Good afternoon, Europe and good morning to the US. Welcome ladies and gentlemen to AstraZeneca’s year-to-date and Q3 2018 results. Before I hand over to 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 Reform Act of 1995. Participants on this call may make forward-looking statements with respect to the operations and financial performance of AstraZeneca. Although we believe our expectations are based on reasonable assumptions, by their very nature, forward-looking statements involve risks and uncertainties and may be influenced by factors that can cause actual results to differ materially from those expressed or implied by these forward-looking statements. Any forward-looking statements made on this call reflect the knowledge and information available at the time of this call. The company undertakes no obligation to update forward-looking statements. Please also carefully review the forward-looking statements disclaimer in the slide deck that accompanies this presentation and webcast. I would now hand the conference over to Chief Executive Officer, Pascal Soriot.

Hello, everyone. It’s Pascal Soriot here. Welcome to our year-to-date and third quarter conference call and our webcast for investors and analysts. We have people on the phone in the webcast. The presentation as always is available on astrazeneca.com for you to download. So please turn to slide 2. This is the usual Safe Harbor statement. As a reminder, today, we will be making comments on our financial performance using co-reporting metrics and at constant exchange rates, CER, which are both non-GAAP measures. All numbers we refer to million US dollars and growth rates will be at CER and for the year-to-date, end of September period of 2018 unless otherwise stated. Please turn to slide 3. We plan to spend about 45 minutes on the presentation and then we’ll keep time for Q&A. For those on the phone, you can get in the queue by pressing star one. There is also an option to ask questions online as part of the webcast. As we would like to provide everyone with an opportunity to ask questions and as a reminder, please limit yourself to one question each in the first round. Thank you very much for that. Today, I’m joined as always by Dave Fredrickson, our EVP of Oncology; Mark Mallon, our EVP of Global Products and Portfolio Strategy, Medical Affairs and Corporate Affairs; Marc Dunoyer, our Chief Financial Officer and Sean Bohen, our EVP of Global Medicines Development and our Chief Medical Officer. Please turn to slide 4. This is the agenda and these are the topics we plan to cover today. In addition, Sean will spend a little bit more time on the year-end update on the late stage pipeline and the look forward to 2019 and ’20 where our pipeline news flow remains busy and very important. Turn…

Great. Thanks, Pascal. I want to take an opportunity to update you on the performance of our new generation of medicines. We'll start with oncology and then I'll hand it over to Mark Mallon who will go through a summary of CVRM, respiratory and emerging markets. We can turn to slide 13. We're really happy to announce that total oncology is now up to 28% of our total product sales, and as Pascal mentioned, it's now growing at 44%. The four new medicines delivered 1.2 billion in incremental sales versus the same period of 2017 and that was with continued growth coming from Lynparza while also Tagrisso and Imfinzi continued their launches in their new indications in first line EGFR mutated non-small cell lung cancer and unresectable stage 3 non-small cell lung cancer respectively. We continue to see encouraging uptake of Calquence in the smaller mantle cell lymphoma indication as we prepare for that larger chronic lymphocytic leukemia indication with pivotal Phase 3 readouts coming next year. As for our legacy business, which I won’t talk more about after this slide, Faslodex trajectory has maintained as the expanded labels and combinations with CDK4, CDK6 inhibitors start to take effect. If we could turn now to slide 14. On Lynparza, we saw ongoing progress with sales of 438 million in the year with growth across all of our regions. As we continue to roll out the broader label in ovarian cancer and the breast indication in the US and in Japan. In the US, sales were 233 million for the year, where Lynparza continues to be the leading medicine in the PARP inhibitor class as measured by total prescription volumes in this very competitive market. Quarter on quarter, sequential sales were impacted by inventory in the US, as we withdrew capsules…

Thanks, Dave. Now moving to new CVRM, our medicines in cardiovascular, renal and metabolism, sales were up 12% despite intense competition with year to date sales at $2.9 billion. Growth for both Farxiga and Brilinta remained strong with double digit increases globally. Brilinta delivered sales of 945 million with 18% growth so far in the year to date and that was driven by strong performance in emerging markets, up 31% and continuous growth in the US and Europe, up 16% and 12% respectively. So, we continue to be very pleased with the performance of Brilinta. Farxiga delivered sales of 994 million in the year to date with 32% growth, maintaining volume market share leadership globally. Farxiga saw a growth of 24% in the US, gaining in class share due to improved competitiveness across healthcare plans. Outside the US, where we have 58% of our global sales, we've seen encouraging performances with volume driven growth increasing, with Europe up 25% and emerging markets up 57%. In China, we are pleased to have delivered reimbursement in 8 provinces so far with further discussion underway. We’re looking forward to sharing the detailed results of our positive DECLARE study during the American Heart Association in just two days' time. Turning back to the US, in the fast growing GLP-1 market, our auto injector Bydureon BCise is a key driver with BCise franchise continue to scope up 19% globally in the quarter and then importantly we estimate that over half of our new patients to the franchise are starting on Bydureon BCise. Next slide please. Turning to respiratory with 5% growth in the quarter, we have now returned to growth in the year to date at 2%. During the quarter, the ongoing challenge of price competitive environment in the US for Symbicort was offset by…

Thank you, Mark and hello, everyone. I'm going to spend the next few minutes taking you through a financial performance in the first 9 months of the year. Please turn to slide 23. As usual, I will begin with the reported profit and loss, before turning to the core numbers. As Pascal mentioned earlier, product sales increased by 2% in the year-to-date supporting our guidance for the year. It is worth remembering that the collaboration with Merck delivered around $1 billion of external revenue by itself in the first 9 months of 2017, which explains that external revenue declined by 81% in the year to date. I do however anticipate a significant sum of external revenue and other operating income to be recorded before the end of this year. Although external revenue will decline as a proportion of total revenue from what is already a small number, I want to be clear that we remain focused on creating further external opportunities in the future. As we have begun a long awaited return to growth, I'm pleased that we have reduced restructuring cost in the year-to-date by 374 million to $271 million. We have also reduced CapEx by 121 million to $728 million. Anticipated declines in both restructuring cost and capital expenditure over the full year. Please turn to slide 24. Moving now to the core profit and loss, our gross margin ratio for the year to date fell as expected by 2 percentage points to 80%, driven by the comparative impact of positive manufacturing variances in the first half of last year as well as the inclusion of the profit share with Merck. Importantly, the ratio was stable in the third quarter at 79% and very much in line with the second half of 2017. Total core operating expenses increased…

Thank you, Marc. I would now like to share an overview of our pipeline progress this year and take a look forward at our upcoming news flow. Can we please turn to slide 30? A level of pipeline news flow continues to be unprecedented. There was strong progress with all our new medicines during the year, bringing new benefits to patients across all therapy areas. Within our new oncology medicines, Lynparza strengthened its lead in ovarian cancer with positive data in the SOLO-1 trial and received approvals in major markets for breast cancer. We continue to establish our lung cancer franchise with approvals for Tagrisso and Imfinzi across several markets worldwide, building on the unprecedented data from the FLAURA and PACIFIC trials respectively. In CVRM, approvals for Lokelma in the US and EU kick started our growing renal portfolio. This year, we had positive top line results in DECLARE with our SGLP-2 inhibitor Forxiga and we also submitted Forxiga for type 1 diabetes in the EU and Japan based on novel findings from the DEPICT trials. Also in Type 2, the Bydureon BCise device is now available to patients in the EU. Approvals in asthma for Fasenra and COPD for Bevespi are delivering our respiratory franchise as is FDA breakthrough therapy designation for tezepelumab in the US. Please turn now to slide 31. With our new oncology medicines, Lynparza is cementing its position as the PARP inhibitor of choice across different tumor types. Today, we can tell you we have submitted the remarkable SOLO-1 data in the EU, Japan and China. Approvals for first line maintenance therapy following chemo and BRCA mutated ovarian cancer coupled with its already widespread use as a second line maintenance treatment will establish Lynparza as the leading PARP inhibitor in ovarian cancer. Added to this,…

Thank you, Sean. Please turn to slide 41. Before we end, let me give you a short summary. We are really pleased by the improving financials and the return to growth in sales, in the year to date, but most importantly in quarter three, more than 9% at CER. We promised an inflection point and we have dedicated it. Total revenue was held back by lower externalization. We see potential for significant income in the fourth quarter from Merck and we also recently announced further streamlining of our portfolio, including next year in some markets. Depending on the structure, such agreements are realized as externalization revenue or other operating income. We remain focused on productivity as shown by the 2% increase in total co-operating expenses, while we continue to invest for growth and in China. Our ongoing launches and the performance of the new medicines keep up optimistic about the future. We’ve realized more than $1.8 billion in incremental sales for the new medicines and 76% growth. China also is doing incredibly well. The pipeline has continued to deliver important news flow, which would support sustainable growth moving forward. So altogether, we are clearly on track with 2018 guidance today. I will stop here and we’ll now go to the Q&A. For those on the phone, please remember to press star one to ask a question. We'll also take written questions from the webcast and as always, I would like to remind everybody to limit questions to one to be fair to all of our callers. Thanks in advance. You can ask a question and come back and ask another one later, but one at a time would be great. Perhaps we can start with the first question from the conference call and we'll ask Emmanuel Papadakis at Barclays to ask his question. Go ahead, Emmanuel.

I have a lot, but maybe I’m able to start with in order of priority, the cash flow question for Marc. I would just love to hear you reiterate or not your thoughts about still delivering a level of free – of cash flow from operations, rather inline or ahead of 2017’s 4 billion. We’re obviously a long way from that figure year-to-date. Maybe you could just remind us in terms of working capital movements as well year-to-date. And if I could throw in a relevant addendum. Your level of comfort on current dividend cover will also be appreciated. Many thanks.

Thank you for these multiple questions. On the cash flow, so if you look at the cash flow for -- the cash flow from operations for year-to-date September, you can see that we are achieving about 400. This is lower than what we had last year at the same time period. This is due predominantly to lower EBITDA and of course the EBITDA of 2017 was supported by the strong external revenue on our contract with Merck. So, this is one of the reason. And on the working capital, if we look at the inventories, receivable and payable, they’re not a lot of differences between ‘17 and ’18. However, we’ve had several settlements of litigation that came out from provision and this has deteriorated the cash outlay for short term provisions. And then you asked me a question about the dividend. I think the dividend cover -- we will have definitively in 2018 -- our EPS guidance is between 3.30 to 3.50, so that's going to be clearly lower than our usual for ‘20 guidance that we have provided in the past. So, it is a fact that 2018 will be a sort of a lower EPS in comparison to previous year. But we will still maintain our dividend policy, which is progressive dividend policy where we maintain or increase whenever we can. So, I have no special comments on the dividend, which has been re-confirmed time and over again by our board of directors.

Okay. Thank you, Marc. We’ll go to the next question, Tim Anderson.

I have a question on Tagrisso, so impressive uptake in emerging markets and especially in China. Yet in China, it's only approved for second line disease. You note how the first line application could be approved in later 2019. My question is how you think that will actually affect uptake relative to today. I'm wondering if it's possible that the first line approval could end up actually not having that great of an effect, because China might push use of first generation drugs like Tarceva and Iressa and the first line setting and keep Tagrisso relegated to second line, even with the first line approval. So you have NDRL or NRDL listing for second line, would you likely get that for first line as well? And then just one quick question on BR31, which is an ADJUVANT trial, is that registrational for Imfinzi? It's a cooperative group trial and sometimes cooperative group trials are not registrational?

Thanks, Tim. Two great questions. Maybe, Dave, you can start and then Sean will add.

Yeah. I think, Tim, there's both a potential and a filing question in the Tagrisso piece. So why don't I talk about the market potential component. Sean can speak to the filing and then you can talk about BR31 afterwards. So in terms of Tagrisso, I'm -- maybe just to start while your question was front line focused, I think it's worth just taking a moment on second line. As you know, we've been launched into the private pay market in China for some time now and we've seen growth in our sales in China as a result of that. Certainly one of the things that we're encouraged by with the NRDL listing is it opens up the market to quite a bit more patients within second line. So we estimate that around 115,000 patients will be treated with the first or second generation TKI and of those, from a market potential perspective, we think that about a quarter of them will be tested and found positive for the T790M mutation. So that gives a sense of the size of the population and I think that we certainly count on our operations and commercial teams to be able to drive the uptake that we’ll have within our DL. In terms of first line, Sean, maybe a good point to transition to the filing discussions there?

Sure. Yeah, Tim. Thanks for the questions. With regard to Tagrisso first line in China, as you correctly captured, we expect a regulatory decision from CFDA next year and I think the question is well, what will that do to uptake, where does NRDL fit? So the first thing is that all those initial approvals, really, China is a private market at that point. So individuals who are able for that indication to pay for the drug then will have access to drug. So there will be some use that way. The other thing I'll point out is obviously the duration of use, because the duration of freedom from progression or death is much longer and the first line is actually per patient who initiates quite a bit longer in the first line therapy. The next part is what will happen with NRDL and I would add then that uncertainty from what the Chinese government will do. Then also when they will do it, whatever it might be. So we just, I think, we go step by step, first approval, approval will lead to a launch, again to a private market and then we'll start the discussions with the Chinese government. On to BR31, adjuvant trial for Imfinzi and non-small cell lung cancer, you are correct, the trial is sponsored by the Canadian clinical trials group. And the question was, is it registrational? Our intention is that it is registrational. The group has a history of doing trials that have enabled registration in lung cancer. They were previously called NCI Canada and during that period, we’re able to conduct trials that support registration and of course we work very closely with them during the conduct of the trial to make sure that the quality of the data that results will enable regulatory filing and will meet muster when inspected by regulators globally.

Thank you, Sean. Next question is from Sachin Jain.

Just a very high level financial question. It’s clearly positive to see the sales inflection for the first time, as you mentioned Pascal for a number of years. I wonder if you could talk to when we would expect that to flow through to EBIT. And I know it's early and we get formal guidance on ‘19 next year, but why don’t you just talk to some of the moving parts as we think into next year, lower down the P&L, so just color on SG&A growth, given the slightly higher guidance for this year, broader comments on the underlying margin leverage and whether you expect that to offset however one-off income play through next year?

Thanks, Sachin. So I think it's a good question and I think it’s important to remind everybody of what it is we're trying to achieve in the strategy we pursue. First of all, we have to rebuild the pipeline, so we went to three steps. One is, we will build a pipeline because we didn't have any new products. Two is, we launched this new products and we turn the pipeline into commercial successes and then top line growth. So we are in this period right now where we are driving topline growth and we have to drive this as hard as possible and the next phase is to drive the top line growth to take the top line growth down to the bottom line. So yes clearly we expect the operating margin to improve over the next year or two. We said that we would expect the operating margin to reach industry norm 30% and beyond, as we move beyond 2020. And certainly in the meantime what we're doing is continuing to streamline the portfolio, generate income through these activities and call the dividend. I just want to repeat, I mean, I want to be clear with the dividend cover, we will certainly maintain the dividend as we go through this period of time. We almost, at the end of the tunnel if I may say so, we can see the light at the end of this tunnel, the top line growth is strong and if we continue and as we look at 2019, it will continue happening like this. So what we have to do is, over the next period of time, ’19, ’20, we will manage our expenses to drive top line growth and we're going to focus now our attention to what you might call underlying growth of IPOs, of our profit, excluding externalization. So, we may get some greater visibility and then you can be reassured that the business is moving in the right direction. I can’t really be too specific, I'm sorry Sachin in terms of giving you guidance for the expenses, we'll do that early ‘19. But suffice to say, we are very clear that we need to increase the operating margin and to do this, we need to drive top line growth, but manage the cost. And you saw in the SG&A, it's a totality of SG&A and R&D that we have to manage of course.

Can I just have a clarification? You mentioned the focus on underlying growth into ’19. Does that imply guidance models have been on an underlying basis?

Well, this is the one thing that we are thinking through. I think certainly, being a little bit clearer as to what the underlying growth is is certainly going to be very much at the top of our mind, but we’ll see what we do with guidance per se, but certainly we’ll make sure that, I mean, our focus is on that and we have to see how much details we're going to give on that, but we're thinking it's through, because it’s really key to make sure everybody understands the underlying business is not in a period of very fast growth and the top line growth is going to drive earnings growth. No question about it. So we're going to move to the next question. [indiscernible]. Sorry, the question is online, apologies, I’ll have to read it. So the question is being early in the product lifecycle and late in the R&D lifecycle, we're seeing 2018 SG&A cost being higher than initially anticipated and R&D lower, but in ’19, 2020, we are still early in the product life cycle and you must not be complacent in order not to fall into a 2006 like patent cliff in the future. Sorry, I'm going through the question here, this line of thought implies continued strong growth in SG&A in 2019 and maybe slightly increasing R&D cost, potentially putting at risk the top line growth into bottom line, the market is expecting, please reassure us on your commitment to deliver on consensus expectations. Yeah. I mean, let me just repeat what I said a minute ago. We’re very clear that we need to increase operating margin, drive profitability up take the top line growth to the bottom line, improve our cash flow import. There's no doubt and we've said it before. I’ll say it again. The plan is to manage through this transition period and then start reducing our debt across 2020, So that is very, very clear and that is what we're going to do over the next year and continue to do in 2020. [indiscernible] at UBS. Jack, go ahead.

A long-term question. There is a new challenge that faces people like us now in oncology and that is that so much is being discovered that drugs that were pretty good two or three years ago are being rendered obsolete, as all that patents expire. So you've done it to Tarceva, Iressa and gilotrif. You're hoping to do it to ibrutinib with Calquence. So if we think about Tagrisso, where – there is a patent expiring in 2032 in the Orange Book, what odds would you give on Tagrisso, making it to patent expiry without being rendered technically obsolete as a first line agent in lung cancer?

Yeah. It's a great question, Jack and I think technology of course is evolving our days, but I mean, it's taken us five years to get to where we are with Tagrisso and I would like to remind everybody and I'll ask Sean to also comment on this, but I'd like to remind everybody that like five years ago, everybody was saying there's lots of competitors around you or around Tagrisso, that is. And many of these products have just disappeared. So it's not that simple to develop an agent as exquisite as Tagrisso that penetrates the blood brain barrier as effective as it is. Mow having said that, we certainly should not be complacent and we’re working ourselves on the next generation of compounds and also more importantly, how do you manage the patients who progress on Tagrisso and develop resistance to escape, escape efficacy from Tagrisso. Can we say by 2032, Tagrisso would be the only agent out there, probably how to say, but I think we have a quite a few years ahead of us where Tagrisso can still enjoy a pretty nice position, but Sean over to you?

Sure, Pascal. Thank you Jack for the question. Again, I will reiterate as Pascal said, 2032 that's a ways off. So certainly a tremendous amount can change between now and then. At the same time, I think, if we were to see an agent that had a really viable opportunity to displace Tagrisso in the normal life cycle of a drug, say 5, 6, 7 years in terms of development to launch, we really haven't seen anything that has the attributes that are required. Obviously T790M is something that a lot of companies, including AstraZeneca have targeted. As Pascal said, I think it's the central nervous system penetration and ability to really treat a common area of the body where metastasis advanced actually are quite lethal and are untreated by other agents probably provides a pretty good period of protection for us.

I will give another example Jack, because I think you raised a very important question. Many of us quite frankly a few years ago thought in immunotherapy, things would move much faster than they actually have. I mean, people saw a number of new targets come up, new combinations and in fact we by now, should have a number of those OX40, I still remember was one of those and there were many others. And in PDX, if I may call it this way, still the mainstay of immunotherapy combination with chemo and I do believe new combination of immunotherapies would emerge, but it's probably not as fast as we could have thought a few years back. Jo Walton at Credit Suisse. Jo, go ahead.

Can I ask about Fasenra, perhaps it’s on the same move, new products coming in and making other ones potentially obsolete, but how could you compare Fasenra and Dupixent now that you've seen the Dupixent label?

Thanks, Jo. I think Mark Mallon would be delighted to talk about Fasenra.

So, I mean the first response is that we continue to be very confident in Fasenra and our profile and we think this is a great medicine for treatment of severe asthma. We've got a wealth of data through our Phase 3 studies, we’ve just shared the most recent results of the BORA study, fast improvement in lung function, great reduction in exacerbations, great reduction in OCS. And with the really favorable dosing approach, with the once every two months or 8 weeks. And the mechanism of Fasenra with targeting directly the IL-5, bringing killer cells to the eosinophilic, unique mechanism is sort of exquisitely designed to target eosinophilic asthma and the feedback that we've gotten from physicians is outstanding and so you take that profile and compare Dupixent, there are going to be patients in which Dupixent patients, what we see in the label will be attractive, for example, overlap patients between [indiscernible] and severe asthma. And obviously, we will be a place that will go, but it is -- we have to really see how the two, the IL-4, the IL-13 really plays out in a broad audience. We need to understand about how that in some patients, we see an increase actually in eosinophilic levels in response to the medicine, all of that has to play out as physicians get a chance to use it. The most important thing I would say is that the unmet need is so high in the severe asthma and so with only one in 10 severe asthma patients eligible for a biologic getting them, I think that, we would be good to see this market expanding and maybe there will be some patients that -- certainly there will be patients that Dupixent is a good option for, but we're confident for our target patients which are very large, where the unmet need is high and where the evidence is both from the clinical studies and the physicians and patient experience is really outstanding and so we remain confident.

Thanks, Mark. I think that's the key point actually Jo is that the growth of the biologics glass treating asthma, I mean, is potentially enormous. It would have been a bit like comparing two new TNF inhibitors, 15 or 20 years ago in rheumatoid arthritis that would have been missing the point that patients need the more biologics treatment and I think it's the same here. Of course, there will be competition and of course, we recognize that Dupixent is actually our tall competitor, but I think also we need to hope that companies collectively will help patients with better treated and through this expanded class, 10% of patients get a biologic and that should be getting one. What we hear from Fasenra as Mark told you is speed of action is important to physicians, full depletion and efficacy is pretty good for patients within disease. And I would say also because there's a lot of talk around the home use, we have a home use program, but what we have from physicians is in severe asthma, we’re not talking about moderate asthma here, we’re talking about patients with severely ill, in severe asthma, being able to bring the patient every couple of months to the office to inject Fasenra is also good from a compliance viewpoint and follow up. So, all those, I think, things make Fasenra a pretty competitive edge into the marketplace. [indiscernible]

Good afternoon. So switching to diabetes, I just wonder whether you have any comments on your dual GLP1 GCGR in phase 2 and whether we might see some data soon and any thoughts on the outlook, given some of the positive data from Lilly recently. And then just checking on Farxiga, the filing timeline and how soon you might expect the DECLARE data to impact sales? And then maybe just to check on Brexit, since you have the slide. It looks like the impact on cost of goods is probably manageable, but are you assuming some of that impact for the stock builds in the fourth quarter of this year, and I assume it’s always in your guidance, if that's the case?

So Sean, do you want to cover the first two and Marc Dunoyer might cover the third one.

Okay. Sure. So thank you for the questions. So starting with MEDI0382, that's the molecule you’re referring to, which is the GLP1, the GLP1 dual. So with that molecules, Phase 2b is ongoing and we are looking at the data. We’ll plan to present it when we have it and really our decision as to where to go next will depend upon our review of that data. You mentioned a Lilly molecule, it’s slightly different, GIP. So we are not – we’ll have to look at the data, look at their data and understand whether the mechanisms might be similar or different in terms of the effects they produce. The DECLARE question. So we’ll present the DECLARE data on Saturday at AHA and I think you'll get a chance to see what we think is a really extremely meaningful effect, both in terms of the reduction in hospitalization, heart failure or cardiovascular death, which is a primary endpoint that was met, but also the opportunity to go into a much broader patient population, which is not only type 2 diabetes patients, who have had an index event that will be secondary prevention, but recall 7,000 such patients in DECLARE, there are also 10,000 primary prevention patients – patients who have not had an index event. So really a much broader spectrum and we intend to file that data. We will do it as soon as we can and can get the filing together and it's probably from Marc to talk about what he expects the impact to be once -- should we get that data on the label and when we do.

So obviously, we can't promote until the label has changed and there's variations of rules around this in different countries around the world. What I think you can say is, as Sean highlighted, there is going to be two important things coming out. One is that, this will be another positive evidence of impact on CV outcomes in particular, really important. Complication of diabetes, heart failure, that's going to be a really encouraging and supportive result as well as the fact that it’s going to be in a much broader population, so I think that's very exciting news to physicians. If you look historically, major data like this has an impact on use, even before approval in many cases. So I'd look to the past bottles to give an indication of what could happen here.

Regarding the impact of Brexit on our guidance, I think it’s fair to say that we have a minor impact on the level of inventories. But it will also have a minor impact by a way of consequence on the EPS. But this is well within our guidance range and therefore there is no -- we don't communicate any impact of Brexit on our guidance for 2018.

Thanks, Marc. So we have a question by email [indiscernible]. The question is for you I think Dave. It's regarding Imfinzi, where do you see a ceiling, a plateau from a purely medical perspective, in other words, what proportion of the unresectable stage 3 lung cancer population is completely CRT ineligible and therefore out of reach for Imfinzi?

Maybe what I can start with is, some of the variability that we see right now across regions, which I think gives some sense for how high this could go and then I'll turn it to Sean to add some color. Just if we take within Europe alone, we see CRT rates between -- anywhere between 40% in some of the countries where it's lowest to as high as 70% in some of the countries where CRT rates have been adopted and we believe that there's opportunity to grow CRT rates in all of the regions that we're in. I think your question appropriately implies that there are certain patients that just simply will not be seen as being eligible for CRT, but in terms of the real world right now, that's what we're seeing at the moment and I think suggests that there is an opportunity certainly to get to the overwhelming majority of patients being treated with CRT. Sean?

Yeah. I think -- to add on to that, I think there are two opportunities that we’re really looking at here. So, the majority of patients who have unresectable stage 3 get CRT, the vast majority do, but the one thing is that if there's a tolerability issue, sometimes, what will happen both tolerability or access to the coordination that's necessary to get concurrent CRT, which is what PACIFIC studied, they will sometimes get sequential CRT either for tolerability reasons or due to availability in the geography. And as I mentioned PACIFIC5 is the ongoing study that really studies that too. So that's one way to access a different group of patients. We are also hearing and I don't think this is something we can really quantitate. But we're hearing physicians start to think differently around how aggressive they want to be with surgical resection of stage 3 non-small cell lung cancer, given the data that's being seen with PACIFIC and whether they may choose to have the patient undergo a CRT rather than attempt a large or complex surgical resection.

So we have a question from Alex at BMO Capital Market.

Pascal, you mentioned there was skepticism about Astra’s ability to return to growth and I wanted to build on that please. As we enter the growth phase, I was wondering if you could revisit your 2023 guidance of 45 billion. I think that about 40 billion to 41 billion FX adjusted now and recent management commentary suggested that it was still possible. Obviously, a lot has changed since you provided that guidance and it is significantly higher than what the Street expects, so I’m just wondering, do you still believe that you can get to 40 billion by 2023 and if so what are the ways that you can get there that you are not well understood by the investment community?

Yeah. It's a great question. The plan, as we have developed it today still takes us to that kind of level. And I see no reason at this point to change this Alex. And to answer whether we’ll get there or not is hard, because to some extent, it depends on how we execute commercially, but this we have, what it takes to deliver and then the other part is the success of our products and new indications in the pipeline. So, that will happen over, I mean, will become clear overtime. Roxadustat, for instance, we’ll know pretty soon. Next year, we'll have more data, we'll have, for instance, the POSEIDON study in immunooncology. We have adjuvant studies, with both Tagrisso, sorry and Imfinzi. We have first line ovarian cancer study in combination with Avastin for Lynparza. We have tezepelumab that is progressing through the pipeline. We have, if STRENGTH, we've not talked about Epanova much, but if STRENGTH is positive and we'll see this weekend, the MRN results, I mean Epanova has substantial potential. So there is a whole range of products that are still potentially underestimated, because we don't know whether they're going to work or that and -- but when we do a risk adjusted search forecast, we still are on track to get there and it turns into, of course, I’m not disclosing anything that you cannot work out yourself, it turns into a compound growth rate of about 12%, 13% a year between now and then. So the question is, can you achieve this. And the answer is, we still believe we can do it, but it depends on the pipeline and how it unfolds. So more to come over the next year or year and a half in term of the clinical news really. Going back to Farxiga for instance that we were talking a minute ago, I think, people sometimes maybe underestimate the potential of heart failure. I think we all have to understand, heart failure is as costly as cancer in term of health care cost. It's a major problem. So if we, in diabetes, can establish that SGLT2 agents actually reduce the incidence of heart failure and also have a potential positive effect on the kidney, which we believe is the case, then the potential for the SGLT2 class is enormous. And of course, there will be competitive class, but still we believe we have a good chance to succeed here. So there is really a lot of opportunities for us to grow. Let's move to [indiscernible] of Guggenheim.

So I just had one question. Can you guys talk a little bit about the safety profile as presented on Roxadustat at the ASM? My understanding is because again we saw a little bit of some safety dynamics there that brought forward some concerns, but I'd love to just hear how AstraZeneca is thinking about those datasets and then when we're likely to see the more complete data set for cardiovascular event differential between epo. And just to expand on that, could you guys talk about the opportunity in pre-dialysis and the ability to gain reimbursement, should you show an ability to kind of expand back into that market?

So one question for Sean and a question Mark. Kindly a reminder just to be short and sharp. So we actually manage all the questions left. So over to you.

Okay. I will try and give you a short and sharp answer to a very complicated question. So I think there was some data at ASM. That's primarily efficacy data from Asia, it's small hundreds of patients, actual events we're talking a handful with a small difference and my argument from a safety standpoint with unusual events, you can't really conclude anything from that data. This is of course a complex patient population that has risk factors for cardiovascular events, because they have chronic kidney disease, they often have an underlying co-morbidity that cause that and they're on dialysis. So with regard to then when -- the rest of the question is, so what will we know, I think the way that this will work is by the end of this year, as we have said, you will have efficacy data from our ongoing program and the largest trials there are ROCKIES an OLYMPUS. OLYMPUS being in the pre-dialysis patient populations and non-dialysis dependent, ROCKIES being in the dialysis dependent and OLYMPUS with almost 2800 patients, ROCKIES with a bit over 2000 to 2100. So much larger data sets. So the efficacy data is based on hemoglobin response, what we have also guided to everyone is that the basis for filing will be an aggregated safety profile across the whole program and because we have to aggregate multiple trials, multiple sponsors in order to do that, it will be sometime next year, first half of next year that we will be able to share that aggregated safety profile and again we'll have two versions of it. One is that that which will occur from more than 3500 patients who are dialysis dependent. And then an even larger number who are non-dialysis dependent, dialysis dependent versus EFAs, non-dialysis dependent versus placebo. And I will let Mark talk about opportunity?

The short answer for pre-dialysis is we see a substantial opportunity and a compelling case we think prepares, we need to do a lot of education of course, because this is something new, but anemia is clearly an issue, people living with is a problem. And CKD patients already have a lot of problems and so having a chance to address this one will be, I think, really important. So we're very excited for the opportunity.

Yeah. Large opportunity for sure. Andrew Baum, Andrew at Citi, go ahead.

I imagine that duration, the median duration of Tagrisso treatment has increased.

Hello. We’re losing you a little bit.

[Technical Difficulty] This is a bit of a loaded question, but where do you think this could go n you go given the use of combination therapies addressing CMS as well as other targets in order to overcome T790M beyond progression using Tagrisso on progression. And then second, could you comment on your level of conviction in that differentiated safety profile of Calquence over Imbruvica, given some recent publications suggesting that the bleeding scene is a BTK related phenomenon rather than off target?

Yeah. So I think there were – there is almost a commercial element in it for Dave I thought, but obviously we've seen a -- we see from FLAURA that we get a considerable extension of the progression free survival and we believe that as first line becomes the standard of care, that will lead to each given patient’s duration of treatment on Tagrisso being much longer, just because they're benefiting for considerably longer. I think the thing that will really drive that home as we anticipate in the second half of next year is if the survival data also shows that benefit manifesting in that endpoint. You asked about MET inhibitors, savolitinib probably is where you're going with that. That's an ongoing casting. We interestingly, at ESMO, we're seeing some variation in the percentage of patients that we -- that may be using a MET over expression or their tumor I should say rather using that over expression to overcome. Treatment like Tagrisso and so that's something we're evaluating in the clinic and then also Dave and his team are evaluating the opportunity. Acalabrutinib, Calquence and bleeding, we believe that the ability to differentiate Calquence based on safety is due to incidents of severe bleeding episodes and also other severe side effects, like atrial fibrillation. We are accumulating obviously a lot of safety experience with the molecule and we believe it has reinforced the opportunity that we saw when we initially did the deal with to acquire the molecule and ultimately the mechanism of action of something like bleeding and severe bleeding is multifactorial. There may be a BTK inhibition component, but it's probably likely that there are other things as well. So we remain as confident as we were that we can differentiate based on safety. I think the question from CL006, when we go head to head with ibrutinib and generate gold standard datas. Do we also have an opportunity to differentiate on efficacy?

I think one small data point just Andrew to add on this, while the front line, it's still early days to know whether or not we're getting to the durations that we saw within the FLAURA study. In the second line, we saw patients being treated to progression and it is a testimony also to the tolerability profile of Tagrisso, which I think is an important aspect of patients getting to what we see in the studies.

So we have a question from [indiscernible] online, which is about the SG&A guidance. Marc mentioned that SG&A, first of all, it’s -- what you see on the P&L and then it's 7%. And I think Marc said that the guidance for the year was that SG&A would grow at the rate we see at the end of 9 months, so the same similar rate, I would encourage you though to look at the totality of operating costs, R&D and SG&A, which at the end of September as Marc mentioned are growing 2% total. Marc, anything you would like to add to this?

No. I think it’s 7% for the quarter three, 7% -- for SG&A, sorry. At CER, 7% for quarter 3, 7% for year to date and estimated to be 7% for the end of the year or I said broadly in line, but around 7% would be the conclusion. Everything is at CER.

Yeah. James Gordon at JP Morgan. James, go ahead and maybe we will take this as the last question.

This is [indiscernible] from JP Morgan. Could you please comment on your appetite to equity funded M&A and if the significant deal was to be perceived in the future, where would the focus likely be, would it be on accretion or growth or pipeline optionality?

I mean if we were to do anything, it would be for accretion for sure, because that's really -- there's not a way we could do anything else, but I have to say our focus is not on this. Our focus is really on launching these products and turning this pipeline into a reality. We still have a lot of work to do, growing out topline and a lot of work to do, driving increase in our operating margin. We are very committed to doing this right now, we are in this transition period and I would say almost toward the end of the transition period where we have to invest to really launch all those products, but it's very clear to us, we'd need to drive operating margin up and that's what we're going to be doing over the next couple of years. So that's really our focus and any acquisition, or M&A would have to be accretive if that's -- but again, this is really not a focus. So Thomas told me we can take one more actually. So we'll take the last question from [indiscernible]

I’m just going to ask one clarification and one question. The clarification was on the roxadustat and safety question. I think one of the concerns with hyperkalemia imbalance that we're seeing in those studies that were presented, I just want to clarify when you look to the integrated safety database of all of the Phase 2 trials, was there any signal of increased hyperkalemia risk, because I know one of the Phase 2 studies, there was an imbalance there. So I just wondered, if you could clarify that? And then on Calquence, you talked about a third of new patients going on Calquence in the current indication. I wondered what anecdotal feedback you are getting from physicians over the drug’s profile compared to your competitor and maybe how the tolerability is flying out in the real world. What is it is driving patients to choose Calquence versus the competition?

Great question. So maybe we can start with the last one, Dave, if you want to cover Calquence and then Sean, you will cover the roxa safety question.

Yeah. So thanks, Richard for the question. I think, as you know, about a third of share of the BTKI class is now Calquence’s. I think that really a very important additional element that we're seeing and this will get to the qualitative aspect is that we're seeing increased utilization in the BTKI naive patients in the second line. So a lot of the initial trial post-launch a year ago was happening in third plus line patients after patients had been exposed to a BTKI. Because of the positive trial experience that physicians have had with the agent in those settings, we're seeing more and more utilization moving into earlier lines of therapy and what that means is that it's becoming the preferred choice within the indication among the BTKI inhibitors and we see that as very encouraging.

Okay, Richard. So the question sort of had two elements. One is now moving to hyperkalemia, I believe, Andrew's question was actually about bleeding with roxadustat. Yeah. So roxa safety, that question is about CV risk. Again, this is small dataset. It’s very difficult to interpret adverse events that are actually known consequences of the underlying condition that you're treating. So these are dialyses. There is a small dataset on dialysis dependent chronic kidney disease patients, these patients’ kidneys aren't able to regulate their potassium levels. So in a small dataset, you can easily get an asymmetry in these kinds of events that are known to be related to the disease and actually how the dialysis can give -- can be given is very, so that can change it incidence as well. Again based on the data we have, we haven't seen anything to suggest that the safety profile is adverse for roxadustat. The limitation there is that it is small and so the definitive dataset will come from the pooling and we will look at all adverse events over that pooled dataset, but the most important one from a regulatory standpoint is the MACE cardiovascular events, because as you recall, the current standard of care, the ESAs have a black box warning for increased risk around those events in their label.