Ladies and gentlemen, good morning or good afternoon. Welcome to the Sanofi First Quarter 2018 Earnings Results Conference Call and Live Webcast. I am Emma, the Chorus Call operator. [Operator Instructions]. The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to hand over to Mr. George Grofik, Vice President, Head of Investor Relations at Sanofi. Please go ahead, sir.

Good morning and good afternoon to everyone on the call. Thank you for joining us to review Sanofi's first quarter results. As usual, you can find the slides of this call on the Investors page of our website at sanofi.com. Moving to Slide 2. I would like to remind you that information presented in this call contains forward-looking statements that involve known and unknown risks, uncertainties and other factors that may cause actual results to differ materially. I refer you to our Form 20-F document on file with the SEC and also our Document de Référence for a description of these risk factors. With that, please advance to Slide 3, and let me introduce our speakers today. With me are Olivier Brandicourt, Chief Executive Officer; and Jérôme Contamine, Executive Vice President and Chief Financial Officer. Olivier will discuss key highlights of the quarter, while Jérôme will review the financials in detail. We'll then follow with a Q&A session. Joining us for Q&A will be Olivier Charmeil, Executive Vice President, General Medicine & Emerging Markets; Karen Linehan, Executive Vice President, Legal Affairs, and General Counsel; David Loew, Executive Vice President, Sanofi Pasteur; Alan Main, Executive Vice President, Consumer Healthcare; Stefan Oelrich, Executive Vice President, Diabetes and Cardiovascular; Bill Sibold, Executive Vice President, Sanofi Genzyme; and Elias Zerhouni, President, Global R&D. With that, I'd like to turn the call over to Olivier.

Thank you, George. Good morning, good afternoon to everyone, and welcome to our first quarter earnings conference call. Moving to Slide 5. We can see we delivered first quarter results which were in line with our expectations. Our financial performance reflected the expected first half headwinds that we outlined on our earnings call in February, but also disciplined expense management. At constant exchange rates, our first quarter sales were down slightly to €7.9 billion, while our business EPS increased by 1.4% to €1.28. If we also adjust for the acquisition of Bioverativ, our sales would have declined by 1.1% at constant exchange rate. On Slide 6, I want to give a clear picture of the headwinds and tailwinds in our business which impacted the first quarter and support our confidence in our return to growth beginning in the second half of '18. So as you can see here, in Q1, we faced a drag from the losses of exclusivity of Lantus and sevelamer in the U.S. Even with these headwinds, we managed to hold sales relatively stable. This is a result of good growth across our global operations, which - with especially strong performances in Emerging Markets and Specialty Care and the addition of our rare blood disorder portfolio. Now when we look to the remainder of 2018, we expect the loss of exclusivity headwinds to diminish, especially after Q2, when the sevelamer generic impact begins to annualize. Furthermore, we expect sales to accelerate due to Specialty Care and to return to growth in vaccines in the second half, which expected to more than offset the LOEs. And in the rare blood disorders, we'll fully consolidated Bioverativ, which, as you will hear later, has continued to grow very rapidly so far in 2018. So taken together, I'm confident in delivering…

Thank you, Jérôme. So to summarize our performance in the first quarter. We're well on track to meet our full year guidance base and to return to growth in the second half. We made good progress on new products with the filing of Dupixent in asthma and our new oncology drugs, cemiplimab, as well as the landmark results of ODYSSEY OUTCOMES. And lastly, we leverage our leadership position in Rare Disease and established an exciting new growth franchise in rare blood disorders. So sitting here today, you find us confident about the future. Now before I hand over to George to start the Q&A, I want to say a few words about Jérôme and Elias. You will have read that each has taken the decision to retire in 2018. We will certainly miss their wisdom and vast experience, but each is leaving knowing he has put Sanofi in a great position to drive future growth. Jérôme has strengthened our financial procedures and cost discipline and been a major force behind the reshaping and repositioning of Sanofi. We are fortunate that he will stay on to assist in the search for his successor and to help the transition to the new CFO. Elias, too, has made a major contribution to Sanofi by successfully transforming our R&D and strategy, as you heard at sustainability - sustained innovation day in December. We look forward to you meeting his talented successor, John Reed, who many of you may know from Roche. However, today, we must give great thanks to both Jérôme and Elias for all of their considerable achievements. With that, over to you, George.

We will now open the call to your questions. [Operator Instructions].

[Operator Instructions]. First question comes from the line of Tim Race with Deutsche Bank.

Okay. First up, just on the progression of earnings this year. We obviously were helped out on the net income line that due to this [indiscernible] €76 million one-off. If you strip that out and just look to the progression between 2Q, 3Q and 4Q, are you saying that second quarter is going to be worse or just as bad as the first quarter in terms of growth? Or you expect actually sequentially second quarter to have a little bit more momentum than third quarter and fourth quarter? Be interested to know exactly the shape of that this year. And then just perhaps moving on to the share buyback, two parts, really. First of all, you didn't really increase guidance despite the share buyback. I know it only has a small effect, and it's only the first quarter. Can you just comment on that? And then second on that point, you mentioned that you feel very confident that's why you did the buyback. Does that mean that you're going to draw a line on the share price here and you'll do further share buybacks if the share price comes under more pressure?

All right, Tim. I think the first question around shape of earnings, Jérôme, do you want to answer that question? Jérôme Contamine: So thank you, Tim. So as you expect, we want - we're not guiding on a quarterly basis on net profit, even if clearly we have said and reasonably said that second half will do better than first half mainly due to top line growth as we widely explained again and again today and as the gross contributors organic post acquisition of Bioverativ will definitely more than offset the diminished headwinds coming from diabetes, from [indiscernible] So I think that's really the picture you should have in mind. I think the first quarter shows our ability to manage our expense lines and efforts to generate good profile. When it comes to the second quarter, I mean, you've heard from Olivier headwinds and tailwinds, which are going to happen. So I think it's a bit early really to guide you more precisely in Q2, but it's clear that the bulk of the growth is backloaded on profit clearly in the second half of the year.

All right. Thank you, Jérôme. So on the share buyback, Tim, first of all, we're not going to change the guidance during the first quarter. We still have three quarters to go. And the share buyback amount, starting in - after our show - annual shareholder meetings and going up to mid-May next year, we have certainly a relatively limited impact on our 2018 earnings. So that's what we have to keep in mind. However, we decided to do it because we want to continue to have a balanced capital allocation strategy, which is basically returning capital to shareholder as well as strengthening our position for sustainable growth, as we did with the two acquisitions during the first quarter. So despite the expected increase in debt due to these acquisitions, we have considered several factors. At the current share price, share buyback is a very good investment. We also anticipate the disposal of our European generic business, which we have announced recently, Zentiva, should generate net cash in the magnitude of €1.5 billion to €1.6 billion. We have kept also a strong balance sheet and credit rating. So overall, we consider our stock to be undervalued, and therefore, again, the share buyback being a good opportunity. And that's basically what we - what we'd say about the share buyback.

The next question comes from the line of Vincent Meunier of Morgan Stanley.

The first one is on Dupixent. Can you give more color on these inventories movement and also the market dynamics in order to help forecast the sales ramp-up for the product for the remainder of the year but also for '19? And the second question is a broad question. With the management changes, I mean, I know it's may be very early, but do you have an idea of the type of profile for the new CFO you will be looking for? I mean, would you prefer to select a CFO with an operational skill and background or more of an M&A background?

So Dupixent. Bill, do you want to answer that question?

Yes, sure. So first of all, just to reiterate what Olivier said, we believe that the Dupixent launch, we're extremely pleased. And it's going very well. All the signs indicate that it remains a strong launch and that we are, as he said, it just at the very beginning. Underlying demand increased by about 25% with total scripts and patient numbers and the number of patients that are added each week continues to be above 500 per week. So the signs are extremely strong. And what happened with inventory is we ended Q4 relatively higher inventory position due to the fact partially that we began only shipping to wholesalers any inventory at the very end of Q3 and beginning of Q4. And so when we finished Q4, we were at about five weeks' inventory, which was on the high end of normal. And we finished [indiscernible] Q1 around 3 weeks. So while there was some inter-quarter volatility in inventory level, we think that this is something that we're through now. We would expect as so early in launch continue to be some inter-quarter variability, but certainly not at the magnitude that's being compared between Q4 and Q1. As far as from a payer perspective, no real changes. We think that we have a very strong position with the payers. Access - patients continue to get access to the product. We see that as people are having their prior operation - prior authorizations approved and are ultimately getting product. Just to remind you of some of the numbers, we had 89% of commercial lives having established utilization management criteria and 40 - 46% of these lives are covered two labels with only one topical failure required. So all in all, we're again very pleased with the progress of the launch and look forward to the continued success throughout the year.

Thank you very much, Bill. Vincent, for the second question, we want everything. So I'm going to compliment Jérôme again. But we want a profile which has both from operational experience in driving the daily financial activities of very large multinational corporation, but at the same time has a very good understanding of what corporate finance is. So we're going to put the bar very high. And we'll see where we're going to that. But at this point, I wouldn't favor one versus the other. We are again looking for the best potential successor to Jérôme.

Maybe just a follow-up on that front. Is it fair to assume that in the short term the priority is on the integration of Bioverativ and Ablynx?

That's a good question, and the answer yes. And we have progressed very well already. We're doing a soft integration for now of Bioverativ. As you know, Ablynx is not closed, and that process is working very well. We are planning for a full integration beginning of 2019. For now, John Cox, CEO of Bioverativ, has - which has become a Sanofi company, is still driving the business and again helping to integrate within our current or future - or current Sanofi organization. And again, that's going very well. And John Cox - I asked John Cox to be on this call, too. He has not been mentioned by George. But in case you had - since he has been so close to this business and know so much about it, he is on the call to answer any potential question related to Bioverativ. So thank you very much.

Your next question comes from the line of Peter Verdult with Citi.

Peter Verdult with Citi. Question for Elias [indiscernible] frontline assets and more general question for Olivier please. Just on the GLP1 - asset, are you willing to discuss in a little more detail the efficacy as well as severity when [indiscernible] GI tox you saw in this phase II study compared to what one would expect with GLP1 agent? And then secondly, it seems like your third is - asset in the pipeline. I just wanted to know - I know we haven't seen the data yet, but at what point of differentiation you'd like to highlight, if any, versus some of the other development? And then more broadly, for Olivier, on rebates and discounts. Olivier, - both talked about increasing the discounts and [indiscernible] in 29 having a fairly material impact on their respective businesses, wondering if you could quantify what you see the impact being for Sanofi. They're my two questions.

Peter, we have two significant primary endpoints that we needed to achieve on the efficacy side, and we actually achieved more than our minimum primary endpoint by far on the GLP-1 and the reduction and [indiscernible]. We're also seeing weight reduction not far from our points from our primary endpoint. The real issue is truly tolerability and the dropout of a 1/4 of the patients because of tolerability issues. And so as you know, if you have - we have a higher GLP-1 effect in the molecule than we expected from the primate data, and we think the titration was too abrupt. And so that's what we're working, adjusting the titration, the dosing - ultimate dose. We will know that. We're doing the trials now. We are looking at very innovative imaging studies that allow us - image receptors so that by the early summer, late June, early summer, we'll know a lot better.

In terms of the third, frankly, the reason we're optimistic is because of the efficacy profile and the safety profile at this point and the dosing is quiet effective. So we think it's differentiated primarily because of the relationship between the nature of the molecules, the dose relative to its toxicity levels. So we'll find out towards the end of the year. That's why we are optimistic about that one.

All right. Thank you very much. You referring into the bipartisan budget act, to close the Part D the year early, so in '19 versus '20. Frankly, we believe and I believe, like many of my colleagues that, that is a wrong approach because again, this new policy shift cost from health plans to the industry. And we draw our companies covering now - not now, but next year, starting next year, moving from 50% to 70%, which again, a very minimal long-term health to beneficiary. So that's the first thing. The second, if you're asking how much is it going to cost us, you have to take into consideration the business and the decline in our diabetes business in the U.S. and the proportion of sales to the part D channel is, of course, expected to decline in the coming years. So by 2020, for us, we estimate that our U.S. sales would be impacted by around 1% as a result of that new legislation regarding the coverage gap.

Olivier, sorry, just one clarification. That 1% of your diabetes - U.S. diabetes sales or 1% of U.S. sales?

No, no, no. I'm mentioning the diabetes sales because it's a large proportion from the U.S. Medicare, the sales, but the 1% is total sales. Jérôme Contamine: Total U.S. sales.

U.S., of course.

Our next question comes from Stephen Scala from Cowen and Co.

It appears on Slide 16, but there wasn't anything in the regulatory update in the release on the status of the rolling filing in the U.S. for some of the math. Perhaps it has not been accepted yet, but the release does discuss sotagliflozin, which hasn't been accepted yet either. So is there an issue with the filing of - And secondly, there appears from the ASCO titles that there's not much at ASCO on the drug. Is that accurate?

Steve, there's no issue on - It's ongoing. We have no signal that it won't be accepted as we are all going to the rolling forward. And as you know, a type 1 diabetes. I think it takes 60 days, and it's ongoing. So as far as I know, there is no regulatory hiccup there. And when you say ASCO, I think we're presenting data at ASCO [indiscernible] I don't have the data in front of me, Steve. I can't give you an update on exactly what those expectations are, but no regulatory hurdle that I know of.

The next question comes from the line of Luisa Hector with Exane.

So on Dupixent, I wondered if you could just comment a little bit more on existing patients, to what extent they are returning to therapy, the persistency of treatment and also whether you're still very much in that most severe patient group or perhaps moving slightly down the line and slightly less of your patients over time? And I wondered on the aisle 33, in the [indiscernible] you've started a Phase II. Could you comment on when we might see some data? What's the profile that you're targeting with this drug? And whether you might need, in fact, have a head-to-head ultimately with the newly launched injectable assay drugs that are coming to market now? And if you're able to give any guidance just now that you separate the corporate cost, the €3 billion amount, just any color on what how we might model those cost lines, ultimately perhaps some decline with cost savings in the near term?

Thank you, Luisa. Bill, Dupixent; and Elias, you take the aisle 33 and I'll take [indiscernible] the €3 billion. So why don't you start, Bill? Yes.

So first of all, with Dupixent, the question of the type to patient. When we started launch, it was tended towards the more severe patients. And I think that was more due to physicians essentially triaging the most serious. Not what we have seen is that there is a more of a balance between moderate to severe patients than had been at launch. This is exactly what we would expect through progression of the launch, perhaps a little more severe and moves back to the moderate to severe, especially as our educational efforts continue and as we recently started the national direct-to-consumer campaign, which is focused on disease awareness. With respect to persistence, as we look at where we are launch date, we estimate that it's about 83% persistence since launch. And we reported in Q4 that it was about 90%. And you would expect, as time goes on, patients are on longer, that number tends to come down a little bit. However, what we're seeing and believe in comparison to other products, other dermatology biologics, that we will continue to have a very good persistence rate in the future.

Thank you very much, Bill. Elias, 33, any update?

Yes. First of all, the reason why we are using IL-33 is based on several genetic findings. IL 33 is identified as downstream area component of the both the Th1 and Th2 pathways. It tends to be active based on a translational studies in the most severe patients both in asthma who do not respond to typically just Th2 type diseases as well as in COPD. So the recent we're developing that in both asthma and COPD is because of the - our understanding the positioning of IL 33 and its potential in certain patients to be combined with [indiscernible] or in monotherapy. In terms of timing, we've already started the trials. We're going to read out in asthma in Q2 or Q3 '19 and then - COPD in 2020. [Indiscernible] 2020.

Thank you, Elias. On the €3 billion, what you've seen, we have realigned our reporting to our management structure and organizational structure. And as a result, we put all those costs related to global support functions. And there are many of them there in that €3 billion bucket. However, we do not expect the €3 billion of cost associated with those support functions change in the midterm. And the reason is you're going to see some of this support functions like medical and external affairs. They may require additional investment and some others may require less. So that's the comment I would make about that specific bucket. However, you know we are paying a lot of attention to cost savings in general, and you have seen that in the first quarter of '18, we have continued to show a very strict cost management, as you have seen our OpEx line growth, which was just at 1.6% before the impact of Bioverativ. And our gross margin, which declined only very slightly despite the adverse impact of LOEs. And one thing we mentioned in February, showing that we're very serious about - that cost line, we have appointed Dominic our head of business transformation, being a member of the executive committee. And he is now in charge of accelerating the transformation of the company. And we expect to give you an update later this year about our cost savings. So that's what I would - thank you very much, Luisa. Next, please.

The next question comes from the line of Graham Parry with Bank of America Merrill Lynch.

So firstly on Praluent. If you can just give us an initial feedback you had from payers on the pricing proposition for better access. What sort of reception and what kind of uptake are you expecting across the piece there? And could point-of-care rebates help at all here in terms of lowering patient co-pays? And secondly on Dupixent, could you just give us an exact quantification of how much of the €30 million stocking versus patient assistant and the extent to which that expect that patient assistance to diminish over the course of the remaining quarters? Should we expect similar amounts or less by the end of the year?

Thank you, Graham. So maybe we start with Dupixent.

Graham, it's Bill. so the relative weighting is about 2/3 inventory and about 1/3 due to the impact of patient assistance. And just - I think I got your question about will that diminish in time, the patient assistance. Yes, we expected to, in fact. We saw it through Q1 that it started to diminish and come back more to an expected range. And we would expect that throughout the rest of this year, which is typical with any of these products as patients hit their out-of-pocket maximums and so forth.

Thank you, Bill. So Praluent, Stefan?

Graham, the initial feedback from payers is positive. Obviously, I think they share our excitement on the data on the ODYSSEY OUTCOMES and the mortality benefit that we see with the drug. What you know we've been public about this, that we're willing to make some compromises on price in exchange for better utilization management. That's something that we're right now discussing with them. We expect that the current, still very high initial fill rate rejections that are in the order of magnitude of 65% to 70% of rejections for new patients come down significantly as a consequence of that. So stay tuned. We will update you certainly on the next call where we land on this one, but we haven't - having a lot of discussions on this right now.

Yes. So we are in the middle of discussion Graham, as Stefan says. It's just making sure we are improving areas of patient affordability, which was also a little bit your question. So we can't really make any comment on the pass-through. But hopefully, that will happen and alleviate the burden on the patient at the end of the chain here. Enhance patient access, of course, by streamlining, creating more favorable UM criteria, as Stefan said. And we're really, really pushing for that in our current discussion. And in order to obtain that, we're ready to provide, as we've been very public about it, an attractive price that is reflective of what we have announced with ICER back months ago. As Stefan said, more to come on this one. It's a little too early to give you a full detailed briefing on that. So thank you, Graham.

Next question comes from the line of Tim Anderson with Bernstein.

A few high-level questions, please. The first is on M&A. Can you say whether doing a larger deal is essentially off the table at this point given the two smaller deals that you've done year-to-date? Second question is just the structure of the company. You're getting rid of the European Generics business. Can we assume there will be no other divestitures or disposals likely coming? And then last question is for Elias. As you prepared to depart the company and you look back at the years that you've been there, what do you think has gone well? And equally importantly, what didn't go as well as you had originally envisioned that you'd hoped to achieve?

All right. So I take the first two. I will answer by saying that on the near term we're focused on the successfully integrating the two companies we just purchased. And while the second one not being closed yet, but hopefully, it's not too far from now. But Bioverativ and Ablynx, that's going to require a lot of attention and focus for now. Second, as you remember, we always talked about the €20 billion envelope we wanted to dedicate to organic growth. We basically have spent about €13 billion of that. So it's - leave on the table, potential headroom if we find attractive bolt-on opportunities, which like we did with those two companies strengthen our core strategic areas. So staying very, very much in what we have said back in November 2015. So it's an indirect way of responding or answering your question on M&A. And on divestiture, it's a little bit similar. We are doing what we said at the time. And of course, we are saying what we do. And when it comes to divestiture there was two, which we highlighted. Animal Health, and we did the swap. And we've also been very clear in the last two years about the divestiture of our Generics business because we didn't feel it was core to what we wanted to do in the future. Beyond that, there was nothing at the time, and we have not added any additional divestiture. So Elias, do you want to...

Sure. So Tim, what went well? Do you have enough time? But no, let me just be - frankly, I think the best thing that we were able to do was to reshape the R&D organization. It was very fragmented. We're able to break it down to just three major sites: Boston, Paris, Frankfurt; reduce the headcount sufficiently to free up resources, to do what we needed to do. I think what we did well was really partner well when we have those empty pipeline and execute outstandingly on it. If you really look at - at our productivity between 2008, 2012, the company had three launches. 2012 to today is 13 launches. And some internal, like Aubagio and some in collaboration. Remember, that we managed the development of the products that we develop where we research with Regeneron. So you have Praluent and IL-6 and - and Dupixent. So I think it's been a very good turnaround in terms of execution. In terms of significant strategy changes, I would say the turn to biologics at a time when the company was a completely small molecule-driven company, has been successful. It's good today, but it's more great. It needs to continue. But what we did within that was to follow the strategy of being able to develop our own proprietary platforms and especially understand what we call multi-targeting. In other words, come up with molecules that are smart enough to attack multiple targets. And Dupixent was our first choice there because it attacks the targets. And you'll see our tri-specifics, which are internally developed that come up. And all of the things we're doing in dual agonist, triple agonist are going to be happening. And last good thing was being able to convince and recruit and attract top talent, including John Reed, who sees a lot of potential. Otherwise, he wouldn't be coming here. Now what didn't go so well is the length it took and time it took to truly do the social planning in Europe. It takes a long time, and it affects research operations. So we weren't done with that process until 2015. As you know, in research, it takes a longer time to mature. So I wish that had gone faster. I think I wish we had also picked some external targets that I thought were exciting, but we couldn't do it because of the constraints of the loss of exclusivity. That, I think, would've helped us. But overall, I think we've come to a place where, I think, the organization was frankly not dysfunctional to a functional place. It's on the highway. I think the development machine is performing above standards of industry. Clearly, I think the emerging portfolio, I hope, will be my legacy, as you'll see it, and time will tell.

Thank you very much, Elias. So next question, please.

Your next question comes from the line of Jo Walton with Crédit Suisse.

Just a couple on products and one on expenses. I wonder if you could outline what you think the opportunities, as I understand that that's now actually being launched? And secondly, for Lemtrada, I - would you characterize the majority of the sales in the first quarter is essentially the second doses from patients that you've attracted some time ago, which means that you've got a very small patient base to go forward? You did say that you hope that Lemtrada would keep its share in a growing market for high-efficacy products. But presumably, what we're seeing is with - coming through, that those high-efficacy patients are moving elsewhere. So just a little bit more on Lemtrada, please. And on the expenses, you've highlighted that you're able to keep your R&D and SG&A growing at only 3% to 4%, including the addition of Ablynx and Bioverativ this year. And in the quarter, we only saw less than 1% growth in SG&A on a like-for-like basis. I know it's tough to look out in the longer term, but with all of your cost-reduction programs in place, and bearing in mind the mix change of your business towards more specialty products, would it be fair for us to assume that your operating cost expenses should be able to be kept below the 3% to 4% going forward quite comfortably, assuming you don't make other acquisitions? So it's a more longer-term cost discussion that I'd be interested in.

Okay. Thank you very much, Jo. So Admelog, we officially launched on the second of April this year, and I'm going to ask Stefan to give you an update. Stefan?

So yes. Thanks, Olivier. Jo, the product is now launched in the U.S. We also introduced a product previously in a number of European countries. So while we do not specifically comment on the precise commercial potential of Admelog, we're making good progress, especially in the channels of managed Medicaid, where we have a very strong interest from basically all major plans and where we've assigned a couple of first agreements here. So we expect of sales for Admelog during this year. Remember that we had advised previously that we see the commercial opportunity in other channels more in the 2019, '20 time frame as the contracting for Medicare 2020 will be influenced by this - for commercial plans for 2019.

Okay, okay. Thank you, Stefan. Bill, Lemtrada.

Stefan, thanks for the question. So just to put the growth in perspective as well, both adding new patients or just burning through all patients. In the 12 months prior, Lemtrada, globally, so we are adding new patients. So now the challenge is the - both of our products react differently to the market dynamics that are going on. So as you've heard me say previously, the high efficacy segment is growing rapidly. In fact, in the U.S., it's over 48% growth and now represents over 20% of total treated patients. And that is good - a good long-term trend for Lemtrada. However, most patients are currently moving - they're moving to high-efficacy therapy are initially being started on - Now the MS prescribers, the neurologists, are very cautious, and many remain reluctant to make that switch to Lemtrada based on the REMS program and product label in the U.S. And then given its durable effect, Lemtrada revenue is mostly driven by new patient starts. So when you get a competitor in the market like - there's an immediate effect on Lemtrada sales. And the other thing with Lemtrada, as you know, it doesn't benefit from ongoing prescription from treated patients. And our patients turn over essentially every two years. And that's unlike any other product that's chronic product and is chronically dosed. So those are the trends that are going on. And we believe that the revenue will be hampered until more patient switching to high efficacy switch to Lemtrada and/or more patients switch from other high-efficacy products to Lemtrada.

Thank you. Thank you, Bill. Jérôme, you want to answer the expense question? Jérôme Contamine: Thank you, Olivier. I think - I mean, as you know, as Olivier has highlighted earlier in this call, we are working on some further cost efficiency and savings. And we said we could come back a bit later this year to give you more precise objectives. Now a way to answer your question, beyond '18, is based on what you say, that I think we clearly assumed that our OpEx is going to grow with the limit of the variation on a quarterly basis, but they're going to move less [indiscernible] I think that's how I would put it. And I think that's how I would look at the coming years. And of course, then this would be driven, like you said, by us moving to more specialty drug as well as benefiting from the cost savings that we are generating.

Okay. Thank you very much, Jo.

The next question comes from the line of Jack Scannell with UBS.

Two questions. The first for Elias Zerhouni. You gave a very interesting interview, which was published a day or two ago on where the industry is spending its R&D money, arguing that some targets are becoming much too crowded. I wonder, with the indiscretion that your approaching retirement allows, could tell us which particular three targets you think are most horribly overcrowded in the industry at the moment? And then secondly, could you give us some more detail on - filing in non-small cell lung cancer? Patient 19 of your report suggested a 2019 filing. If that happens, but - on what basis - so what would be the comparator data? And what would be endpoints in which you'd be filing?

All right. Both for you, Elias.

I think the comment I made was really related to the utilization of capital within the industry. And frankly, the fact that you have many, many different targets and what happens when you have that is that you basically do not have that study that you used to have for a single product. So the dynamics in the macro sense is not favorable because you invest a lot around the same targets. When the first one gets to the market, there is - they're followed very quickly by others. So your PCSK9 is a good example of that, that you're living in. I think the PD-1s are a good example. So PD-1, for example, to me, we're third in that bucket, but we do have plans to combine it with other targets that we developed, [indiscernible] TGF-beta. I think what happens is that the - from the policy point of view, that's what I wanted to say, is that sometimes capital allocation and return on this capital, because the crowding is suboptimal relative to what it could be if you had a different world. So I think what you see is a very short time. And it's not - we're not exempt from that relatively short time, where you have - that you have today to recover your overall investments. So you're winners and losers. That's no question. But overall, I think the industry needs to be careful about that. Now that's - how - what the solution is, I'm not going to speculate. Talk to me after I retire. And the second thing is on - So what is our strategy? Frankly, to us, the PD-1 is a mainstay of immuno-oncology combinations. We have many other combinations. And frankly, the non-small cell lung cancer, we believe, being the third PD-1, that if we are able to demonstrate the equivalent - we will - efficacy beyond what we have today, we will - to the best-in-class, KEYTRUDA, I think we will have a foothold. However, the follow-up and the follow-on to what we're doing here that is important. I mean, where mix - we're combining PD-1 with CD38, for example, with TGF-beta, with other targets. So to me, it's basically having a foothold in the non-small cell lung cancer space. 2019 is what we think we can achieve in terms of submission given the recruitment rate we're experiencing. Hope that helps.

The last question is from Philippe Lanone with Natixis.

Two, if I may. First, on Bioverativ and hemophilia. Now we have Hemlibra in the market, and it seems to be taking off well. That's what Russia is saying. So I wonder whether you're confident that this is in line with the plan you make when you bought Bioverativ and you still have the same assumptions on the non-inhibitor patients for which it will be available at the end of the year? And the second question is on diabetes. Where - now we have the dual agonist, which is kind of a drift. And you have two candidates. One is - coming early in the market in '22. So I wondered where you will be interested in doing some acquisitions like - in licensing in this area. Consider that the midterm is sufficient? Or if you can probably increase speed on the other agonist that you have in Phase I, which is the -

Thank you very much. So we're going to start with diabetes. We never said no. If there was an opportunity, we committed to diabetes, because you've seen what we have decided since 2015, and we're moving full speed ahead with something - [indiscernible] and we're launching in type 1 diabetes, as you know. You're right. Our strategy was to get - using those two assets as really stepping stones in order to get through new mechanism of action or new combinations such as a dual agonist. I'm going to ask Elias to talk about the backup programs we have when it comes to dual agonist. What you heard is not the end of the story. So I just would like - can you, Elias, please? And then the - I'll ask - the Bioverativ question, and John Cox, as I mentioned, was on the call to also give his opinions. So Elias, if you can?

No, I'll go quickly. So on the dual - our strategy for diabetes is sort of go beyond the next wave - I mean, the current wave of products, which is single-target, single mode of action, is to really go into what we call dual, AAA and even quadruple action points. So we do have backups for each one of these programs. As I told you, I think that our current dual agonist is not hopeless because it came very, very high above the diabetes endpoint - target profile that we wanted. And frankly, hopefully, we'll get the titration and the dose right in the next few months. However, the next in line, really, to me, is the triagonist, which is entering clinical development right now and the approach that platform that we have developed, once a week, once every two weeks or even once a month possibility. So there's a whole program here that obviously will depend on results that we see in the clinic, but our idea here is to go into multi-pharmacology and really push the expansion of the duration of action from once a day to once a week and longer.

Thank you, Elias. Just to come back to Hemlibra. I understand - I think I understand your question. So we've done extensive work first, right, to analyze the hemophilia market and its trend before moving on Bioverativ. And what we are seeing insofar is not a deviation from what we did in our analytical work and assessment at the time. So our position is factor replacement therapy will remain the standard of care for patients with hemophilia A because it's a very well understood biology. It has an excellent safety. We are working, as you know, on superior extended half-life profile with both Eloctate and Alprolix. And there is something which is emerging very strongly. It's a benefit of those two because of the very specific Fc fusion technology on joined half. And finally, patient can dose based on their specific needs, which is not medically something Hemlibra will offer as a possibility. But before I go too much beyond, I'll ask John Cox to give you what his assessment and what he's saying at this point. John?

Thanks, Olivier. I would first just for [indiscernible] when you say it's on the market - or Hemlibra is on the market, it's on the market for a small portion of the market or a niche portion, the inhibitor market. So far, what's important for our product with Eloctate particularly is in the market and the non-inhibitor market, although we do use and doctors are going to use our product for inhibitor treatment or for immune tolerance induction. The question right now with Hemlibra that's being raised is more around the safety profile. And Olivier alluded to the safety profile we have with Eloctate, which is very promising and very good improvement. And there's a growing safety profile with it with a new modality. We'll see how that - what that means for whether the product - Hemlibra is approved for prophylaxis. And if it is, the question will be with a lot of physicians really more on the safety focus and relative to the benefit. But we see with the Eloctate, we continue to see with the growth with Eloctate is not only very low bleed rates, but the ability of doctors to treat the patient and adjust the treatment according to their lifestyle and also to see what Olivier pointed to in some of our publications, the terrific joint health and joint health improvement that we see when people are using this type of factor VIII replacement. So we do very strongly believe in the future of these long-acting products. And we're looking forward to advancing our clinical trial in the next generation, the next half-life product, which we're calling BIVV 001.

Thank you, John. And Philippe, add up because very clearly John cannot answer that question, but valuation of Bioverativ, we did give market share to early - model related to non-inhibitor two. So it's not like we have ignored that possibility. Not at all. But we feel very confident that we are getting a very strong return even with - getting some market share in that population over time. So with that, I want to thank all of you and wish you a very good end of day. And thank you very much.

Ladies and gentlemen, the conference is now over. Thank you for using Chorus Call, and thank you for participating in the conference. You may now disconnect your lines. Goodbye.