quarters, and so far 2006 is following this pattern. Our analysis leads us to the conclusion that the second and fourth quarters are most impacted by the fact that the holidays occur at the very beginning of the first and the third quarters, affecting shipping -- and that being, really, New Year’s Day and the Fourth of July. Wholesalers raise their inventory levels typically to the upper end or even above the upper end of the normal range in anticipation at the beginning of the following quarter, when order taking and shipping is on hold. In Q2, while inventory levels grew sequentially and it was in the normal range for all of our products, inventories at the end of Q2 were actually the same or lower than Q2 of 2005. We have previously discussed this topic in terms of shipping days, and while total shipping days do vary quarter to quarter, we believe that more than the shipping days, it's the timing of those shipping days which are specifically the first few days of the quarter that have the most impact. Moving to slide 10 for a discussion of our outlook for the second half of the year, as I discussed, we do expect continued favourability in gross margins driven by a more a profitable product mix, the production efficiencies, and the lower royalty expenses. On the R&D front, we do expect R&D expense growth to accelerate, due to the momentum of our seven mega-trials we initiated in the first half of the year, as well as additional large-scale trials commencing in the second half, as well as the continued growth in staffing to execute those trials. We expect SG&A expense growth to moderate versus the first half of the year, as we spent more money in Q4 last year on sales and marketing initiatives than we expect to this year. Finally, as I mentioned, we still believe our full-year adjusted tax rate will be lower than 2005, even though the quarterly rates may fluctuate based on potential audit settlements. So I'll move you now to slide 11 and tie all this together for you for guidance. We do remain confident in our revenue guidance and with a half of the year complete, we are narrowing the range from $13.9 billion to $14.4 billion to $14 billion to $14.3 billion. The important point here is that our mix of sales has changed in favor of Aranesp. So with Aranesp's higher margins, coupled with our favourability in cost of goods from production efficiencies and royalty reductions both, I will reiterate what Kevin mentioned and that is we’re going to raise our guidance by $0.15, up to $3.75 to $3.85. My final slide is slide 12. Just a couple of updates on the balance sheet. You'll recall in the first quarter this year we issued $5 billion of convertible notes and in the second quarter we paid approximately $2.5 billion in cash in conjunction with the Abgenix acquisition, and bought back almost $3.5 billion worth of stock. These transactions were the primary drivers of the changes in our cash and debt position. Capex for the second quarter was around $233 million, versus $205 million in the second quarter of ‘05. We are continuing to add to our manufacturing capacity and expand our sites in Puerto Rico, and we are investing in our ERP system, some of which is capitalized. And then lastly in the second quarter we repurchased 13 million shares at a total cost of $876 million. Year-to-date, we've now bought almost 60 million shares, a total investment of around $4.25 billion. In December of '05 the Board authorized to us spend up to $5 billion on the share repurchase program. We currently have around $2.3 billion remaining under that program. So as you can see, average diluted shares for adjusted EPS were 1.181 billion, versus 1.25 billion in the second quarter 2005, that reduction reflecting our aggressive share repurchase efforts. Now, I will turn the call over to George for a commercial update.

Thanks, Richard. I will start with some highlights on slide 14. As Kevin mentioned, we had our first ever $1 billion quarter for Aranesp, and also for Filgrastim. Aranesp remains the key driver of our overall growth, representing about half of our total growth. We are encouraged that Enbrel is showing some early signs of improved performance. In the rheumatology segment, we've had a stable share for the last two quarters, and the segment rebounded to over 20% growth for the quarter. On the other hand, derm has not rebounded yet and we will talk more about what we are doing about that in a moment, but it is a point of focus and we do see a way forward. Also, as Kevin mentioned, very strong international revenue growth of 18%, excluding foreign exchange impacts. Ex-U.S., biosimilars appear increasingly manageable. In particular, we're more optimistic that complex proteins such as EPO are not likely to be substitutable. On the left side of slide 15 we have quarterly international sales from the last 18 months. As you can see, my international colleagues had a terrific performance in the second quarter, with 18% growth, again at constant exchange rates. I should point out, however, that second quarters tend to be very strong. Turning to key drivers, the biggest drivers were Aranesp, nephrology, and Neulasta. With respect to Aranesp, extended dosing regimes are now the standard of care in Europe and by that I mean, every three-week dosing in oncology, every two-week dosing in dialysis nephrology, and every four-week dosing in nephrology. We continue to take share with Aranesp inching away towards 50% in both nephrology and oncology. Turning to Aranesp on slide 16. Aranesp grew 26% second quarter ‘06 over second quarter ‘05 and as Richard mentioned, since inventory was relatively flat between those two quarters, 26% is a good measure of the true dollar demand growth during this period of time. Sequentially however -- and that's second quarter ‘06 over first quarter ’06 -- inventory was up even though the second quarter inventories ended the quarter in the normal range. In fact, a little bit towards the high end of normal range. So in other words, projecting from the first quarter to the second quarter would not be a good indicator of true underlying demand growth. As far as normal ranges is concerned, it varies a little bit by product, but it's generally between ten and 15 days. Turning to the key drivers, our every three-week dosing approval in the U.S. is being rapidly adopted and as a result, is driving continued Aranesp share gains in the U.S. and it might even add in Europe, where we were approved for Q3W in chemo last year. In Europe, we saw an uptick in segment growth in both nephrology and oncology in the second quarter, which helped our business there as well. The next slide is slide 17. As the title states, we are taking steps to improve Enbrel growth. One quarter doesn't make a trend, but we believe we are seeing early signs of an inflection point in our underlying growth curve. As I mentioned up front, the rheumatology segment growth is improving. It went from 16% in the first quarter to 21% in the second quarter. We've seen a stable share for Enbrel at 41% throughout the first half. Dermatology segment growth is still slowing slightly. It went from 21% in the first quarter to 16%. But we are still confident that recent initiatives will cause a turnaround. For example, we are back on air with a non-branded TV ad and are working with the FDA to get a branded TV ad on in the near future. What we've experienced through our market research is that since we've gone dark with our direct-to-consumer branded Enbrel advertising, we've seen a big drop off in the number of patients asking for Enbrel, and that certainly is an issue. Also given the continued managed care focus on biologics, we are implementing new support services that will help patients and offices navigate through such things as prior authorizations. Dermatology offices historically have not been really set up to manage the extra work in the usual two to three weeks that it takes to get a prior auth approved, so we are helping them do that. We're also increasing patient assistance program funding, and this is a way to deal with the increase in co-pays that we are seeing in some segments of the market. We're also about to launch SureClick, our autoinjector, which will make it much easier for patients to self-inject. So just to summarize on Enbrel, given the low penetration of biologics in both the rheumatology and dermatology addressable markets, we believe the long-term growth prospects for Enbrel are excellent and I have tremendous confidence in the team that we are going to get after it. Turning to slide 18, NEUPOGEN and Neulasta grew 12% second quarter ‘06 over second quarter ‘05. Again, inventory was flat between those two periods so it's a good indicator of true demand. Sequentially, however, second quarter ‘06 over first quarter ‘06 inventory is up, but again still within the normal range so I wouldn't recommend projecting sales based on that increase in first quarter to second quarter. The key driver remains increasing penetration of first-cycle prophylaxis of the appropriate addressable market. We estimate this addressable market consists of about 600,000 chemo patients; that's 600,000 of the 1.4 million who get chemo every year in the U.S., and less than half of whom are treated first-cycle today. So, in other words, we feel there are about 300,000 patients who could be treated prophylactically who are not today. We also believe that the recent release of ASCO guidelines and European guidelines, which are now aligned with NCCN guidelines and our new label is going to put the wind at our back with regard to driving this with both providers and payors. Turning to EPOGEN, slide 19. As you can see on the chart on the left quarterly sales of EPOGEN tend to be choppier, which makes calculating a true underlying demand curve more difficult. The primary reasons for the choppiness are wholesaler inventory changes, end user inventory changes -- and that's oftentimes contractually driven, spill-over true-ups -- and that's spill-over between Procrit and EPOGEN in the various markets, and, finally, Aranesp hospital conversion. So without running through all the math, we are confident in saying that underlying demand in free-standing dialysis centers is consistent with annual patient population growth of about 3% to 4%. Conversion to Aranesp in dialysis has stabilized at about 200 million to 240 million, so it should wash out for the rest of the year. On to Sensipar, slide 20. So the news here is that, on top of a steep growth curve since launch, late first quarter continuing into the second quarter, we had a step change resulting from SAFETY NET and new Medicare patients transitioning on to Part D and providing a good prescription benefit for those patients. We are also seeing demand grow through appropriate dose titration and segment penetration, and in particular, greater use in patients whose PTHs are less than 800, which is a sign of increased physician comfort with the drug. Slide 21 provides a brief overview of reimbursement. The fact that I'm not going to spend a lot of time on it suggests that things are a little calmer than they have been in the past. First, ASP values are stabilizing, reflecting stable product discounts and small price increases. Adoption of ASP is less than 25% in managed care today, although it is increasing. I should also say that, although it's variable, the average is about ASP plus 15%. As I mentioned, payors are cost-shifting to patients through increased use of co insurance and higher co-pays, and this is the reason we have increased our patient assistance through, again, co-pay foundations and patient support systems that help ensure access. As the title of slide 22 states, we are prepared for follow-up biologics in Europe. As Kevin mentioned, we are very pleased with the EMEA regulations. They do reflect a real sensitivity towards ensuring patients' safety. Every regulatory agency has come out and said that follow-on biologics are not generics; they are different. Now we are working very hard at the country level to ensure that automatic substitution of follow-on biologics is not permitted. Doctors should be the people who make the decision of whether or not a patient gets put on one of these products, not some pharmacist or fiscal intermediary. Lastly, for planning purposes, we do not expect follow-on biologics until some time in 2007. My last slide is simply to let you know that panitumumab launch preparation is on track. We are excited. As you heard we have a brand name, Vectibix, and we'll look forward to reviewing more of the details of our launch plan on the third quarter conference call, including pricing. So before I turn it over to Roger, I would like to close by saying that our business performance is rock solid. We're intensely focused on Enbrel's growth opportunities, and we don't see any insurmountable obstacles to future growth.

Thank you, George. I'll have the opportunity to review this afternoon with you highlights of the R&D performance for the second quarter of 2006. As you know, and as we presented at the beginning of the year, 2006 is a year of enormous investment in our late-stage pipeline. We've begun an awful lot of clinical trials, seven in the first half of the year. I won't have the opportunity to go through all of that information with you, but we've tried, through this slide deck, to present more information about this than ever before. You will see that a great deal has been accomplished. Let me just say that with respect to Vectibix, as we've indicated, we do expect approval in the third line colorectal cancer setting in the United States and launch by year end. For AMG 706, we were able to confirm clinical activity, which we previously presented from Phase I studies at ASCO in the gastrointestinal stromal tumor setting. At the same time, we have a new safety observation which I am going to spend some time talking about with you, which is currently under investigation. For AMG 531, we did complete enrollment of the second of our two Phase III studies in immune thrombocytopenic purpura, the lead indication for this molecule and we also began other studies that we will talk about. For denosumab, we have now had the opportunity to take a look at three-year data from our Phase II study in post-menopausal osteoporosis. I will show you a little bit of data from that study and it suffices to say that denosumab continues to show the expected treatment effects, and we've seen no safety signal in those studies. We've also initiated some large-scale studies in denosumab, which is just part of the six large-scale clinical trials that were initiated in the second quarter of this year, including the denosumab Phase III skeletal-related event study. Building on the strength of our Phase II studies, which we presented at ASCO, demonstrating that denosumab behaves at least as well as does alendronate in terms of ameliorating the effects of skeletal metastases in tumor settings. We also initiated our heart failure study with Aranesp. We initiated our Enbrel safety registry study. For Vectibix, we initiated our Phase III study in second line colorectal cancer. So let me turn now to slide 26, which is a slide for Vectibix. This slide illustrates the general format for all of the following slides, with respect to research and development. Namely, for each product we will show the indication that's being pursued, the phase of the study, the number of patients that we plan to enroll in the study, and the current status of the study. Then in the last column on the far right, you see the projected data publication date. Now please keep in mind that this is the date at which we expect to be able to present in a scientific forum the data, either through a scientific meeting or through publication of manuscripts. This date does not represent the date at which data will become available. In all cases, where material information becomes available to us, we will make a top-line summary available to you, either through the earnings call or via press releases. So now if we just focus on Vectibix and consider what the important new developments are for the second quarter, I'll say that we completed file submission for Vectibix in the United States, the EU, Australia, Canada, and Switzerland. Our PDUFA date in the United States is September 28th, and we are having frequent interactions with the FDA. We are on track in terms of those interactions, and we are expecting, as I've indicated, that we should be able to launch this molecule before the end of the year. We are very enthusiastic about it. We are really looking forward to completing that negotiation. We also began our second-line colorectal cancer study, a Phase III study involving more than 1,000 patients. That study is enrolling and very shortly we'll to be enrolling our first-line colorectal cancer study. We also have a study ongoing in Japan, which will ultimately permit registration in the Japanese setting, we believe. With respect to adjuvant studies in colorectal cancer, this is part of a cooperative group study. There has been some delay in the initiation of that study because we are still working with the NASBP in terms of study design. We also delayed our head and neck cancer studies, both for locally advanced and for metastatic disease in order to respond to changes in the commercial environment, with the approval of Erbitux in that setting. Lastly with respect to the PACCE study, we continue on track with PACCE in terms of enrollment. And if you'll turn to Page 27, I'll remind you that the PACCE study is a non-registration-enabling study. It's a medical affairs study that was designed to give us information about the behavior of Vectibix in the context of different chemotherapeutic regimens. It's a study that involves enrollment of 1,000 patients, and we now have more than 900 patients enrolled in this study. In fact, we are nearing completion of the oxaliplatin-containing arm. This is a study in which we look at patients who receive chemotherapy with Avastin, plus or minus Vectibix. The primary endpoint of the study is progression-free survival. Those data should be available in the second quarter of '07, as we've said before. We will have response-rate information available from local reads, that is, a not centrally-adjudicated response rate data, but local reads available in the fourth quarter. On our fourth quarter earnings call, we should be able to make that information available to you. But suffice it to say the study is on track, has enrolled well, and we will be completing enrollment in the third quarter. Let me turn now to AMG 706, which you'll recall is a small molecule, multikinase inhibitor which has been studied in a variety of tumor settings. In the second quarter we had the opportunity to review the response-rate data from our Phase II study in gastrointestinal stromal tumors. This is a study in patients who had rigorously failed treatment with high-dose Gleevec. It was a single-arm study, which was intended to support accelerated approval in the event that there was no drug that had already received full approval for this indication. As it happened, in January, Sutent received full approval for this indication. So the study is a Phase II study that is designed to demonstrate efficacy. I'm not going to give you a detailed rundown on the study. I don't want to jeopardize publication of the results. But I'll say we were quite encouraged by the data in terms of response rates and progression-free survival. The median follow-up at the time of analysis was 10 months, and at that point, medium survival had not been reached. So that looks pretty good. This survival data, of course, are preliminary, and the final survival analysis is pending. The data, obviously, will be included in any subsequent filing for AMG 706. Now in the course of evaluating our AMG 706 program and, in particular, in looking at some of the other tumor types where we're treating patients, we have more than 600 patients who have been exposed to this drug. We've noticed that there is what we believe to be, an increased frequency of cholecystitis, that is inflammation of the gall bladder. It's not such an easy thing to call out because cholecystitis is a not infrequent problem in the normal population, as well as in the cancer population. To date, somewhere between 2% to 3% of patients have developed cholecystitis. The number is not precise, because, of course, it's a clinical definition that's based on the physician's assessment of the patient's condition. These events have been managed with standard clinical practice, including cholecystectomy under some circumstances. As we've looked into it, we've decided that our ongoing studies can continue, but we are going to introduce some protocol amendments to ensure that physicians are aware that it is possible that cholecystitis may occur at increased frequency, and that they should manage those appropriately. While we're studying this issue and trying to understand what the true underlying frequency of this adverse effect is, our Phase III studies in first-line breast cancer and first-line non-small cell lung cancer, which we had planned to start in the fourth quarter, will necessarily be delayed. We are currently investigating this phenomenon to gather additional information. We have, of course, been in touch with our investigators, as well as with the FDA. And I think everyone agrees with the plan that we have going forward to understand the risk-benefit ratio for AMG 706. If you turn to Page 30, I want to emphasize that our ongoing studies with AMG 706 are continuing. In the thyroid cancer setting, we did enroll 184 patients in this Phase II study, which is potentially a registration-enabling study in these patients who have failed all prior therapies. We do expect to have data available by the end of the year. We also have initiated our study in the setting of the first-line HER2-negative metastatic breast cancer, head-to-head with Avastin. That study is enrolling, and we are continuing that study, and, of course, following for all safety signals, but, in particular, for the cholecystitis safety signal with AMG 706. So the program is moving along, and we are following this with interest. Turning then to AMG 531, which is, of course, the peptibody that is a thrombopoietin agonist and stimulates platelet production. As you can see on slide 31, we did complete enrollment of our second Phase III study, in immune thrombocytopenic purpura. The database launch for both of these studies will occur in early 2007 and assuming that these data recapitulate what we've seen in our Phase II study and which we've discussed before, should permit filing some time not too long thereafter. We also began studies in chemo and therapy-induced thrombocytopenia in the lymphoma setting and in low or intermediate-risk myelodysplastic syndrome where we're evaluating the ability of AMG 531 to improve platelet counts in this population. So things are moving along according to plan for this important and interesting new program. Now if we can turn to slide 32, we will spend a minute talking about denosumab. As I've told you before, denosumab is by far and away the largest clinical program that we have ever undertaken at Amgen, and the largest program in post-menopausal osteoporosis that anyone has ever undertaken. Our Phase III study, which was enrolled sometime before, is continuing along on track, as are all the other what we call wave-one studies that look at hormone-ablation associated bone loss, as well as prevention of osteoporosis. But in addition, we've begun some other important studies for denosumab. We've begun three different studies looking at the ability of denosumab to delay skeletal-related events in the setting of metastatic cancer, prostate cancer, breast cancer, and in solid tumors. You can see on the slide that there's quite large studies involving 1,400 to 1,700 people. They are being conducted on a worldwide basis. We also began a study last quarter on prevention of prostate metastases. So we are actually looking at the ability to delay the appearance of clinically significant bony metastases in patients with established prostate cancer. We have a Phase II study going on as well in the treatment of multiple myeloma, which should provide some earlier data, and that will be quite interesting. Let me emphasize to you, again, in the far right-hand column, you see projected data publication date, and those dates look like they may be later than what you had imagined previously, but we are not changing these dates from our prior announcements. These dates are the publication dates, not the dates at which time data might be available and we are not delaying our sense of when it might be possible to file the molecule if, in fact, we achieve the kinds of results that we hope to achieve with denosumab. Now, one of the things that we have with denosumab is the opportunity to continue to look at our Phase II dose-ranging study in post-menopausal osteoporosis. On slide 33, you will see some information from that Phase II study. This is looking at three-year data. Again, I have provided only a subset of the data because I don't want to jeopardize publication. What you will see there is increase in lumbar spine bone mineral density as a percent change from base line. Looking at denosumab at the Phase III dose, which is 60 milligrams given subcutaneously every six months, as compared with placebo, which is the line that is trending downwards at the bottom, as you can see that the results between Month 24 and Month 36 -- that's the last year of study for which we obtained results just a few weeks ago -- continue to demonstrate what we had seen before: that administration of denosumab on a once every six month basis continues to improve bone mineral density in the lumbar spine. Alendronate you will see terminates at 24 months, and that was part of the plan of the study is that we have stopped administering alendronate and have followed the decline in bone mineral density that occurs in that setting and also in two of the arms of the study in which denosumab was withdrawn. What I can tell you is that we noted that the effects of denosumab are reversible, that bone turnover increased, and bone mineral density declined. We will have an opportunity to show those in a scientific forum in the not-too-distant future. But suffice it to say that from our perspective, this is important information because it tells us that the bone is awake, it wakes up as the denosumab washes out of the system, and bone turnover then does increase. So that's an important piece of information that we obtained looking at the 36-month data. Again, we will discuss it in more detail in a scientific setting. Turning to slide 34, I will just present some information on Aranesp and on the Sensipar/Mimpara studies. With respect to Aranesp, as I've already indicated, we began our Phase III study in heart failure. This is quite a large study, 3,400 patients, looking batter a composite endpoint for cardiovascular health. That study is now enrolling. We are continuing to enroll the TREAT study. We have, as of the present, more than 2,200 patients enrolled in TREAT. Based on our current enrollment rates, we do expect that enrollment will continue beyond the year end. We had hoped to complete it by the end of the year, but it will take somewhat longer. As turns out, because the event rates are on track, the fact that enrollment will continue beyond the year end does not change our planned timelines with respect to the availability of TREAT data. Keep in mind that this is an event-driven study, and so the patients who are already enrolled in the study contribute disproportionately to the number of events that will be required in order to achieve statistical significance, Asking the question on whether or not administration of Aranesp to improve anemia in the setting of diabetes and renal insufficiency improves outcome. So we do expect to have data available for publication in 2010. Of course, the data themselves will probably be available substantially earlier, based on the event rates. If we look at Sensipar, we did announce that we are beginning our outcomes study for morbidity and mortality, asking the important question of whether managing calcium-phosphate balance and lowering PTH really does have favorable outcomes, as we suspect that it does. A large study of 3,800 patients. The study is just being initiated, and we will begin enrolling patients soon. So that's where we stand in terms of our late-stage development programs, but I can't end my presentation without making a few comments about the presentations provided by Roche at the EDTA meeting on peg-EPO. I think the headline is, as you see on slide 35, that peg-EPO appears to be a fairly typical EPO receptor agonist. Roche reported that peg-EPO has a relatively slow on-rate from a biochemical perspective. That's perhaps not surprising that if one takes erythropoietin and derivitizes it with polyethylene glycol, that diffusion will be limited and the on-rate might, in fact, be slowed, in addition to any stearic hindrance that might occur when peg-EPO binds to the receptor. As a result of that, and because of the fact that the off rate is not affected, peg-EPO looks as if it has inferior binding characteristics. With respect to the Phase III clinical data that were presented, they were able to demonstrate a non-inferiority, but certainly not superiority, as compared with current therapies in measures of hemoglobin stability. I think many of us found it surprising that Roche did not choose to report a complete safety profile for the peg-EPO molecule, particularly since the safety data for their Maxima registration-enabling study for maintenance of hemoglobin were adverse with respect to peg-EPO, as compared to Epoetin. There were no data presented to support the use of peg-EPO to correct anemia in treatment-naive patients, and, certainly, I think everyone is interested to see those data. But on balance, it appears that peg-EPO does not offer any clinical advantage over Aranesp or EPOGEN and of course, that's very important as we look towards the competitive profile in Europe. We do not expect, in the end, that we will be competing with them in the United States for patent reasons. So in summarizing from the perspective of R&D, we now have 42 molecules in our development pipeline. It's by far and away the most robust development pipeline that Amgen has ever had by leagues. Research productivity in the first half of 2006 was very good. Five new molecules were introduced into clinical development, by which I mean they were put into a position where we are going to do formal GLP toxicology and produce enough material to enable us to bring those into the clinic. Two new molecules were actually brought into the clinic. We terminated three programs, representing the typical kind of attrition rate that you expect in small molecule programs. But I want to emphasize that our pipeline continues to reflect a broad range of therapeutic modalities and important disease settings. We are focused on grievous illness, and I think we are making a difference. I will now ask Kevin to summarize.

Thanks, Roger. So let me just summarize a couple of comments before going to Q&A. It, obviously, clearly, was a strong and importantly, a high-quality financial quarter. We had strong execution in the clinic and in the market and in operations across the board. I'm pleased that we continue to see a really meaningful progression in our pipeline. Finally, I just want to reiterate we have a firm conviction that Roche will be judged to infringe our U.S. EPO patents and we are eager to get to the merits of that case in court. We would now like to turn it over to questions and answers; the questions, we will try to provide the answers. Arvind Sood : Gerald, would you go ahead and review the procedure for asking questions, please?

Certainly, sir. (Operator Instructions) Your first question comes from Eric Schmidt with Cowen & Company. Eric Schmidt - Cowen & Co.: Thanks for taking my question. Let's see. I couldn't tell from Richard's comments on the gross margin in the quarter whether the 86% or so margin is expected to continue in the second half of the year at that rate or whether it might decline slightly? Could you clarify, Richard?

You couldn't tell, because I didn't address that directly, Eric. Frankly, I don't think we are going to necessarily talk about the actual gross margin for the rest of the year compared to the second quarter. But I did think I said that it would be lower than prior year's gross margins. So we are going to see continued improvement in gross margins. Whether it's exactly at the second quarter level or not, we are not sure yet, but we will see continued improvement because we do pay lower royalties on several of our products and we are experiencing some production efficiencies, especially with respect to Enbrel. Eric Schmidt - Cowen & Co.: Thank you.

Your next question comes from Elise Wang with Citigroup. Elise Wang - Citigroup: Hi, thanks for taking my questions, and congratulations on a good quarter. Kevin Sharer : Thanks. Elise Wang - Citigroup: What I'm trying to reconcile is, I know that you indicated that sales was really driven by demand this quarter. Clearly with Aranesp in particular. What I'm trying to reconcile that with is with the fact that your revenue guidance, nonetheless, appears to be relatively conservative, even taking into account the trends that you've indicated has occurred historically. So I guess what I'm trying to understand is how do you address that issue since demand does seem to be quite strong, and yet inventory levels seem to be normal, so to speak, and going forward? I am just trying to again, try to reconcile that with your guidance at this point, which does appear to be conservative for the rest of the year.

Thanks, Elise. I think, honestly, we are pleased with the second quarter. We are kind of at plan with respect to our revenues. So the fact that they are high quality and there's no inventory or spill-over is good, but not necessarily indicative of the fact that we see any particular trend breaks versus our plan. I think George commented to you that, I think, all of his products, maybe other than Enbrel, are doing as well as we expected, but they are not doing much, much, much better than we've expected. We haven't seen trend breaks. So we are still at a point where we are having the kind of year we had projected to have from a revenue side. From a profit side it's a little bit more profitable, because the mix is different and the cost of making the product and the royalties are lower. The seasonality we showed you is certainly following through for the first two quarters of the year. We will see what the last two quarters look like. But historically, the last two quarters have had the seasonality as you can see on that chart. George, do you want to add something to that?

Yes, it just seems that the second quarter seems to embody a lot of the growth for the entire year. So we think we are rock solid, but the trends would take us to within our guidance at this point.

Your next question comes from Mark Schoenebaum with Bear Stearns. Mark Schoenebaum - Bear Stearns: Hi, thank you very much for taking my question. Thanks for delivering some much needed good news in our sector. I had a question regarding some of Kevin's earlier comments. I just wanted to ask something a little more specific. You mentioned that CERA, you believe, is simply pegylated EPO. Roche put up a slide out at the Scottish meeting over the weekend showing a molecule that appeared to have three sugars, three glycosylations on it. Do you believe CERA has three sugars on it? Kevin Sharer : We don't want to get into that level of detail. I just want to say that we are convinced that peg-EPO is EPO and it violates a number of our patents. Everything we have seen: Roche's comments, clinical data, everything we see leads us to that conclusion. Obviously, the important judge of that is in Boston, and we will contest it there. But, basically, I think that's the best I can do today, but we have a growing conviction. Mark Schoenebaum - Bear Stearns: May I have a follow-up, since I couldn't get you to answer that one? Kevin Sharer : No. Mark Schoenebaum - Bear Stearns: Okay. Kevin Sharer : Let's give your colleagues -- I sort of answered it, but maybe not exactly the way you wanted it. Mark Schoenebaum - Bear Stearns: No problem, thank you.

Your next question comes from Gene Mack with HSBC Securities. Gene Mack - HSBC: Thanks for taking the question. Actually, on a different topic, the cholecystitis that you are seeing in the GIST trial, are patients symptomology, are they just presenting with really bad abdominal pain or is there any other symptomology that is going on with that?

Gene, I am caveating this by saying that we do not have a detailed understanding. In some respects, it is just alert monitoring by our clinical and safety groups that brought this up, because it is not the sort of thing that would just amazes you. As an example, just to be clear, in our GIST study, we actually did not see a signal at all. Part of the reason that is true is because a lot of those patients with GIST who are heavily pre-treated had already had their gallbladders removed as part of their treatment, so that could have had an effect on it. Again, we have exposed hundreds of patients and what was seen was pretty typical kind of cholecystitis presentation. Patients with right upper-quadrant pain were evaluated. In some cases, they had associated diarrhea or other gastrointestinal symptoms, nausea. In some cases, they had had a little bit of a fever. Some of those patients went on to have their gallbladders removed, cholecystectomy. The determination of cholecystitis is strictly made by the physician involved in caring for the patient so that is the definition, if you will, of cholecystitis. We are going back and looking at everything that we can to understand what is the nature of the signal here, how strong is it, et cetera. What typically occurs in these patient populations, because again, cholecystitis is a not-infrequent condition in the general population and in patients who have cancer. Gene Mack - HSBC: Thank you.

Your next question comes from Joel Sendek with Lazard Capital Markets. Joel Sendek - Lazard Capital Market: I have another gross margin question. I am wondering, versus the first quarter this year, whether the improvements were due to any reaching of any sales thresholds or anything like that. If you can comment on that, and as to whether the improvements we saw in the second quarter are permanent, or how many of them are, if they are? Thank you.

I think our assumption is the bulk of the improvement in the second quarter on the cost of sales lines are pretty continuing. The royalties that we do not have to pay anymore on Neupogen, Neulasta obviously stopped at the end of last year. The royalties that we bought out so far towards the end of last year, early this year are obviously gone. We have begun to make the product more cheaply in all the places that we make Enbrel. We are doing well on our other products too, as Kevin said, to congratulate our operations folks. That is why we are comfortable to say for the remainder of the year, we expect cost of sales to be lower than last year. We are doing well on the cost of sales line.

Your next question comes from Steven Harr of Morgan Stanley. Steven Harr - Morgan Stanley Dean Witter : Roger, is it possible to give us some additional granularity on the rate of acceleration of bone loss after the cessation of treatment with denosumab? Is it the normal rate or is it an accelerated bone loss?

Steven, I think what I would say is that we are still in the analytical phase of this. One of the things we need to do is see more than one year after discontinuation. But if you look at the data from estrogen or selective estrogen receptor modulators, or from intermittent PTH, what you see is that when you withdraw a drug, there is a decline in bone mineral density that occurs back down to base line over a period of time. That is the kind of thing that we are seeing when we discontinued denosumab.We are following it. We like the fact that it indicates that the bone is awake and turns over, from the perspective of any concerns one might have about frozen-bone type syndromes. On the other hand, we want to make sure that we evaluate it and that there is not excessive bone loss. It looks very much like what you see with estrogen receptor modulator. Steven Harr - Morgan Stanley Dean Witter : No changes in minerals?

I'm sorry, this is bone mineral density. Steven Harr - Morgan Stanley Dean Witter : I am sorry, but no changes in serum electrolytes, like calcium.

Oh, no. So far, we have not seen anything that concerns us from the standpoint of safety tolerability.

Your next question comes from Geoff Porges with Sanford Bernstein. Geoffrey Porges - Sanford C. Bernstein & Company, Inc.: Related to the side effects issues, Roger, could you comment on the 706, whether you are seeing the cholecystitis in the background of the typical multikinase inhibitor profile, hypertension, thrombosis, and that sort of thing? Could you answer whether there has been any osteonecrosis of the jaw in the extended dosing with denosumab?

With respect to 706, we are seeing the typical kinds of mechanism-based hypertension -- in particular, hypertension. Each one of these molecules is going to look a little bit different, I think. We have not seen any change in that. We are still digging through the issue of whether there is a correlation between adverse experiences which might well be mechanism-based, like hypertension, and the frequency of cholecystitis. Frankly, there really are not enough cases and there is not enough data to get that analysis done, but we are pouring through it. The reason I have stressed it here, I certainly want people to be aware of it. That is the way we have treated it in terms of our discussion with physicians and with the FDA. I think everyone agrees in terms of the way forward for 706. As regards to osteonecrosis of the jaw, we have not seen a case of osteonecrosis of the jaw in the extended studies we could talk about. We have one example of a case of osteonecrosis of the jaw that occurred in a patient, just a very short period, two weeks after treatment with denosumab, who had prior treatment with zoledronate for a long period of time. At this point, it was considered to be a zoledronate problem. That is the only example. There is one case. Geoffrey Porges - Sanford C. Bernstein & Company, Inc.: Thanks very much. That is helpful.

(Operator Instructions) Gentlemen, your next question comes from George Farmer of Wachovia Securities. George Farmer - Wachovia Securities: Thanks for taking my question. Back to 706 -- Roger, could you comment as to whether any of the other VEGFR kinase inhibitors may have shown similar problems that resemble cholecystitis, or maybe even just generally upper-quadrant pain?

George, first of all, these are patients in many cases with gastrointestinal malignancies. They have a lot of abdominal pain. Because cholecystitis is not at all unknown, it is the sort of thing that might be occurring anyway and would be difficult to pick up. I give a lot of credit to our clinical and safety groups for having noted this signal, because there are a lot of people who have abdominal pain and cholecystectomies occur in this setting with some established frequency. We do not know whether there are other multikinase inhibitors that have this kind of effect. Over time, I guess we will all find out. George Farmer - Wachovia Securities: Thank you.

Your next question comes from Jennifer Chao of Deutsche Bank. Jennifer Chao - Deutsche Bank: Thanks for taking the question. The first is, the question really is with respect to Aranesp patient market share, could you discuss what patient market share percentages are in your estimation versus J&J in the oncology setting, and whether we are seeing more meaningful penetration of Aranesp into the pre-dialysis market? Thank you.

Jennifer, the second part of your question was..? Jennifer Chao - Deutsche Bank: Just whether we are seeing more meaningful penetration into the CKD market, whether you are seeing some interesting trends there?

I think that market is still growing. We would like it to grow a lot faster, and that is the reason we embarked on the TREAT trial. We have to create a larger clinical imperative for doctors to treat more and more of the patients and actually get more referrals from primary care land. Jennifer Chao - Deutsche Bank: Thank you.

Your next question comes from Shiv Kapoor of Montgomery & Company. Shiv Kapoor - Montgomery & Co.: Thanks for taking my question, and congratulations on a great quarter. It is a strategy question that I have. As we traverse through what seems like low sentiment in the industry, is there an acceleration in your acquisition strategy? Are there specific technology or disease areas that you are more interested in now than before?

We have a consistent view in terms of what we call outreach -- which is acquisition in-licensing partnering -- that is always to get product candidates outside the company. We are consistent. Dips in the market do not change our strategy there. We have historically gotten about half of our molecules from outside the company one way or another. I think that is a sign of health. We aggressively look outside the company. From time-to-time, we in-license, sometimes we acquire. We look for high-quality opportunities. Those are not usually inexpensive, but we are going for innovative medicines to treat grievous illness. We have been on the look-out for those for a long time. We still are. We are not trying to look for a particular technology or a particular therapeutic area. We just want good molecules that are innovative, will treat grievous illness, and be competitively successful. I guess the short answer to your question is no, the current market conditions have not changed our strategy. We remain alert and hungry and aggressive. Shiv Kapoor - Montgomery & Co.: Thank you.

Your next question comes from Geoff Meacham of JPMorgan. Geoffrey Meacham - JPMorgan Chase & Co.: Thanks for taking the question. A question on Enbrel. Last quarter, you talked a little bit about competitor discounting. I am wondering if that has continued, or that slowed down the second quarter, and if that was a key driver of the strong sequential growth that we saw.

Actually, last quarter we talked about it happening in the fourth quarter of '05 and it was Remicade. They juiced up their contracts, probably in advance of the new inject-able products coming onto market. To my knowledge, that has not continued and I do not think it is a factor today in the marketplace.

Your next question comes from May-Kin Ho of Goldman Sachs. May-Kin Ho - Goldman Sachs: Roger, a question again on 706. Do you know what the mechanism is there for the cholecystitis?

May-Kin, no, we do not. We have not established at all what is going on there. It is not, as you will appreciate, it is not an easy thing to do. Right at the moment, we saw the signal in our clinical trial database. We were digging into each case, one after another, trying to understand what is going on in this population. So more to come. May-Kin Ho - Goldman Sachs: In animals, you did not see any signal at all?

No, in animals, we have gone back over the data in animals and again, there are not really good animal models, because spontaneous cholecystitis is pretty common in the human population but is not a feature of most of the animals that we use in our safety study. We are now looking back at some of the long-term exposures at high dose to see whether or not there is any relevant information there but it is still early days. May-Kin Ho - Goldman Sachs: Thank you.

Your next question comes from David Witzke of Banc of America Securities. David Witzke - Banc of America Securities: Good afternoon. Any update on how you are thinking of pricing Pmab? Do you think you will have the PACCE data necessary to make a decision, possibly to price lower and go for the first-line colorectal market?

Yes, we are going to probably talk about pricing when we talk about our overall launch strategy on the next conference call. I do not think we need PACCE data necessarily to price the product at this point. David Witzke - Banc of America Securities: Thank you.

Your next question comes from Bill Tanner of Leerink Swann. Bill Tanner - Leerink Swann: Thanks for taking the question. For you maybe, Roger -- you mentioned CERA has a relatively slow on-rate. Could you remind us what the difference in the on-rate between darbepoetin is in alfa? How much does that actually then contribute perhaps to the ability to dose more infrequently as opposed to the glycosylation to increase half life?

The affinity of Aranesp for the receptor as compared to erythropoietin is not the determination of the dosing frequency, so the dosing frequency is a function of the half-life in the circulation. The affinity of binding to the receptor is determined by the off-rate, and Aranesp is a less-avid binder of the receptor than is erythropoietin, but that lower avidity of binding is more than compensated for by the fact that it has a long dwell time in the circulation. That was the basis of the initial studies that we did on Aranesp, and the basis of its favorable effects. I think the argument had been made by Roche in the past that there was some very different mechanism of binding of a peg-EPO to the receptor and they seem now, from the EDTA presentation, they seem to have backed away from that.

Your next question comes from Stanley Grossman of Rudman Capital. Stanley Grossman - Rudman Capital: Thanks for taking my call, and it is a great quarter. Congratulations. The one question I had is, Roger, I think you mentioned in the first quarter call that you might take a peak at the SRE reduction in metastatic lesions, bone in denosumab. Do you plan to do that at all, because I think there must be a marked reduction in radiation treatments and also in pathological fractures?

Stanley, we have three ongoing Phase III studies in patients with metastatic bone disease that we started this quarter, where we are looking at skeletal-related events with denosumab. Those studies should provide us with the information that would let us know whether or not we are reducing those skeletal-related events. I will remind you that when we looked at those data in Phase II settings in breast cancer patients with metastatic bone disease, we found that denosumab treatment did in fact lower skeletal-related events. Denosumab did also improve bone mineral density, even in those patients who had progressed on zoledronate. We are very enthusiastic about the potential for denosumab in the setting of oncologic bone disease. Stanley Grossman - Rudman Capital: Right. Thank you very much.

Your next question comes from Mark Schoenebaum of Bear Stearns. Mark Schoenebaum - Bear, Stearns & Co.: Thanks a lot for taking the follow-up. I was just reviewing the slides. I noticed on the denosumab slide that you, I was not aware of this, you have started a trial head-to-hand against alendronate?

Yes. Mark Schoenebaum - Bear, Stearns & Co.: Can you give us some information about that trial? When you actually expect the data versus publication? Can you tell us whether it is a superiority trial, please?

We are in fact doing, as we had indicated that we would, head-to-head study versus alendronate for bone mineral density. We expect that we can demonstrate in certain settings, as we have shown in our Phase II study, superiority with respect to that, but we do not expect that we would be able to demonstrate superiority with respect to fracture rates, because you would have to do enormous-sized studies in order to do that. In terms of treating post-menopausal osteoporosis for bone mineral density, that is the basis of that study. Of course, we will have data in advance of the publication date, but it is kind of a function of enrollment. It is early days yet, Mark. Mark Schoenebaum - Bear, Stearns & Co.: Do you expect that to be in the label?

Really, it is a function of how we do it with respect to enrollment and what sorts of data we see. Customarily, you would need multiple studies in order to have a strong case from the standpoint of labeling. We will just wait to see what it looks like. Mark Schoenebaum - Bear, Stearns & Co.: Thanks.

Gerald, as we are going on about 3:15 west coast time, maybe we can take one more question, please?

Certainly, sir. Your final question is a follow-up from Geoff Meacham of JPMorgan. Geoffrey Meacham - JPMorgan Chase & Co.: Thanks for taking the follow-up. I hate to beat a dead horse on 706, but is the delay in the Phase III starts due to more exposure or more so from guidance from the FDA? Can you help us out with that?

Geoff, there has been no guidance from the FDA on that. It is our wanting to understand what the basis is of this safety signal and whether there are some things that we should do differently in terms of our clinical study. I have to emphasize to everyone that we just sort of found this out, and so we are still sifting through this analysis. We decided it was best to let you know that we plan to delay these studies until we understand better what we are dealing with in the Phase II environment before we embark on a large Phase III trial.

With that, I would like to thank everybody for their participation in our call. If have you any follow-on questions, comments, thoughts, whatever, the Investor Relations team is going to be back in their respective offices shortly so please feel free to call us. Thanks again.

Ladies and gentlemen, we appreciate your participation in Amgen's second quarter financial results 2006 conference call. You may all now disconnect.