Good morning, and good afternoon. Welcome, ladies and gentlemen, to AstraZeneca's Q1 2019 Results Conference Call and Webcast for Investors and Analysts. Before I hand over the call to Pascal Soriot at AstraZeneca, I'd like to read the safe harbor statement. The company intends to utilize the Safe Harbor provisions of the United States Private Securities Litigation Reform Act of 1995. Participants on this call may make forward-looking statements with respect to the operations and financial performance of AstraZeneca. By their very nature, forward-looking statements involve risk and uncertainty, and results may differ materially from those expressed or implied by these forward-looking statements. The company undertakes no obligation to update forward-looking statements. There will be an opportunity to ask questions after today's presentations. [Operator Instructions]. We will now hand you over to AstraZeneca where the call is about to start.

Hello, everyone. It's Pascal Soriot here. Welcome to the first quarter 2019 conference call and our webcast for investors and analysts. We're in London today for the Annual General Meeting that will take place this afternoon. The presentation, as always, is available on astrazeneca.com, and we've also sent it to those on our distribution list. Please turn to Slide 2. This is our usual safe harbor statement. We'll be making comments on our financial performance using core reporting numbers and at constant exchange rates, CER, which are both non-GAAP measures. We'll also discuss other non-GAAP measures deemed helpful for investors and analysts. All numbers we refer to million U.S. dollars, and growth rates will be at CER and for the first quarter of 2019, unless we state otherwise. Please turn to Slide 3. We plan to spend a good half hour on the presentation and then go to Q&A. [Operator Instructions]. Thank you so much for your help with this. And today, I'm joined by Dave Fredrickson, our EVP for oncology; Ruud Dobber, our EVP for Biopharmaceuticals; Marc Dunoyer, our CFO; Mene Pangalos, who's our EVP for R&D BioPharmaceuticals; and José Baselga, our EVP for R&D Oncology. Also very pleased to have Susan Galbraith with us today. And Susan, as you know, is our SVP of R&D Oncology. She will join us for the Q&A. Please turn to Slide 4. This is that the agenda we plan to cover all key aspects of our results today. So if we move to Slide 5. In March 2013, we launched a new strategy for AstraZeneca where we outlined the plan for achieving scientific leadership, returning the company to growth and being a great place to work for all our colleagues around the world. We have more than 60,000 colleagues working for our…

Great. Thank you so much, Pascal, and I think that is a nice segue to bridge from a conversation about the oncology strategy to the performance within the first quarter. After I do that, I'll hand over to Ruud Dobber, who'll give an update on CVRM, respiratory and Emerging Markets. Please turn to Slide 13. 2019 has really started off well for oncology. Sales of $1.9 billion in the quarter represented 59% growth, and the 4 new medicines contributed $700 million of incremental sales. In the lung cancer franchise, Tagrisso and Imfinzi continued their launches in the new indications of first-line EGFR mutated non-small lung cancer and also within unresectable Stage III non-small cell lung cancer. Lynparza continued to cement itself as the leading PARP inhibitor with the first-line ovarian cancer launch now under way and expanding across the globe. We continue to see encouraging uptake of Calquence in the smaller mantle cell lymphoma indication now with sales of $29 million in the quarter. The majority of sales is coming from the approved indication in the U.S. where we estimate as many as 40% of the patients are being treated with Calquence. And we saw the majority of use in patients naïve to BTKi inhibitors. This launch allowed us to build the infrastructure as we prepare for the larger chronic lymphocytic leukemia indication where we have 2 pivotal Phase III readouts coming in the second half of the year. And finally, while it's not mentioned on the slide, I do want to note that Faslodex held ground, with sales of $254 million in the quarter and growth of 4% mainly driven by performance in Emerging Markets that were up 28%. And we do note that we anticipate to see the first generic fulvestrant in the U.S. market relatively soon. With…

Thank you so much, Dave. Today, I'm here to talk to you about the BioPharmaceuticals business. Total sales of the 2 therapy areas was $2.3 billion in the quarter, growing at 16%. We're very pleased with the continued growth of Farxiga and Brilinta and the ongoing successful launch of Fasenra. Symbicort and Pulmicort also continued to deliver strong sales. We look to build on this growth, including through further launches for Lokelma during 2019. Please turn to Slide 18. Moving to new CVRM. Sales were up by 19% despite intense competition in diabetes with total first quarter sales at $1 billion. Growth for both Farxiga and Brilinta remained strong with double-digit increases globally. Farxiga delivered sales of $349 million in the quarter with 23% growth maintaining volume market share leadership globally. Farxiga saw U.S. growth of 3% in the quarter with growing slowing due to increased competition and formulary plant changes taking effect in the quarter. Outside the U.S., where we have 62% of sales, we have seen encouraging performances with increasing volume growth. Sales in Europe were up by 30%, and the emerging market sales were up by 51%. Brilinta delivered sales of $348 million with 24% growth driven by strong performance in Emerging Markets, up by 38%. We also have continuous growth in the U.S. and Europe, up by 33% and 3%, respectively. European underlying demand growth was high single digit despite some negative inventory impact. We continue to be very pleased with the performance of Brilinta, which is still outgrowing the market in all regions. Bydureon, including the autoinjector Bydureon BCise, continued to perform well with sales up 4% despite the impact of supply constraints for the new BCise device. We are on track to resolve this throughout the year and be back to normal supply in…

Thank you, Ruud, and hello, everyone. I want to take you through our financial performance in the quarter as well as our financial priorities and guidance. Could you please turn to Slide 22. As usual, I will begin by the reported P&L before turning to the core performance. As Pascal mentioned earlier, product sales grew by 14% while there was minimal collaboration revenue in this quarter. Other operating income of around $600 million included the impact -- including the impact of divestment of right to synergies for the United State. Please turn to Slide 23. Moving to the core P&L. Our gross margin ratio improved by over 2 percentage points to 80.5%, reflecting the phasing of our mix of sale. Operating cost, which increased by 5% in the quarter, represented 61% of total revenue, which was a 4% point reduction versus the first quarter of last year. With product sales growth ahead of operating cost, we are beginning to deliver operating leverage, and our core operating profit margin reached 30% supported by other operating income. The core tax rate was 23% impacted by the geographical mix of profits and disposals. Despite the higher-than-average core tax rate, our core earnings per share doubled to $0.89 underpinned by top line growth and cost leverage. Please turn to Slide 24. This is a slide I first showed you at the full year results in February, and I will use it going forward to demonstrate our progress. It describes our financial journey and what our priorities are. We are delivering strong and sustainable sales growth, driving operating leverage and improving on margin and profitability, but the story will not stop there. Our focus will be on taking this down to generate, in 2020, more cash so that we can direct it to deleveraging our…

Please turn to Slide 26. Now thank you, Marc. Good afternoon, everyone. I'm delighted to be here today to provide an update on the progress in the pipeline since our last announcement. I'm also joined today by my counterpart, José Baselga, who'll discuss the oncology pipeline and our upcoming news flow. Firstly, and this is something I'm very passionate about, I'd like to take a moment to speak about the improvements in our R&D productivity over the past few years. Please turn to Slide 27. As you know, we're committed to the productivity and quality of our science across the portfolio, and I just want to highlight a few examples of the progress that we've made over the past 5 years in our attempt to bring new and better medicines to patients. First is the number of accepted high-impact publications have increased tenfold to over 100 in 2018. This, as you know, is a clear measure of the high quality of sciences underpinning the research and development in our organization. Secondly, the number of Phase II projects has increased by 1/3. And with this, we've also achieved 30 projects and validated proof of mechanism in the same period. Again, this is a demonstration of the quality of molecules we're putting into the clinic, which improves the probability of transitioning them from early development to late. Finally, we've received more than 50 regulatory designations in major markets. And I'm very proud of what we achieved in 2018. It was a record-breaking year with both new molecular entities and major life-cycle programs, seeing 23 major market approvals in total, the most ever attained in 1 year in AstraZeneca's 20-year history. Please turn to Slide 28. In respiratory, our partner, Circassia, received U.S. regulatory approval for Duaklir, a medicine to the treatment of…

Thank you, José. So in the interest of saving time for Q&A, I'll skip the last slide and just remind you product sales grew by 14%, and operating profit is up 96% with an operating margin of 30%. We reconfirmed our guidance for the year, and our pipeline is making good progress. So with that, we'll now move to the Q&A. [Operator Instructions]. We'll also take written questions from the webcast. [Operator Instructions]. Thank you, and we'll now get started with Richard Parkes at Deutsche Bank. Richard, over to you.

Since I'm first, I'll ask the inevitable cash flow question, and it's got two parts, unfortunately. So obviously, the decision to raise equity as part of the Daiichi collaboration is raise overall focus cash flow and your balance sheet, and I think that was triggered by the decision to raise more cash downs required for the initial upfront and milestones. So I just wondered if you could talk about that decision. And was that driven by the fact that the rating agencies take into account the longer-term milestones with them looking at your rating? Or was it because you're seeing these additional payments from legacy business development that is impacting cash flow in the short term? So That's the first part. The second part is just on the cash flow in the first quarter. You've given some clarity on that in the release. Obviously, it's still a significant negative in the first quarter when you look at working cap short-term provisions and noncash movements of more than $1 billion negative. Can you help us to understand those and get a sense of how those will develop for the full year?

Thanks, Richard. So I think that the line was not really good, but hopefully, we captured the two parts of your question. The first will relate to the equity raise, and the second relates to the cash flow itself and some of the negative movements in the cash flow statement. Marc, do you want to cover both of those?

Yes. Let me start. Richard, thank you very much for these two informative questions. So the first one on the equity. So we -- as we have said, the equity was -- the purpose of the equity was to pay for the upfront and the milestone in the early years for the Daiichi Sankyo product right acquisition as well as to strengthen our balance sheet as we have a repayment of $1 billion bond in September of this year. So this is exactly the way we explained it. Now turning to your question on the credit rating agencies. They tend to look at the cash flows and other -- a lot of financial metrics usually within a 12 to 24 months horizon. So they are informed and well aware of our 2019 cash flow progression. You will certainly remember that at the full year results in February of this year, I have flagged that in 2019, we would have larger proportion of business development payments to be made. Some of them linked to deals that we have done recently, but some of those linked to the deals that are much older. In particular, we have had a substantial payment for a true-up of a joint venture with Merck in the United States that started in 1988 for Nexium. So some of them are recent. Some of them are more historical. So that, I think, answers the situation of why the credit rating agencies were looking at our credit rating with some level of scrutiny. Turning to the current quarter cash flow and what can we learn from the quarter cash flow for the full year. So first of all, we have to be very much aware that quarter-to-quarter cash flow comparison are very delicate. You could have in 1 quarter a positive, and you can have similar quarter the following year a negative. And of course, this exacerbates the difference. And I think the quarter 1 of 2019 is a very good example. We have, as I've just described, a true-up with Merck for historical joint venture. This was a negative of $400 million. In 2018, we had a positive for litigations that we have settled. We -- so we have one-off on one side, one-off on the other side. Since we have both of about $400 million each, so you have a variation in that quarter of $800 million, which of course is not reflective of the progression of profitability. So this is why quarter-to-quarter comparison are very delicate. Now if you go for the full year, this is a bit less variable, but still, you could have some variability. So we have said that in 2019, we will have more CD payment than we have had in the past. And therefore, we said that the results of cash flow for 2019 will not be as good as they have been in 2018, and I think I can confirm again the same for that.

So we'll move to Mark Purcell at Morgan Stanley. Mark, over to you.

As part of the ongoing U.S.-China trade talks, it's been reported that China's offered U.S. pharma companies 8 years of regulatory data protection in China for the biologics they develop. Pascal, what are your thoughts on this and other potential offsets to the political agenda, which is driving earlier and more rapid uptake of innovative medicines in China? And if I may, just a point of clarification on the last question. Could you tell us which year the dividend is going to be covered by cash flow before financing activities and which year is going to mark the start of a sustained improvement in your net cash position before M&A and internalization consideration? So just turning to my last question, an early comment that you made at the beginning of the call.

Okay. Thanks, Mark. So the first question as it relates to China, I will really start by saying that everything we've seen happening over the last year or 2 in China goes in the right direction, quite frankly. I mean -- and I could list, for instance -- the government is really protecting -- going in the direction of protecting IPO, respecting IPO rights at least in our industry. There is an emergence of an innovative industry in China, and of course, that's something that they can only develop and flourish with an IP system in place. So clearly, China is focused on IP rights. Now we've seen acceleration of review and approval of new, innovative medicines, which is a real positive. We've seen funding being deployed to reimbursing and facilitating access to new, innovative medicines like Tagrisso; Farxiga which, quite frankly, not long ago, nobody would have thought would be reimbursed. On the other side, we've seen -- of course, we've seen the emergence of generics and the price pressure on older products. So you can see China transitioning to a model that is more aligned with what we see in the Western world and driven by innovation, quite simply. We've seen that coming since quite some time, and we spend all our time and effort preparing for it and accelerating the development of our new products. Now as it relates to these ongoing discussions between the U.S. and China, I -- and I would think that it is a little bit early to comment. I think we should wait until the discussions conclude. But again, I think the fact that we are debating 8 years or 10 years or more is actually reflecting that there is a discussion around IP rights, which is a good one, good discussion to have. I will only say that the gold standard really for biologics protection is 12 years. International norm has started to establish itself around 10 years. That's what you see in this recent agreement between the U.S. and Canada and Mexico. So I can only say that I would hope that China and the U.S. can agree on 10 years, but we'll have to wait until the end of this discussion. But really, I think I would look at it with a positive approach really. I mean the discussion reflects really good progress on IP protection. So dividend, Marc, do you want to cover that?

Yes. So I did not understand your question very clearly, but I assume that you wanted to ask about the payment of the dividend in 2020.

Yes. Sorry. I would just repeat it. The question was just based on what Richard was asking, some of the consensus -- the -- which year will the dividend be covered by cash flow before financing activities? And which year will mark the start of a sustained improvement in your net cash position before considerations around M&A and internalization?

Very good. So I think the -- so first of all, one remark on the underlying trends. Obviously, the continuous margin expansion and operating leverage will lead to a better cash generation from 2020. So as we -- regarding the share issuance and the relationship with the Daiichi Sankyo investment, we have said that the share issuance was going to cover the upfront and the milestone in the early years. So if we exclude the impact of Daiichi Sankyo, we anticipate to reach a coverage of our dividend in 2020. So first. And then the job will not be finished. We will continue to work on increasing profitability, increasing operating leverage and margin expansion from 2020 onwards.

Thanks, Marc. I mean I think Mark, we can only -- Mark Purcell, that is, we can only reconfirm what we've told you before. I mean there's no change to what we've said. We will basically cover the dividend from 2020, start reducing that from 2021. And as Marc Dunoyer that he said a minute ago, essentially, the equity issuance will enable us to cover the cash outlays related to this DS-8201, which again I think people will learn to appreciate the significance of this asset and its enormous potential. So let's move to Andrew Baum at Citi. Andrew, over to you.

First question relates to a topic you've already touched on in AstraZeneca's success in building a very sizable pipeline. So by your own admission, I think you call out 157 projects. I'm counting around 133 for Roche. You spend about $5 billion per annum. They spend $10 billion. Without getting too hang up on the absolute numbers and talking about probability of success, I guess what I'm getting at is, could you give us some indication on the rate of increase for R&D spend given the opportunities in front of you? And then related to that, could you talk about big step to which there can be much further prioritization of your portfolio, including throughout licensing, especially now given the integration of MedImmune? And then separately, if Susan is on the line or to José, there's now been two sets of data with PARP inhibitors showing that prostate cancer having ATM mutations are not PARP-responsive, and this is obviously part of the population for the primary endpoint of your PROfound trial, which reports at the end of this year. How do you think about the risk sparing and the size of the market in the event this pans out within that trial? And apologies for the second question.

Thanks, Andrew. That's great -- two great questions. So maybe what I would propose, Mene, if you could cover the first one, and then if José [indiscernible] the mother of this great medicine, Lynparza, to answer the second question, if that's okay. So Mene, do you want to cover the...

Yes. So size of the pipeline, so as you articulate, we do have a rich array of molecules across our therapy areas in early to mid-stage, and I think the key actually is generating confidence in the dataset that enable you to be comfortable and confident moving in today's stage development world disease, the big investments you made. So I think rigorous prioritization, which is something that we're actively doing on a regular basis, and then making sure that we ask really good questions in early development to demonstrate proof of mechanism, early signs of efficacy so that when we move things into Phase III, they have a high probability of successfully launching the medicines. I think is key, and it's a much better place to be than having not enough molecules in your pipeline, where you're actually just moving in stores because as we can. And then the surplus molecules that we have, if there are things of a low priority, Andrew, absolutely, we'd outlicense them or work out how to partner them in a way that gets them to patients as fast as possible.

And we're on this process, Andrew, of prioritizing a few mid-stage projects and accelerating them to the extent we can and doing this prioritization and selection of the most promising project. But I think that, as Mene said, we should keep in mind, our productivity has increased. And we have been able with the low R&D budget to deliver a very rich pipeline relative to some of our competitors. But still, we need to continue prioritizing. Lynparza, Susan?

Okay. Thanks, Andrew for the question. So the one thing to note is there will be more data that's coming out from expanded dataset from [indiscernible] at ASCO coming up. So look for that from Johann de Bono's group. Clearly, across the panel of genes that can produce homologous repair recombination and deficiency, there is some variation in the relative sensitivity to PARP inhibition. Nonetheless, we've got very good strong proof of principle from the data in the image indiscernible] that patients with ATM loss do have activity as being with long and durable responses were the last group treatment. So one thing to bear in mind is the testing for ATM. As with all of these big genes, I think there's probably increased sensitivity when you've got biallelic loss rather than just monoallelic loss. I think still getting the testing right is important. We've taken that consideration into account where we've designed the studies that we have ongoing. So we remain confident that the study is designed well to look for the size of benefit that has been out to detect across the patients that were enrolled.

Thank you. So thanks, Susan. So let's move to Naresh Chouhan at Intrinsic Health, sorry. Naresh has a question that is asked online. I think it's for you, Dave. And the question -- I will read the question so everybody can benefit from it. And it is, please can you help us to understand the slow down in Tagrisso? I guess the question relates to the U.S. mainly. Underlying demand only grew mid-single digits quarter-on-quarter, but penetration is only 60%. Some discussion about where you expect penetration to peak out would be helpful.

Thanks, Pascal, and thanks for the question. I think, first, to start with, so I think you're pointing out that in the U.S. that there was a decline in the sequential growth quarter-over-quarter, and you think that it's important to point out that obviously the year-over-year growth of Tagrisso has been quite tremendous. And I think that given where we are in the launch of Tagrisso, we're quite pleased with mid- to high single-digit demand. We look at the demand as the primary measure of the health of the brand. And at the end of the year, there are always inventory movements and one-offs. And you'll recall that we have described that as happening in the fourth quarter last year within China, and we saw China rebound back quite nicely, as we said that it would in the first quarter this year. I think the other piece that I'd like to highlight here is just that I think what you see is also the strength of the diversity of the geographies that we operate in. And again here, you saw sequential growth globally of 6% for Tagrisso. And I've been working in oncology a long time. You see these S curves, where the goal is you get as rapidly up the adoption curve as possible in various regions. And you know that U.S. will be first. And usually, Japan and Europe follow on the heels of that, and then China comes after that. So on your question about what's the top penetration that we can expect, TKI as a class have about 80% to 85% penetration of the frontline market space. There are remarkably still patients despite TKIs having been in the market for decade that are still getting chemotherapy or even I/O chemo as their first therapy. That's part of our educational effort, but that's what I'd look at in terms of what the maximum potential of the class is.

Thank you, Dave. So Jo Walton at Crédit Suisse, please. Next in line. Jo, go ahead.

I wonder if I could just return to China and get a little bit more information about it. Sanofi pointed in their strong results in China today to a pull-through of demand into 1Q that would otherwise have been in 2Q because of the change to value-based pricing. I wonder if you felt that there was any distortion in your numbers. And then as you reflect on how things go going forwards, should we see -- are there any particular products where we should see a decline in growth? And I wonder if you could also help us by giving us the absolute sales of Tagrisso in China.

Thanks, Jo. So I'll ask Dave if he -- if you have the number to answer the second question. The first question is clearly no. We haven't had any effect, as you described that Sanofi mentioned. In fact, I would say it was the other way around as we are concerned. We had an inventory decrease in Q1. And at the end of Q1, trade inventory is frankly the lowest it can get to. I mean we couldn't get lower, or we would have supply issues. So we have very low inventory in the trade in China. We haven't seen that effect. In fact, again, we have had an inventory decline. So the Tagrisso question in China, Dave, do you...

Yes. So Jo, we don't share the specific sales levels for individual brands within China. What I can say is that we certainly saw a very nice increase in oncology sales in the emerging markets, oncology sales growth of 43%, and Tagrisso was a significant contributor to that.

Thanks, Dave. So Keyur Parekh. Keyur, go ahead.

Two questions, please. The first one, coming back to cash flow for Marc. Marc, when you talk about covering the dividend in 2020 from operational cash flow, is that pre- or post-CapEx annual royalty payments to Bristol and [indiscernible] and things like that? Just linked with that, how -- what is it that is likely to change so dramatically over -- between kind of the cash you're generating today versus 12 months' time? Because inherently, what you're talking about is an increase in quarterly cash flow generation of about $1 billion. So just if you can help us bridge the gap, that would be great. And then secondly, Dave, to -- just a clarification on your comment regarding the Imfinzi growth outlook. I just want to make sure I heard you right in saying you -- we should be thinking about Imfinzi growth in the near term as being driven by ex U.S. rather than the U.S.?

So on the cash flow, so what I just said a few minutes ago that since we have used the share issuance to cover the upfront for -- the upfront and early milestones for Daiichi Sankyo, I'm looking at the coverage of the dividend excluding the impact of the Daiichi Sankyo product acquisition. So you need to remember this caveat. So the coverage of the dividend will be done after CapEx, after restructuring expenses, after the normal outflows, but excluding the impact of the cash payments related to Daiichi Sankyo. I'm sorry. And then why is it -- I guess why is it getting better? Well, I think I have also outlined that for 2019, we have reasonably high level of payments for business development deals. Some of them, we have done in 2018. Some of them, we have done much earlier. And I mentioned the true-up joint venture with -- historical joint venture with Merck in the U.S. But also we have some molecules which are progressing for the development momentum, and hopefully some will be approved. And there are milestones, which are accompanying this regulatory approvals in the course of 2019. So I see it as a good news in a way. Because to me, these products will reach market expansion. So two effects really. Okay, I mean the obvious one is the improvement of our operating margin that goes down to the cash flow. And the second, as Marc said, some of these one-off payments will decline. And remember, in Q1 of this year, we had a big one, $400 million relating to next year which is 1-year, 30-year-old joint venture. And the final milestone relating to the undoing of this JV. I think there was a question on Tagrisso for you, yes?

So it's on Imfinzi actually. So the payer -- the first part, we -- I certainly have the expectation for continued growth within the U.S, but I do think that growth rate on a percentage basis sequentially is going to be lower than what we've seen. We saw a fairly significant catalyst with the overall survival data at ESMO, and we saw a sequential growth rate of 27% on the heels of that within the U.S. And we saw a lower rate in the mid-single digits in terms of the growth rate that we saw at 7% in terms of sales in the U.S. We see CRT rates now of really well above 50%, probably getting close to 60% within that. We think that there's some opportunity for continued growth there, but we're also under the belief and we listen to what divisions are speaking through. I think that there are patients for whom CRT just isn't a viable option, and so that's probably getting close to near where its max is. Again, we also see opportunity for continuing to have more patients treated with Imfinzi post-CRT, but we have a greater opportunity to get to U.S. levels of performance outside of the U.S. We have just recently launched within Japan, where we saw really strong uptake in the quarter. And we're encouraged by the trajectory that will come in to Q2 on. In Europe, recent approval, and we're starting to seek reimbursement coming onboard, where demand is increasing in Germany, increasing within France, and we're still yet to see reimbursement on a number of other countries. So U.S. growth still there. Other markets outside of the U.S. have an opportunity to get to the U.S. levels, and that was the comment that I was making.

Thanks, Dave. The next question is online...

Pascal, sorry, just a clarification to Mark's answer.

A minute there. I'm just thinking forward to answer the second part of Jo's question about China a minute ago. She was asking what product could be impacted moving forward on a negative basis. And as I said, the second half in China will be impacted by some of those new policies of tendering products that are patent-expired. So I would say that the one product that you could see being impacted in the second half will be Crestor because we lost a tender. In fact, as you probably noticed, of all the tenders that were issued in China, only 1 tender went to an international company, and that's with us with IRESSA. But we did lose Crestor tender, so certainly, that product will be impacted. So the online question related to PACIFIC 2, and it is do you still estimate that the PACIFIC in eligible population due to progression of after chemo that could become eligible based on PACIFIC 2. It's still at 25%, or are you seeing a lower number in the real world given the difficulties associated with defining progression in the setting on the availability of Imfinzi as an option? And this is a question for you. It's for you, Dave.

And so, Marietta, on this question, I guess the main focus for PACIFIC 2 is testing the hypothesis that can we actually get more patients through CRT and those that progress or drop out within it. The rate that you're using there is a reasonable one in terms of what we see of patients that while initiating therapy do drop out. Now I think it may be implied within this is another question, which is we do not promote to this, but we do see that there are physicians that are utilizing Imfinzi in patients post maybe just radiotherapy or if they're not able to actually give a full course of CRT in the way that they want to for tolerability reasons. But again, that's a minority of use that we're seeing. And I think that PACIFIC 2 provides us the opportunity to be able to give evidence in the setting, which we believe is an important area of opportunity.

Thanks. Dave. So we'll take last 2 questions. Hopefully, we can deal with them within the time allocated. We're a bit behind time. So the first one is Seamus Fernandez at Guggenheim. Seamus, do you want to go ahead?

Yes. So maybe just if I can make one suggestion, and then I'll follow-up with a question. But just in terms of the cash flow dynamics, I think it would be helpful to everybody if we had a slide or a schedule of potential payments going forward so people can understand more clearly the specifics around cash flow dynamics that's become an over focus relative to the future growth of the company. So just a suggestion there. And then separately, my question really is in terms of our thoughts around the growth opportunity for the overall portfolio and the various programs that you have studied outside of the Daiichi Sankyo program. Which programs are you particularly excited about to really improve cash flow going forward? I think probably the biggest focus right now is on roxadusat. And I feel like that program maybe isn't appropriately understood in terms of where the pushes and pulls are for roxadusat. So I'd really love to know a little bit more AstraZeneca's thoughts around that program and how you drive the opportunity in anemia.

Thanks, Seamus. First of all, the comment when -- so thank you. The question growth opportunities is least of -- I mean not least. There's more than one of [indiscernible] an important one. We will have SOLO-1. We'll have a lot of -- well, we'll have, well, PT010. But it's probably more for 2021 and beyond in term of cash flow generation. But -- so roxa, we will have Lokelma that we have to launch. We'll have some of our I/O readouts. So there's quite a few. But maybe what we could do is focus a little bit on roxa since we must focus on it. Or do you want to say a few words about the opportunities, Ruud?

Yes, of course. First of all, we need to see the safety analysis, and we have said that it will be done in the first half of this year. But the opportunity, of course, is very substantial both in the dialysis segment. Not to give you a few numbers, roughly, there are 600,000 patients in the United States on dialysis. It's growing. But the -- especially, the nondialysis population is at least 4 or 5x bigger than that. And at the moment, in the nondialysis setting, only iron ore, iron ore, IVR is used because of the black box warning of EPO. Equally, the opportunity in China is very substantial. In China, roughly, there are 800,000 patients on dialysis, who already have secured approval in that indication, and we are doing our best in order to get on NRDL as we speak. So we need to wait and be more patient, but we will launch roxadusat in the second half of 2019. So all in all, depending, of course, of the outcome of the safety analysis, we feel very good about this opportunity. And that the medical need is very, very substantial, but we just need to be a bit more patient in order to see the full data set.

Yes. Thanks, Ruud. I mean -- and maybe that gives me an opportunity to make a comment on this. That I said it is because I know some of you have been wondering when is it coming, and we communicated earlier that it will be first half, and we're on track on this. You have to keep in mind that it's a very complex database. I mean it's -- first of all, large database studies. But importantly, it's not your standard one company sort of analysis. We have three companies involved. So we have to consolidate all of these. FibroGen is taking the lead, of course, consolidating these. Then when it is all done, we'll have to have the 3 companies working together to analyze the data and agree on the conclusion. So this is not a very simple exercise but [indiscernible] process. As soon as we have the results, we're very well aware of the obligations here, we will communicate those results immediately when we know the safety analytics. So with that, maybe Sachin Jain at Bank of America. Sachin, go ahead.

Sachin Jain, Bank of America. One question on roxa and then one clarification. On roxa, Slide 30, I know I'm overinterpreting, but will it be one safety press release or two? Because you referenced the pooled safety analysis followed by totality of evidence. So is there a splitting of the CV versus a broader safety? Or am I overinterpreting that? And then just the second question just on -- back to the cash flow, the BD payments. Marc, just to clarify, you had said that the predominant payments were in 1Q. In your prepared remarks and an answer to a prior question, you said there were further payments across this year. And so is it fair to assume that the further payments are smaller than the $400 million for Nexium? And is there any larger payments you could call out? For example, I know you're due $600 million on development milestones on roxa. So should that be positive? Would we expect that cash outflow this year?

Thanks, Sachin. So the first question, maybe I can cover. The second question, Marc, you would cover. The first question, very quickly. Again, I just kind of kind -- I can only repeat what said a minute ago, Sachin, it's that it's a very complex data consolidation, 3 companies involved. The lead partner here is FibroGen. This is their product. We take their lead, and so we're working as fast as we can. But today, I can't answer your question in the end whether we will have a safety, safety -- CV safety analysis that we release. So we will release everything in one go. It will depend on how quickly we can get those analysis completed. And of course, we know everybody is focused on the CV safety. So as soon as we have this, we know we have to release it. Ideally, we would release everything in one go. But it may be that we have two sets of data releases. Again, we're going to have to take our lead from FibroGen on that one. Marc, do you want to cover the...

Yes. So as I said, there won't be a payment as large as the true-up for the joint venture with Merck later in the year, but there will be multiple other commitment that we'll have to satisfy and pay, and I described most of them. They are usually milestone of some form of success, progression of a development program, approval of that development program. So they are -- there is still a quite a few program that we will have to pay for in 2019. But we have also given an indication that the spend into the outlay in 2019 first quarter was relatively large quarter. So I think from there, you can probably project the rest of the year.

Yes, that's an important piece. And the majority of this large payment is just taking place in Q1. So maybe the last one question because we are out of time, unfortunately. There's a few more questions, but we'll take one last one online. That relates to Tagrisso, Dave, for you, and it's saying that U.S. and Japan are well penetrated in the existing indications. China is the future key growth driver. And I would not underestimate Europe or so. I mean, unfortunately, I have to say really unfortunate in Europe. Patients at the back of the queue. They have to wait until a reimbursement is achieved. That's a sad reality of our reimbursement system in Europe across all products in all markets, quite frankly. So they -- but we are getting there in terms of getting second-line reimbursement. As soon as we can, we'll get first-line. So Europe still has to make a contribution. But the question here relates to China. Where is the China penetration now, Dave? And where do you see these verticals filling in?

So let me take James' questions in reverse order. So in terms of the second line, which is where we've got both the approval and the NRDL listing, we estimate that there are, within China, as many as 80,000 patients. Now we take in our own mind about 20% to 25% access cut on that. So we think that there is between 16,000 and 20,000 patients treated in the second line that have the ability to really get access to. We have a small percentage of those patients, and we've got opportunity for quite a bit of continued growth, and that's just in the second line. So obviously, we look forward as well to frontline indications, and then we'll be looking forward to frontline NRDL opportunity. So there's really a tremendous amount of China opportunity for Tagrisso. And I can only reiterate what Pascal said, which is that we still see quite a bit of frontline opportunity across the rest of the globe, and I think it's important to look at the China numbers and the composition and see that it is a product that really, U.S., Japan, China, Rest of World, all make important contributions, too.

And again, on China, remember, we have the tender. We won the tender for IRESSA. And it's actually very important because it is important for IRESSA is important for first-line lung cancer patients in China. But it's also important for Tagrisso because in doing this, we kind of create, if you will, the second-line market, and the opportunity there is still very large in China. And as Dave said, we'll then move on to first-line, hopefully. So we'll close here, and I would like to again thank you for your interest and repeat that we believe we are very much on track with what we told you before. We're now focused on driving this top line growth, taking it to the bottom line, improving our operating margin and improving our cash flows. And what we have said to you before remains true. And I also want to maybe close by saying this DS-8201 opportunity was a unique one, and hopefully, over time, people do realize the potential of this agent, which is we believe enormous and explains and justifies why we did this equity raise. So with that, again thanks for your interest, and I wish you a good rest of the day.