Thank you very much. Good afternoon and welcome to this call for GSK's Full Year 2016 Results. I'm pleased to report that sales and profits were up in all three of our businesses, Pharmaceutical, Vaccine, and Consumer Healthcare. Total group sales were £27.9 billion, up 6% CER. And core EPS was at £1.024, up 12% CER. This was towards the top end of our financial guidance. which as you know we increased during the year. In sterling terms core EPS was up was 35%, reflecting a significant movement in the currency in 2016. And if sterling rates were to remain in line with January average rate for the rest of 2017, we would expect a 9% benefit to core EPS during the year. Total EPS for the group was £0.188, down on last year primarily as a consequence of the comparative to the £9.2 billion gain in the Novartis transaction during 2015. We've declared a dividend £0.23 for the quarter, bringing a total dividend of £0.80 for 2016. And we continue to expect to pay a dividend of £0.80 for 2017. The positive momentum we saw in 2016 delivered Pharmaceutical sales of £16.1 billion, up 3%; Vaccine sales of £4.6 billion, up 14%; and Consumer Healthcare of £7.2 billion, up 9%. On a pro forma basis sales growth was respectively +4%, +12%, and +5%. On a geographic basis U.S. accounted for £10 billion, Europe £7.5 billion, and international, £10 billion. Operating margins also improved in all three businesses, reflecting good cost control and delivery of organic and transaction related savings with a group core profit margin of 27.9%, up 3.9 points on last year. These performances reflect the investments we've made to build scale and sustainability in the group and to deliver new products. Sales of the 11 Pharmaceutical and…

Thanks, Andrew. The results that we've reported today demonstrate the progress we've made in delivering on our strategy, as well as the financial goals we set out in our financial architecture. All three of our businesses are contributing to the delivery of more broadly based revenue growth. Our continued focus on the execution of our integration and restructuring programs has accelerated the delivery of the targeted benefits, allowing us to improve our margins and operating leverage, while still making substantial investments behind our new products, supply chain improvements, as well as progressing the R&D pipeline. We've also maintained our focus on financial efficiency in the P&L and in the allocation of our capital, allowing us to deliver core EPS growth ahead of sales and at the top end of our EPS guidance as well as a significant improvement in our cash generation and a dividend of £0.80 per share. We expect continued progress from the business in 2017 with all three businesses continuing to benefit from recent new product launches and other investments, including supply chain capacity as well as the completion of the integration and restructuring programs. The guidance we've given today for core EPS performance in 2017 reflects that momentum, but also takes account of the possibility that a substitutable generic alternative to Advair may be launched in the U.S. this year. This is a situation that is bound to evolve during the year. And we will update our guidance as and when there is more certainty on the competitive position. Given that this will depend on a number of variables, including pricing and supply availability of any generic, it seems unlikely that we'll have any greater clarity before the middle of the year. I'll come back to details of the guidance shortly. Our earnings release provides an…

Thanks very much, Simon. Let's open up the call to questions. If maybe the operator could just remind everybody the protocol for how to request a question please.

And for your first question comes from the line of Graham Parry from Bank of America Merrill Lynch. Please go ahead, Graham.

Great. Thanks for taking my questions. So firstly on the guidance, if you could just run through some of the reasoning for your assumptions. So for example, the midyear approval of a generic when GDUFA is at the end of Q1. And the what appears to be $0.75 (sic) [£0.75] decline post generic, given that you've previously questioned the ability of generics to erode sales due to pricing and manufacturing restraints or potential pricing and manufacturing restraints. And secondly, if I look at consensus at about the $1.36 [to £1] FX rate, Advair sales and consensus would've looked to be about £1.2 billion versus the £1 billion consensus. EPS is flattish for the year. So is it fair to say that what you're trying to tell us is that you think you could just about make something close to current consensus EPS, even if Advair sales in the U.S. were about 20% lower than consensus is currently forecasting? Thank you.

Thanks, Graham. I think in the – just on the assumption piece, it's clear that to get this right on all the dimensions – i.e., the timing of an approval, the timing of a launch, the amount of supply, the pricing dynamics – impossible to get that right. So what we've aimed to try and do is come up with an answer, which we think it essentially covers most of the likely outcomes. So of course it's possible people could launch a bit earlier. But it's also possible they don't have as much supply as you would need to get 75% erosion. Maybe they – something else happens in the marketplace. But we think the combination of a midyear launch with a pretty aggressive decline of 70%, 75% in the second half after a 15% decline in the first half, getting you to about that £1 billion, that feels like a reasonably sensible balance of probabilities. Because you might say the timing moves a bit. You might say the erosion curves move a bit. But net-net, you might come out at a broadly similar number. It could end up being a bit better. It could end up being a little bit worse. But I think this is a pretty reasonable estimate, realistic estimate of what could happen. Obviously if we've seen no product launched by the middle of the year, then we're on the upside territory and the like. I think in terms of what we're trying to tell you. We're trying to tell you we think it's £1 billion if that scenario plays out.

Yeah.

I mean I think what's clear is if any of those assumptions go our way a bit – so if we see the decline being a bit less than 15% in the first half, if we see the generic being a bit delayed, if the generic only has 50% supply – then clearly we're going to do better than that number. But we're just trying to give you a sense of what we think is a reasonable estimate for what the bottom end of the bracket is for the year, so that you can model from there. We don't really think it's very far away from where consensus was this morning, maybe a point or two, but certainly doesn't feel like it's very far away. And I think when looked at the general view of the market, I think most people's view of the market is actual arrival of the product into the marketplace probably is midyear. I mean I remind you, when we had Advair approved, it took GSK six months to get from Advair approval to launch. Now we're 15 years, 16 years down the road. People are more sophisticated now. Give people credit for doing things better now than we did in those days. Everybody's had more experience. But not super trivial to just produce 20 million packs of supply overnight for a product like Diskus. I mean I think that's a fair area to think, even if you did get a first pass approval, not super trivial to turn on maximum supply overnight. And so we think this is a reasonable – this is a reasonably simple way to pull together two or three elements of the assumption set, recognizing they won't be right, but probably covering most of the outcomes. And as we go through the year, obviously once facts becomes facts, we can start to fine tune this for you. But I don't think it's going to be a million miles away. If there's no generic, I think the 5% to 7% is a good estimate for where we'd be. And if there is a generic, I think the flat to slightly down is a good estimate to where it could be at the bottom end of the curve. Thanks, Graham. Next question?

Thank you. Your next question comes from the line of Andrew Baum from Citi. Please go ahead.

Good afternoon. A couple of questions. Before that, I just want to say that I'm sure occasionally dealing with sell-side analysts has been as much fun for Andrew as getting his teeth drilled. But I wanted to express my thanks for the candor, insights, and openness over the years. On the questions, two things. Number one, Andrew, could you address value-based pricing? It's obviously highly topical in the U.S., given the new administration among the industry. In particular how easy is it going to appear to implement, given the very significant challenges across the range of therapeutic? And then, Simon, again in terms of the new administration, obviously U.S. tax reform, border tax proposals look like they're going to happen. Could you talk how that will impact GSK? In particular I note that you committed to new manufacturing sites in Scotland rather than the U.S. And whether these are indeed going to be supplying the U.S.? And then second, I know you also relocated some IP out of the U.S. into the UK.

Thanks, Andrew, and thanks very much for your kind words at the beginning. Value-based pricing. The first thing to say is I think if you ask four different people to describe value-based pricing, you'd probably get five different definitions. So I think there remains a bit – this is a bit of a bucket description, which different people interpret different things. And it does vary a bit by country and payer model, what is viable. I think in the U.S. it's quite tricky to see how you get to value-based pricing without more transparency in the pricing system, Andrew, in the first instance. And I suspect it's going to require some simplification of some of the regulatory thicket which exists in the U.S. pricing environment. As you think about in any given zip code in the U.S., you've got every single piece of legislation which touches Medicare, Medicaid, Veterans, private, all over superimposed on each other. And so to try and drive a new pricing model through that thicket of regulation, sometimes they're quite inhibitory in the way in which you might want to innovate your pricing approach. Those two areas need – we probably need to be – frankly, we probably need to see a pretty coordinated effort from industry intermediates, payers, and government to try and rethink what this space is like to create a more fertile environment for some innovation in this space. For me what it really all boils down to is companies starting to take more risk in terms of the price that you charge for the product and having more dynamism in terms of the likely price received for the product over time, as your contribution basically ebbs and flows according to data and all the rest of it. I do think it is…

Yes. Thanks, Andrew. I mean I think as you will anticipate, the devil is going to be in the detail of what the proposals finally turn out to be. But particularly in relation to any border adjustment and which products it covers and how it covers cross-border flows. Because we're not alone in having our supply chains stretch across those borders. On balance from what we can see today, we think it's likely to be a net positive. Exactly how much, impossible to say at this point. But we feel reasonably well hedged, given the manufacturing footprint that Andrew just described. And we have retained structural flexibility to move parts of the group around, including our R&D investments and intellectual property to respond to where governments place the incentives. And that would include the U.S. as well as the UK. So at the moment we're watching. We're participating in the debate. And we'll see what proposals appear.

Thanks, Simon. Thanks, Andrew. Our next question?

Thank you. Your next question comes from the line of James Gordon from JPMorgan. Please go ahead, James.

Hello. James Gordon from JPMorgan. Thanks for taking the two questions. The first question was just how are you thinking about overall U.S. integrase uptake over the next few years? Because I know you're going to have new doublets. But there could also be a competitive integrase combo from Gilead. Do you think that's going to significantly increase the pace that integrase market expansion occurs? Or is it going to be more of a share battle? And then the second question, just pipeline wasn't a big focus on the presentation. But in terms of pipeline readouts for this year, what's important particularly in – maybe in oncology? And could we see anything progressing to Phase III in the pipeline this year?

Yeah. No. Thanks very much. So in terms of integrase, I mean I would – we've seen a very significant expansion of the integrase market since we launched dolutegravir. And I would expect to see, A, our progress continue. We've seen very little change in the uptake curves of the product. We're seeing tremendous amount of new starts coming into our product as well as switches from non-integrase backbones. Clearly if a competitor brings out another product, it just depends how good that product is. But at the very least you'd expected it to probably self-cannibalize some of its own portfolio, so again increasing the size of the market. I think it's a reasonably challenging ask to start to take business away from dolutegravir, given the extraordinarily effective resistance profile that dolutegravir has got and the kind of a lead we've got in terms of establishing ourselves. If we then are able to demonstrate that the doublet approach really does deliver both the efficacy and the resistance protection at a wide range of viral loads, which is clearly what we want to achieve, then I think that really starts to reshape the whole game. So I think one way or the other, it's a pretty safe bet, James, that integrase market size grows. I think the degree to which it grows dramatically obviously revolves around whether we win in the doublet space and/or our competitors develop a product, which has got any point of differentiation which is meaningful to clinicians and patients. Those two things are going to get played out over the next 12 months. So we're going to know probably on both of those dimensions over the next 12 months the kind of shape we're going to get. In terms of products coming through. An awful lot beginning…

Thank you. The next question is from Richard Parkes from Deutsche Bank. Please go ahead, Richard.

Hi. Thanks for taking my questions. The first one was just expanding on Graham's on the assumptions around generic Advair. Obviously I think consensus is assuming a 30% decline in the U.S. this year. And I think you're assuming around a 45% decline. Just is it safe to say that your assumption is what you'd see as essentially a worst-case scenario if generic Advair is approved and there could be upside if the rate of erosion is slower? Is that the right way to think about it? And then secondly, the guidance for 5% to 7% underlying EPS growth. I'm just wondering how much of that is driven by continued delivery on the cost saving versus organic growth and margin leverage? I'm assuming it's largely the latter. But just interested in your perspective on whether that's a good guide for the longer-term organic growth potential from the business. Thanks.

Yeah, thanks very much, Richard. Well, on the second point I think very much the latter, so very much driven. I mean still some cost savings coming through, but as Simon said, the lion's share of that is now delivered. We're in the tail end of that. And we've done that early. That program has kind of done its job. So very much driven by what we expect to see, which is top line growth, a lot of new product momentum carrying forward. I mean it's worth just reminding you. I know you've seen it already. But £4.5 billion of new product sales in 2016, £1.4 billion in Q4. So that's annualizing at £5.6 billion. So it's clear that there is significant further organic growth flowing through the business. So that's going to be essentially the driver of that 5% to 7%. As far as the Advair assumption is concerned, what we've tried to lay out here, Richard, is what we think is a properly realistic conclusion. And you can get there any way you want. You can decide to launch a bit earlier. You can decide to have a bit lower decline curve. What we simply picked was a scenario where we said, it's the middle of the year and it's a pretty aggressive decline curve. Could the actual situation be different from that? Absolutely. Could it be better than that? Yes. Could it be a bit worse than that? Yes. So I'm not saying to you this is the exact number and I'm not saying to you this is the absolute worst case. Is it something where there's a, in my view and I think in the company's view, is this something which is likely to be close to a realistic outcome? Yeah. That's why we've made this estimate. And we think it's a perfectly sensible scenario, given all of our experience in the complexity of manufacturing these products and given everything that has to go right for a full-on generic attack on Advair. And we've essentially assumed that somebody figures that out in July, and they get a good hit of the product from July onwards. That may or may not happen. Next question?

Thank you. The next question comes from Jo Walton at Credit Suisse. Please go ahead. Jo Walton - Credit Suisse Securities (Europe) Ltd.: Thank you. I wonder if you could talk a little bit about your Vaccine and your Consumer business? In particular you've already reached the target margin for the Vaccine. So I wonder if you could tell us a little bit about the capacity constraints and how you are looking to free those up for this year, or whether we should assume that effectively all of the extra Bexsero that you were able to deliver, you haven't actually got that much more runway for this year. So just to help us think about that Vaccine business. And to dive a little bit more into the Indian problems and the turnaround or when we should start to see Consumer improve again. Because I think that the Consumer, certainly the profitability was slightly less than expected in the fourth quarter of the year.

Yeah, thanks very much, Jo. I think in terms of the Vaccine business, so we've been investing very significantly in fixed infrastructure, capacity, and also process redesign over the last many years actually. And we are beginning, we saw during 2016 and we continue to see the continued benefits of that. So capacities are going up all the time, as Simon rightly said. In some, for example Bexsero, we acquired Bexsero with a certain demand curve and a certain capital base. It takes a while to adjust to the higher kind of sales level that we're currently running at. So there will be within any given year probably situations where we're not in an unconstrained supply position for every vaccine. And the reality is actually, Jo, that at some price you can pretty much sell 100% of your vaccine output on a global basis. So there is an almost unlimited marketplace. And the pace at which you can expand is to some degree the limiting factor. As we look forward between 2016 and 2017, I fully expect us to continue to grow the Vaccine business. I just wouldn't expect it to grow as fast as it grew in 2016. Now partly that was because we had the effect of Bexsero coming off a very, very low base into a very much bigger base. But partly it was a consequence of having a very big flu season, which may or may not repeat. But I think, we all know flu can be seasonal. So I think there is a likelihood the Vaccine business might grow a bit slower, but it will still grow I think during the year. In terms of margins, I think we've shown really very quickly how we could essentially take the Novartis and the GSK vaccine businesses,…

Thank you. The next question comes from the line of Tim Anderson from Bernstein. Please go ahead, Tim. Timothy Minton Anderson - Sanford C. Bernstein & Co. LLC: Thank you. If I could go back to the HIV category. So in the integrase area Gilead has their bictegravir. And it seems like there's a fair amount of excitement among KOLs about this. And they certainly seem to be taking on dolutegravir head on in a variety of their trials that are head-to-head studies versus your product. And I'm wondering if you can see at this point how you think they're going to try to differentiate their product? I know that once in a while with dolutegravir for example you hear about CNS side effects. I'm wondering if that's an area for example that they might try to go after. And we've got data coming up here, I think this month at CROI, Phase II data. I'm wondering if you have any visibility on that? And then the second question was on your zoster vaccine. You recently initiated a Phase III study to look at the impact of the reactogenicity with the product on quality of life. And I'm wondering if you can give us additional context here? Whether that's just for commercial purposes or if that's actually a regulatory requirement?

Great. Thanks very much, Tim. And thanks for the questions. I think on the HIV front, obviously it's not my job to figure out the positioning strategy of a competitor. So I'm going to resist the temptation to do that. I think looking at dolutegravir, and I've worked in HIV since AZT days in 1989. And the same is true now as it was then. What people want are effective, resistant – effective drugs, which have high barriers to resistance with a good safety profile in the short and the long run. And I think when you look at dolutegravir you've got a very, very attractive molecule there. No resistant isolates identified during the Phase III programs. Extremely, extremely impressive resistance profile with over 300,000 people dosed. And a very, very strong track record of efficacy into the marketplace. And as we've started to develop the combination product, obviously that's opened up a lot of opportunity for further growth. That is going to be quite – once you have a zero resistance profile, it's not easy to beat that, right? I mean it's like an antibody, which is like 99.9% effective, how do you top that? So in terms of the thing that really matters, I think dolutegravir remains an extremely impressive molecule that people may try to look for peripheral things. The reality is what it will boil down to is physicians' personal experience of what they've seen with patients. And, yes, you might see a patient with some idiosyncratic kind of reaction to a drug. But that's true of any drug in reality. My understanding of the CROI data next week – and obviously I have no idea what that data shows. But my understanding of the study that's going to be released next week is in 75…

Thank you. Your next question comes from the line of Kerry Holford from Exane BNP Paribas. Please go ahead, Kerry.

Thank you very much. Two questions please. So firstly, on cost savings and how that related your guidance. So the current £3 billion selling program is nearing an end now. And I notice very little incremental savings this year. And these savings have seemed to have come through faster than you anticipated. But could you see the ability to cut costs further? And then exclusively, is that assumed within your 2017, if you like, worst case guidance for earnings this year? Are you assuming that there's some reduction in sales reps or promotional spend or something behind Advair within that guidance? And then secondly, for the put options there, so following the news on Shionogi today, should we consider that you might be having similar discussions with Pfizer? Do you have similar call options there that you could look to exchange? Thanks.

Thanks, Kerry. I'll ask Simon to comment on those two. Just on the specific of if there were a generic Advair, then there is cost associated directly with the Advair promotion in the U.S., which we would obviously cease. So to that degree there is some relating. It's not a huge amount, but there is some. And of course that is factored into the downside guidance that we've issued. But let me ask Simon to comment on the put and anything else you want to say on cost savings within the guidance structure.

Yeah, thanks, Kerry. I mean I think as I said in my remarks, you should assume the £200 million to complete the program is what we have baked into our 5% to 7% guidance. Clearly as we've done with previous restructurings, we'll continue to look to see whether there are other opportunities. But there's nothing that you should be including at this stage. And as Andrew said, on the downside there is a small amount of cost still attached to Advair. Not a lot, because part of the restructuring we've been doing over the last two years or three years is to get ahead of this moment. And so that will drop out and is factored into the downside. Although not all of it, as some of it will get reallocated to continuing to drive the new products. So the profitability position you're looking at is a net one. And on the put options. I think there's a distinction between Shionogi and Pfizer in that this is a very significant part of Shionogi's business. And I think as we talked with them as shareholders, it seemed extremely unlikely to a nonexistent risk that they were going to ever exercise the put. And we thought it was a good idea to remove that uncertainty. Pfizer is perfectly comfortable with its position, and we're perfectly comfortable with their position. So there's no discussions going on with them at the moment in terms of a similar initiative. And let's see what Pfizer want to do over the longer term.

Great. Thanks, Simon. Thanks, Kerry. Next question?

Thank you. Your next question comes from the line of Seamus Fernandez from Leerink. Please go ahead, sir.

Oh, thanks for the questions. Andrew, I hope you've already gotten the invitation from Richard Branson. But just as a follow up on some of the questions asked earlier. As we think about the opportunity to really grow the profitability of the Respiratory franchise, we keep hearing about declines in pricing. When do you think that this really stabilizes? And the overall Ellipta franchise can actually drive incremental cash flow growth? And then the second question. As we think about – we've heard some good things about your BCMA product. As we think about oncology and sort of building around that, we've seen single-asset oncology areas be quite successful. But do you see GSK really working to continue to build around the BCMA asset? Thanks.

Yeah, thanks very much for the question. And yeah, I haven't been practicing my water.

Kite surfing.

Kite surfing. Kite surfing. So I haven't been doing that. And you probably wouldn't want to see me try I don't think. In terms of your questions, thanks for those. But I think what we're seeing with Ellipta is very – first of all, I think pricing is stabilizing. It doesn't mean there isn't still pressure out there. But I think the lion's share of the adjustments which we needed to make have been made. And to reassure everybody as Advair pricing came down, the Breo pricing kind of came down with it. So we're not sat here with some big discontinuity if and when a generic comes along, which gives us quite a lot of confidence. We've never had more contracted access for our products across the board in the U.S. than we do this year. Extremely good position in terms of access. And if you look at the NBRx as well as the NRx and TRx curves, you can see there continues to be significant upswing momentum behind all of the new products, implying a continued good performance there. So I think from that point of view, very good. Of course we're invested heavily in those new products, as you'd expect. And Advair has been a very generous support during that period of the last two years or three years in terms of helping us fund the growth of the replacement products. But as we move forward, the new products are going to be the ones that are going to be kicking in the cash flow and the profitability. And as you can see, they're all now beginning to move into scales, which you would expect to be very much net contributors as we move forward over the next year or two years. So I think that has…

Thank you. Your next question comes from the line of Michael Leuchten from UBS. Please go ahead, Michael.

Thank you. Michael Leuchten from UBS. One question about your longer-term guidance please, and one question about your net debt profile. So on the 2016 to 2020 range that you've given to mid to high single digit EPS CAGR, does that include at the upper end of the range Novartis putting their stake in the Consumer business to you? Or is that end of the range achievable without? And then on your net debt profile, could you help me think about how that looks like as we go into 2017 and 2018, given the slide you presented on the currency impact on your net debt, the restructuring charges that you talked about, but also Advair potentially going away in the U.S. if we get a substitutable generic? Thank you.

Thanks very much, Michael. I'll ask Simon to comment on the net debt. As far as the shape we outlined in 2015 for the 2020 group, that assumed no change in the Novartis put, no change in ownership of the Consumer company. So if that were indeed to happen, that would obviously be an upside to that scenario. That was not part of the base scenario. I will take the opportunity though, to remind you that an assumption on essentially the more or less total loss of Advair in America during the period was part of the assumption. So the guidance we're giving you today on Advair is absolutely consistent with what we had embedded within the 2020 outline we gave you back in 2015. Simon, do you want to just comment on net debt?

Yeah, on net debt, given the investments that I've described in my remarks, I think we'd expect debt to start to come down in 2017, not by very much, and then fall further as we go forward from there as the cash generation comes out of the other side of the Advair impact. So that's a trend that I've described earlier, back last year. And I think the picture still looks very much the same going forward.

Do you want to just, Simon, comment on how you feel about the currency effect on the debt, given where the debt is versus our businesses? I think that might have also been part.

Yes. I mean the translation effect is very significant. As I called out in the comments earlier, we saw £2.2 billion of impact from currency just in terms of cash and translation of foreign currency borrowings on the balance sheet. And clearly we could take a different approach and hedge those. We don't. We match them up to the currencies that we generate earnings in, so that they are naturally hedged across the businesses. So that translation effect is something we keep an eye on. but it's not driving the economic value or the capital allocation decisions we're making. And the comments I just made assume that currencies don't really move from the January rates that we've got in front of us and clearly that may swing the picture around a bit. But the underlying position peaking at these sorts of levels and beginning to now start to come down as we come out of the other side of the restructuring and integration programs, the impact of the Advair generic, and then into the regeneration of the cash generating capability within the company on the back of the new products in Vaccines and Consumer as we've discussed before.

Thanks, Simon. And I think we just have time for one last question.

Thank you. So your next question comes from Keyur Parekh from Goldman Sachs. Please go ahead.

Good afternoon. One big picture question for you, Andrew, and then one kind of on the HIV business. First on the big picture stuff, given everything that's happening in the world kind of from a pharma perspective, pricing, kind of innovation, FDA, et cetera, how do you see the shape of the industry changing over the next few years? And what do you anticipate Glaxo's role to be in that change? And then secondly on the HIV stuff, what do you think doctors are going to need to see to feel comfortable moving to the dual regimen? Is it going to be longer-term data? Is it going to be more experience when the product's on the market? How do you think that shift happens? Thank you.

Thanks. On the second point I think what they're going to want to see is a high barrier to resistance in a range of viral load patients. And I think that's clearly what we're aiming to show through our trials. I think that's key. What we know in this marketplace, Keyur, is that people are more than willing to try new medicines, new molecules, new combinations based on the typical kind of Phase III trial duration time. So it's really about the data that's generated in those trials. I don't think that this will necessarily require a very different kind of set of experiences. I think people will look at it based on what they see. People are very used to looking at 48- or 96-week data. They'll look for what's happening on the resistance profile. That's what I think will really drive this, just as it always has done. As we've gone up the drug regimen, I think it will drive any choices to come down the drug regimen. And we'll obviously see that as the Phase IIIs start to conclude. The big picture question, good grief, in all the world what do you see might happen. I think it's this industry is one of the most fascinating industries. It's always been fascinating. There's always been a major dynamic going on. I think what we're seeing now more than ever is a kind of global focus on pricing and affordability. And really for the first time in the last two or three years, the U.S. marketplace has really become a central player within that discussion. I think the number one question is whether or not something fundamental changes in the U.S. As I said at the very beginning of this call, there is tremendous complexity and rigidity in the…