Good afternoon, ladies and gentlemen, and welcome to the Analyst Call on the GSK Second Quarter 2019 Results. I will now hand you over to Sarah Elton-Farr, Head of Investor Relations who will introduce today’s session.

Thank you. Good morning and good afternoon. Thank you for joining us for our Q2, 2019 results which were issued earlier today. You should have received our press release and can view the presentation on GSK’s website. For those not able to view the webcast slides that accompany today’s call are located on the Investors section of the GSK website. Before we begin, please refer to Slide 2 of our presentation for our cautionary statements. Our speakers today are Chief Executive Officer, Emma Walmsley; Iain Mackay, Chief Financial Officer; and Dr. Hal Barron, Chief Scientific Officer and President of R&D. We have a broader team for Q&A. We request that you ask only a maximum of two questions, so that everyone has a chance to participate. Our presentation will allow for approximately 45 minutes slightly longer than usual time to update you on our R&D progress. And with that, I will hand the call over to Emma.

Thank you, Seth. 2019 is an important year of execution for GSK and I am pleased that we’ve delivered continued good operating performance with growth in group sales and earnings in our first two quarters of generics competition to Advair. Group sales growth of 5% in CER terms reflected the particularly strong performance in vaccines with Shingrix of course, but also with our meningitis portfolio and we also saw good performance in consumer healthcare. Our group adjusted operating margins were down 1.4 percentage point from a CER basis. We expect tight control of SG&A spend while investing behind on new product launches and increasing investments in R&D as planned as we work to strengthen our pipeline and accelerate our priority assets. On a total basis, earnings per share were up over 100% to £19.5 and adjusted earnings per share increased 4% to £30.5. Reflecting our good start to the year, we are today upgrading our 2019 guidance and Iain will walk you through that shortly. Our free cash flow year-to-date was £535 million in line with our expectation. As guided previously, we expect cash flow to be weighted to the second half of the year. Few years ago, I laid out my long-term priorities for GSK Innovation, Performance and Trust. All to be powered by a necessary culture change. This year, we’ve continued to make good progress with a number of meaningful achievements in the quarter. For Innovation, we continued to execute on our new product launches and have demonstrated strong growth with Nucala and Trelegy in respiratory with the oral two drug regiment in HIV and most notably in vaccines from the Shingrix. Strengthening our pipeline is critical to our long-term success and we’ve made some good progress here also. Among our achievement, we had positive data read out…

Thanks Emma. All the comments I'll make today will be on a constant currency basis, except where I specify otherwise, and I'll cover both total and adjusted results. On Slide 9 you'll see a summary of the group’s results for Q2, which was a strong quarter with 5% group revenue. Overall, this was driven by strong performances in vaccines and consumer, offset by a 1% decline in pharma as expected. Total operating profit is up 80% with total earnings per share up over 100%, reflecting lower charges for the quarterly measurement of the ViiV contingent consideration liability and the conclusion of the Novartis’s JV in Q2, 2018. On an adjusted basis, operating profit declined 1% and adjusted earnings per share was up 4%. We’ll see the drivers behind these in more detail in a moment. We’ve lowered £370 million of free cash flow in the quarter in line with our expectations, and as guided previously, we expect cash flow generation to be weighted towards the second half of the year. On currency, weaker sterling particularly against the U.S. dollar and Japanese yen results in a tailwind of 2% on the sales and 5% to adjusted earnings per share. Slide 10 summarizes reconciliation of our total to adjusted results. Main adjusting items in the quarter were major restructuring focused on the supply chain representing non-cash charges relating to ramp up of the program we announced in July 2018. Within transaction related, our management of the ViiV contingent consideration liability primarily driven by changes in exchange rates. And within disposal column, the main contributors again from the revaluation of the embedded derivative in respective GSK's exposure to movement in Hindustan Uniliver shares price. Our comments from here onwards are on adjusted results unless stated otherwise. Slide 11 summarizes the pharmaceutical business where…

Thank you. In July last year, I shared with our new approach to R&D and made a commitment to being more transparent about the decisions we’re taking and the progress we’re making to regular update like this one. I'll spend the majority of this presentation on our pipeline highlighting the progress we’ve made since we set out new approach one year ago. To recap what I said at Q2 last year, our new approach to R&D is based on the multiplier effect of science, times technology, times culture. We define this is strengthening our pharma R&D pipeline by focusing on science related to the immune system, the use of human genetics and the applications of advanced technologies such as functional genomics, machine learning and cell therapy. I believe we made significant progress over the past 12 months resulting in a much stronger pipeline. I’ll take you through this in more detail in a moment. We’ve also made good progress on our technology strategy, which I will cover at the end of my presentation. And on culture, we strengthened our peer review process with a focus on smart risk taking and our single accountable decision making model. We’ve also hired some outstanding people and established some exciting new partnerships. Turning back to the science, in the last 12 months, we’ve advanced eight assets in the Phase I, three assets in the Phase II and four in the Phase III. In addition, we’ve also progressed three vaccines into the clinic. Importantly, we’ve also gained approval to three new medicines including Dovato, Dictova and the Nucala pre-cell syringe. In this slide, you can see a summary of what I've just mentioned with specific details on each asset. These blue do show progression, green for approval and red for termination. As you can see…

Thank you, Hal. So, as a reminder, we've seen some good progress this quarter on our priorities of innovation, performance and trust; and we are on track with our key areas of focus. We're driving improvements in our operating performance. We are progressing our pipeline with a number of further key readouts to come. We're investing in our executional capability for specialty portfolio and we're working towards the successful integration once the consumer JV has completed shortly. Successfully delivering these priorities over the coming years will provide a clear pathway for the creation of two great companies, one focused on pharma and vaccines, and the science of immunology, genetics and new technologies, the other on consumer health. So, we're now joined for Q&A by Luke, David, Brian and Roger. And with that, operator, this team is ready for everyone's questions.

[Operator Instructions] Your first question comes from Emmanuel Papadakis from Barclays. Please go ahead.

Thanks for taking the questions. This is Emmanuel Papadakis from Barclays. I should probably start with the obligatory guidance question. Perhaps, sustainability of the guidance of great we've just seen in terms of thinking about the run rate for Ventolin and the Established Products will be particularly helpfully reference, some inventory at Relenza benefit in Q2. Just trying to think about how those run rate still looking in the second half will be very helpful? And then maybe on similarly financial topic, free cash flow, you did call out H2 waiting for the year. Given its important to your dividend outlook, perhaps, you could give us just a bit more color on the second half of the year? And do you think it's now feasible with the revised guidance to get to or beyond the level of 2018 free cash flow i.e., just over 5.5 billion? Many thanks.

Thanks, Emmanuel. So listen, both on guidance of free cash flow time they want to…

Absolutely. Yes, Emmanuel, on free cash flow first. Look, the overall outlook that we provided at the fourth quarter, and I think then went over again the first quarter was certainly driven and informed by genericization of Advair and then the timing of pre-genericization rebase and that run-off. And we would expect to see a step down from overall free cash flow in 2019 versus 2018, which saw a lot of progress made in our regard. Overall, the theme currently as it relates to guidance around the profitable that we’d referenced momentarily has a similar read across to the overall guidance from a free cash flow perspective. So, as in previous years, we’d expect to see strong cash flow with generation over the course of the second half of the year we’re very much in line with where we expect to be at the half. And in terms of overall for the year as guided earlier, we’d expect to see a bit of a step down against 2018 informed by Advair genericization principally. In terms of the overall guidance, the guidance that we provided in the fourth quarter and in terms of what we’re driving to down 5 to down 9 adjusted EPS is. Those key factors remain obviously consistent, so the genericization Advair obviously keep the broader impact of the genericization of pricing in the ICS/LABA class is an important factor because certainly when you look at Breo/Relvar in the second quarter and the first half year, we’ve seen that tick across in the pricing and the overall ICS/LABA class. So that’s some of the pressure that we’d expect to see continue over the second half of the year and directly related to ADVAIR genericization. One of the welcome features of the first half was Ventolin perhaps somewhat unexpectedly,…

Thank Iain. Next question please.

Thank you. Your next question comes from Graham Parry, Bank of America Merrill Lynch. Please go ahead.

Great, thanks for taking my questions. So, firstly on vaccine, if you could just help us understand some quantification and the benefit of CDC stocking where that hit aided Shingrix and the right run rate for Shingrix for the year? And then secondly on HIV, I wondered if you can comment on the multiple data that was published at IAS. In vitro data looking at viral resistance emergence with dolutegravir plus 3TC versus Biktarvy suggesting increased viral resistance emergence with dolutegravir 3TC and any thoughts on why that is this quarter with the clinical data suggest GEMINI and TANGO that you’re seeing any critique of that study would be useful? Thank you.

Thanks, Graham. So, I'll ask Dave that to pick on your second question. I’ll just comment briefly on Shingrix outlook for the year. And you remember at Q1, we guided that we thought it would be roughly maintained run rate. For the rest of the year, we would -- as far as Iain has alluded as part of the upgrade, we would expect actually that the second half would be more in line with the full first half. This year, we’re really pleased with the ongoing strong demand for Shingrix and we’re equally pleased with the progress that Roger’s team are making in terms of our capacity expansion. There is a, let just say, a possibility of further progress beyond that, but we will update you on that at Q3. And beyond that and perhaps go to David on the HRD question.

Yes. Thanks, Graham. I mean I’m not really getting into much detail of the logical because obviously it wasn’t our study. Although, as you said, it was a poster I think presented yesterday at IAS in Mexico and it's an in vitro study. What really matters I think and what we’re very pleased about is later today in Mexico, investigators and the team is going to present the GEMINI 96-week data, read out and also the TANGO study which was switch study we did Dovato. And I think and we believe that data would be very positive both from an efficacy point of view and importantly across around the thousand patients in the study on the Dovato arm. We’ve seen zero treatment emergent resistant there in a clinical with patients doing right across the world. I think that would be seen as very reinforcing of durability of Dovato. I mean the one -- so the technical point I would add that will come out in the TANGO data is, one of the things we did is some archived DNA analysis to look for resistance and particularly resistance from the 184V mutation which potentially end cause resistance in lamivudine and therefore make in the Dovato setting the argument as it could make dolutegravir become monotherapy. And we did the archive testing to pick up very low levels of resistance. What was interesting is overall out of the 650 patients in the study there is any picked up seven patients of 184V even at the low level. So in the instance of it is very small for the Dovato arm and none failed. So, I think we’ve now got some very specific evidence around that, which I’m sure you would be very helpful. So, we’re feeling I think increasingly confident of the durability and importance of two drug regimens.

Thanks, David. Next question please.

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

Thank you, guys. Two questions please. Firstly, just following up on ViiV. Thinking about the outlook for the remainder of the year. So in the quarter, U.S. sales were down 6%. I wonder, David, if you can talk to how you explain that to evolve in the second half of the year? And in the first half of the year, that equated to just 1% sales growth globally. Are we still expecting as the ex-U.S. regions to deliver more significant growth to drive that higher? Or are you looking at sort of that remaining similar level for the remainder of the year 1% growth or so? And then secondly on Zejula for Hal, I wonder, I know you can't give us some more details on PRIMA, but if we're to say that data can comment at ESMO. Can you tell us whether or not we would get the detail results by subgroup and here I am talking about HRD positive versus negative. I am also interested in any view you may have on how that data could compare to Pillar 1 for the abstract, if that is also positive. Can these two approaches co-exist? Many thanks.

Thanks Kerry. So we'll go to David first on ViiV and then come back to Hal on the two pronged set of your questions around Zejula. David?

Yes. Okay, thanks Kerry. So, dolutegravir was flat in the quarter in Q2 and not 3% year-to-date, and then of course we have the drag which is about 2% of the older product particularly at the [indiscernible] for entry. So overall that put ViiV into slightly negative territory for the quarter and not 1% from the year. And I think as we said in the U.S. future growth will primarily come from our two drug regimen, Dovato and Juluca. Obviously, we have also in 2020 from cabotegravir, our long-acting for which we have been going to the priory review PDUFA data at the end of the year and also a much smaller amount from fostemsavir. I think we do remain very confident in the growth potential of our HIV portfolio. Overall, it's early days for Dovata first in the U.S. I think overall the launch is pretty much what we anticipated. As Iain said, MBRX is now about 2.5%. A weekly TRx has gone up to just over 350 strips a week. We've got about 80% reimbursement coverage now so that's gone very well and very fast, and overall Dovato is about 50% ahead of Juluca at the same point. I mean of course it is challenging as we've always said a very well established treatment paradigm and it will take time to build momentum. And as physicians gained experience, we're very encouraged by the data that I talked about earlier in response to Graham's question. And obviously, we hope that should have a big impact. Outside the U.S., we continue to grow very strongly; and international, particularly driven by Japan and Russia; and in Europe actually, we gained market share dolutegravir volume up about 8% with a bit of pricing drag there, but that was reduced overtime. So, overall, we remain confident in our growth outlook for the HIV portfolio.

Thanks David. Hal.

Yes. Hi, Kerry, thanks for the question. You know, as you know, we've just un-blinded the data. So, we've only completed limited number of analyses. As you know, we're very pleased about that. The hypothesis that Zejula would benefit women beyond those who have the gBRCA mutation in facts the hypothesis that the HRD-positive patients, those with the genetic defects in the homologous recombination pathway could benefit was validated. And the way the study was designed as once we validated that hypothesis, we could ask the question. Does the overall patient population benefit? So those are the analyses we've done. Obviously, there's a lot of interesting data in the dataset. We will be exploring that, figuring out what are the most interesting questions to provide data on. But as you know, we can't comment further on either the data or the analysis because it's just too early. In terms of [Solo-1] that you mentioned, again we are looking forward to seeing that data as well later this year. And I think it’s important to remember first of all that [Solo-1] is exploring whether PARP inhibition benefits the women with ovarian cancer in the first-line maintenance setting. Actually, we think that Zejula and the PRIMA study I've answered that question. Zejula does benefit women in the front line maintenance setting. The real question that the world will have I think is the incremental value that Avastin adds to PARP inhibition there unfortunately Paolo is not answering that question, but for the 25% of women who are being treated with the Avastin currently will I guess have information as to whether the combination works. I think it’s also important to realize that questions that will be evolving relate to Avastin and the benefit that it has both in terms of the improvement of progression free survival, but also the recent data from the GOG-218 on overall survival where there is really no improvement in the overall survival and there is obviously toxicity and there is cost. So all of this will partly put together in a way that we can digest that and understand how to think about benefiting the women in the first line setting. But again, we’re very excited that hypothesis that the PARP inhibition in Zejula particularly will benefit the women who have the homologous recombination defects and excitingly in the all-comers analysis, the benefit was there as well. So, we look forward to sharing more that data when we have it.

Right, Hal. Thanks very much. Next question please.

Thank you. Your next question comes from Seamus Fernandez from Guggenheim. Please go ahead.

Thanks very much. I guess I just had to ask a requisite question around the Senate's bill and the proposed impact relative to some of those differences that have been proposed. I just was hoping, Emma, if you could just give us your any broad stroke thoughts? And then if possible where GSK kind of fits in the context of that proposal since then I know it’s the moving target, but I think investors would certainly benefit from hearing your thoughts there? Thanks.

Thanks Seamus. So listeners as everyone know, these proposals were published late yesterday. So I think a 40 page document with over 30 detailed provisions, so we need to take a bit of time to analyze this and understand the provisions and understand any implications they might have. It’s also and again just looking at the issue very important in those that all of this is subject to discussion potentially amendment, and even in this ultimately past many as not due to big impact until the reflecting so 2021, 2022. And all that said, of course considering the size and the importance of the U.S. market, we take these issues very seriously. And today, we don’t have significant exposure to part B, but as we’re thoughtful on how our portfolio will continue to evolve, within the pharma business noting that this quarter our vaccines business of the U.S. was about 30% of the our sales excluding consumer. So the key area we’d be looking at net on this is part B, but we’d be thoughtful about any evolution in part B in terms of our approach with new assets. And there are few print support I suppose that we would be supportive of just the broad base rather digitally talking about this specific proposal, and we are absolutely supportive which is why we were also supportive of the previous rebate and discussion. But all things that reduced the out of pocket cost for as many patients as possible in any of the price reductions that we pass onto payers by no way to patients. That’s already key principle, likewise we’re supportive of principles that incentivise responsible pricing, and most fundamentally that incentivise innovation and ongoing innovation and access to that innovation so the patient that need it. So, I will keep, we’ll keep watching it obviously the answer to these kind of environmental context continues to be, make sure you price responsibly. I think if you look at GSK's track record and where its performance has come from that is something that the Company has been long committed to and make sure you innovate or value which is why all the work that Hal has that progressed with, with the R&D team is so important and so we want to outline that with you today. But we will be I’m sure watching this space and we'll see what comes through that. Next question please.

Thank you. Your next question comes from Mark Purcell, Morgan Stanley. Please go ahead.

Yes. Thanks very much for taking my questions. I have two. Firstly for the Zejula, can you help us understand your plans to support the drug outside ovarian cancer on the back of the PRIMA and then obviously the AVANOVA data, which are very interesting and implying a role I guess the potential -- hypoxia caused -- did the damage to the, synergistic effect with this molecule. I’m asking this question because there are some unique advantages of it over other PARP inhibitors including better blood flowing barrier penetration and potential, possibly in tumors such as lung cancer for example. And then secondly on the [indiscernible] and the neutropenia SAEs we saw recently. Could you discuss the risk maintenance initiatives you've put in place? And whether how successful you believe these are in terms of improving the risk benefit of this drug for patients as you move forward into early lines of therapy? Thanks very much.

Right. Thanks. So I think that's both of you, Hal.

Yes. Okay. Good questions, complicated questions, so let me take a few minutes to think about that. But the question about the Zejula beyond the frontline ovarian, I think it's a great question and we have a very robust development plan that includes a number of studies as you highlight in ovarian cancer. I think you’re right the AVANOVA study is very exciting. It is important to realize that was a treatment paradigm study that looked at the combination of Avastin in combination with niraparib as a potential treatment. And this is exciting on a lot of levels. One, we did see a pretty significant effect in a uncontrolled single arm trial, but we’re excited about where that might be able to take us in terms of maybe later line therapy when patients have progressed on their platinum-based front line therapy. It’s really tough on patients to take platinum continuously in the second, third and fourth line. And if we can potentially design studies to find drugs like niraparib and Avastin that would be as good, maybe even better. But certainly with the better safety profile, I think that would be a really significant advance for these women because you know they’re very sick by the stage and likely to die and improving the quality of life is critical. So that’s an exciting approach. I’m also excited by the TOPACIO data, small numbers, but there's some intriguing signals and there again small numbers in particularly when you look at the combination of niraparib plus dostarlimab where we're starting to learn a little bit about what PARP inhibitors may be doing beyond simply blocking the replication fork and the homologous recombination repair process and looks also like when you do that, you actually activate the STING pathways and you almost by…

Thanks Hal. Then the comments on [belantamab]?

Yes, great -- sorry. So, belantamab, so BCMA, you know it's important to remember that myeloma is a plasma cell dyscrasia and so just the disease alone causes thrombocytopenia that’s known. BCMA is known to cause thrombocytopenia, and so the combinations of various drugs where we know there’s going to be additive talks, it’s important to get a strategy to minimize that as you point out. And so without going into excruciating detail, the modifications we’re putting in place are several folds. First and probably most importantly is that these findings are usually dose related and we are using a lower dose in the Phase II studies DREAMM 6 in the IFS, and that should help us figure out whether it's a C-max or C-min area under the curve et cetera. How do you optimize the dosing to minimize this problem as well as are there ways of certainly educating the sites to ensure that when a patient does get some toxicity that are managed well? And we think the combination of doing both of these will result us in having a safe and effective window for providing this drug and in particularly given how active it is we can set the risk benefit is likely to be something we can demonstrate, but of course that’s why we’re doing the trial.

Thanks Hal. Next question please.

Thank you. The next question comes from James Gordon from JP Morgan. Please go ahead.

Thanks for taking the questions. James Gordon from JP Morgan. I have two questions both about Zejula please. And for the first specifically, clarification around the PRIMA study result, but in your press release it says that the study is statistically significant and why there was cross by market data, and in the study the original endpoint was just an HRD positive patient. And then, how we state on the call it sounded like the study due to deterioration of primary that did show benefit in just HRD positive first and also in the whole population. So could you just clarify, what was the primary that was tested to this work to the some population and in all-comers please? And the second question was also on Zejula. So the press release referenced AVANOVA and talked about the strong data we saw, presented at ASCO so Zejula plus avastin. Were you slightly misdated because there is the potential to get this patient labels so you could actually have mono and an avastin combo? I mean that’s why things like or do you think that whatever the mono and avastin combo that’s only the Astra likely tied as a please?

Hal?

Okay, so let me explain the design of PRIMA and it’s a relatively common approach to designing trials and then binding. The primary endpoint was tested hierarchically, so by that I mean the first question you asked of the data is the one that we have the most confidence in, which is the HRD positive population. And the analysis plan calls for answering that question and if positive, only if positive, do you go on and then assess the all-comers? And because the HRD population was positive, we then went on to test the all-comers analysis and what I said in the script is that. That was also positive. So that’s how that works. And it’s done to avoid having to spread or I should say share alpha, but that doesn’t matter. But it’s a common way of doing what's called hierarchical testing. Then the other question was about AVANOVA. Yes, we were pleased with the data. It's very intriguing the combination is as I mentioned is a possible treatment regimen is interesting. I think that in terms of treatment setting it's going to be important clinical data to look at, but the plan is not to be filing that with the PRIMA study.

Thank you. Next question please.

Thank you. Your next question comes from Tim Anderson from Wolfe Research. Please go ahead.

Thank you. A question on PRIMA again, so a short top line press release you said it was clinically significant today. The top line release only describes safety as in line with prior data set, but PRIMA looked at the lower starting dose to try to avoid thrombocytopenia. And from our prior conversations with you guys, I think the goal is to have thrombocytopenia in line with competitors PARPs like in lynparza. So I’m wondering on a safety, if you can say that was achieved in PRIMA with that dosing modification? Second question is the Shingrix opportunity in China. My understanding it will be a cash pay product, no government reimbursement, which is not uncommon with multinational vaccines. If I think of cortisol also cash pay, explosive growth, consumer awareness is high. There is a big different here with your product in that cervical cancer is potentially fatal and shingles is not. So, I’m wondering if you can just layout, what you think is the commercial case for Shingrix will be over time? What the awareness of shingles areas?

Okay. So just to give Hal a break for a second, I’m going to come to Luke to comment on the Shingrix launch in China. Obviously, there is a question both of capacity and market creation from our point of view, but Luke do you…

Sure, Thanks Tim. So, and I think the analogy of the fellow you make with HBD is interesting. I mean if you look without the Shingrix, right now, we're doing 90,000 doses a month. If you go back to a couple of years ago, the awareness is relatively low. So, I think once your product enters the marketplace and companies begin to be active and physician has a solution with indications to become engaged. You’re correct our assumption is that it’s going to be a private market product. If you look at pricing for celebrex and cortisol, again these are very much in line with the prices in the U.S. and that’s our assumption in China. In terms of what would drive that, I mean we’ve looked at this as you can imagine in some depth there is a, few things which are very favorable, you’ve got obviously in all the population, you have relatively concentrated families with children working and also as a pretty well developed appetite for information on the web. So, we think a number of these things are in line with it. I think the broader question in and this is really where fits in the context we supply with Shingrix is, we try to work out beyond the U.S. and Canada and Germany where we now which the market that we would enter and right now the focus is likely to be in the private market.

Thanks, Luke, Hal.

Yes. It’s a good question. So, the PRIMA study as I said the primary end point was in the HRD positive and subsequently all comers if positive. There was as you point out and it’s an important subgroup but a smaller subset of patients who are randomly allocated towards the end of the trial to the so called weight complete dosing regimen where the women whose body weight was less than 77 kilograms and whose platelets were I think less than 150,000, were administered a lower dose 200 milligram. So I should point out that the concept of lowering dose for lower body weight patients particularly those with low platelets would make sense given this is non target effect. And in fact clinician often do this in the real world, it’s not uncommon to see patients actually identified and do those but we thought it was important in the PRIMA study to actually study it in a controlled way. As you point out I can’t comment further on the data in terms of subgroup or sub-analyses now, but all of this data or much of this data were presented at coming scientific meetings and we will look forward to discussing with you then.

Thanks Hal.

Luke, did you have anything you want to add there?

I mean I think it's interesting as you look at usage rates in the U.S., where we have the best data I mean obviously patients start on 300, around 50% automatically drop down to 200 mg immediately so physicians are adapting the behavior in line with the protocol.

Thanks, next question please.

Thank you, and your last question comes from Jill Wilson of Credit Suisse. Please go ahead.

Thank you, two quick questions. On the vaccine, I wonder if you can help us on the improvement in profitability. How much of this we should build in to our future modeling because it comes from the leverage of Shingrix, which you know doesn't look like it's going away and therefore should give us a strong sustainable base? And how much could be attributed to one-time practice? And the second question for Emma on the new marketing and scheme whereby you are paying -- you're paying your reps more in line with the prescriptions that they are responsible for? How are you going to decide whether that is successful or not? and what should we look as shareholders investors to see how that is maturing?

Thanks very much. I'll let Iain comment on the vaccine margin delivery and outlook and then I'll come back on your last question, Jill.

Great, thanks very much. I think overall, clearly, we're very pleased with the PARP performance to vaccines business and Shingrix is a standout in that regard. With the margin where we have it now, obviously in the high 30s, that's informed by an effect in the first quarter and the inventory adjustment, which we mentioned. And also you'll see that both SG&A and R&D fairly tightly controlled within that business. As we move forward and continued, we probably see a step up to some degree in SG&A to support as we gradually moved to expand market launch from a Shingrix prospect we go forward into longer-term. And then also from an R&D perspective, as Howard mentioned earlier, there's a couple of interesting priority assets coming through vaccines where the step up in R&D to support that development are key features. So in terms of the medium-term, from an operating margin perspective, we very much see that margins for vaccines in the mid 30s, notwithstanding the very good performance that we've seen on higher volumes and good cost control coming through the first half of the year.

Okay, thanks Iain. And then finally just on the update of our policies which as you know has been around sales force incentive, but also around engagement and with HCPs so that they can hear from practicing HPCs where we have new data. We've laid out several key priorities for the Company. We know we want to deliver an improvement in operating performance and we want to deliver a much stronger pipeline and I've been pleased to be able to demonstrate. There is always lots more work to do and some progress on both of those. Within that we are talking fundamentally about the shift in our portfolio to more specialty medicines and the update to sales force incentive was linked to that. We are building under Luke's leadership an ever strengthening specialty team and we want to be able to attract the very best, including in the field and there is specific expertise around that competitively. And as you know when the initial policies were rolled out, there is no follow on from other players in the industry. So we think this is very important, but it contains to our specialty workforce in a certain key geographies particularly where we’re launching new assets in the same way as our HCP engagement is comprising, physicians is contained to new data either or around existing products or new launches. And the answer to Hal knows whether it's working to make sure that we are successful with the new products that we bring to market over the label we expand or new launches we’re talking about coming through in 2020 and beyond. And that’s why we're delighted to seeing some of the progress we had on data that we’re going to be filing with six new registrations hopefully in the next six months. And Luke's building out the team that he has, so we’ll see with the numbers that come through. So, with that everybody, thank you very much. Thanks for joining the call. Obviously, it's a busy day today and we look forward to talking to you soon. Thank you. Good bye.

Thank you. That concludes your conference call for today. You may now disconnect. Thanks for joining. Enjoy the rest of your day.