Severin Schwan - CEO Daniel O'Day - CEO, Roche Pharmaceuticals Roland Diggelmann - CEO, Roche Diagnostics Alan Hippe - CFO

Sachin Jain - Bank of America Merrill Lynch Andrew Baum - Citi Jeffrey Holford - Jefferies Richard Vosser - JP Morgan

Good afternoon, ladies and gentlemen. Welcome to our year-end briefing. Let me get right into the numbers. You have seen sales up by 4% in constant currencies, core EPS up by 5% and the Board proposes to increase the dividend to aid francs 20. So, fully in-line with our guidance. Turning to the divisional results, you see Pharma with a solid 3% increase. Diagnostics doing really well with the 7% sales growth, clearly outgrowing the markets. So that brings us to the 4% on the group level. What you see here the first time is that we have a positive impact from the currencies - 5% in Swiss francs and this of course is very different to previous year. You'll remember when we had the contrary effect. This time we benefited from the strong dollar in particular and also the Japanese Yen. Low to mid single-digit growth over the years and we expect that to continue. Contribution from all regions, I should point out that we have seen good momentum in particular in the international region driven by China, which is picking up again. And first and foremost of course 2016 was a critical year for us in terms of launches and readouts. We'll come back to that in a moment. Let me just highlight here three items. First on the launches, Tecentriq, our first cancer immunotherapy and I should say that it had a better uptake than we originally expected. So doing very well and I'm sure Dan will elaborate on that in much more detail and as you know, there is a lot more coming in the space. Now on the readouts, certainly the most significant one was Ocrevus. We saw the positive data both on RMS and PPMS on both indications for multiple sclerosis. So based on that, we are very confident to be able to launch in the United States in April and we have filed it both in the U.S. and in Europe. And finally I'd like to point out the launch of Cobas e 801. This is a really important launch, a new platform in the area of immunodiagnostics. Immunodiagnostics in diagnostics is the biggest segment we are in and that puts testing in this field on a completely new level. We have launched this now in Europe - extremely good customer feedback, very good development of this platform and we plan to launch it in the United States this year. Now it's not only about the quantity and the breath of the pipeline, it's also very much about the quality of the pipeline and it's good to see that it is also recognized externally by the FDA. We got another five breakthrough therapy designations in 2016 alone. In total we have now 14 such designations, which I believe speaks to the quality of our portfolio and the medical benefit we potentially bring to patients. In more practical terms, we clearly see the benefits in terms of an accelerated pathway towards a [indiscernible] medicines who are granted this status. Margins are stable at 36%. You have seen that core EPS up by 5% as we discussed and you see here the exchange rate effect up by 8%, actually in Swiss francs. And you see that's also reflected in how the payout ratio develops over the last couple of years. We had actually the same absolute increase in '15 and in '16, but last year, the payout ratio went up pretty dramatically in one year to about 60% due to the negative exchange rate development we had for the Swiss franc, relative to the other currencies and now it has reversed and we are back to our payout ratio in the mid-50s. Let's turn to 2017. This is the slide I really like most. It has been an extraordinary year for us in terms of product launch. You see here four launches within only one year. This is unprecedented in the history of Roche. We've never seen combination of launches in such a short period of time, but we had also important clinical readouts and we are now looking forward to launch Ocrevus in MS and we did see very good date in Hemophilia A. It will be the first study reading out in inhibitor patients in December. We'll communicate more detailed results mid of this year, but I can tell you that we are all very much looking forward to sharing these results with you. So we should be ready to launch and we see some up by the end of the year. Now let me shift to the guidance. There's a lot of moving parts here and of course we all focus a lot around biosimilars and AFFINITY. So I just want to emphasize that the guidance does account for the erosion we expect for biosimilars and it's irrespective of the outcome of AFFINITY. What I should also say and I see less discussion around it and I think this is not appreciated enough so far, is the effect of Tamiflu. We've had 800 million of sales in Tamiflu 2016 and you know that the patents have expired also in the U.S. in December and you know the pattern with small molecules, what that means going forward. So this guidance, very importantly accounts for the erosion, which we will have for sure in Tamiflu, which again was 800 million in 2016. What I should also mention on the EPS side is that we had as you know, a significant income from the change in the pension scheme, the PSI effect and again, our guidance accounts for that specific item which helped us of course in 2016. Right, '16 was a very good year in terms of portfolio progress. Now, '17 remains very exciting as you know. There's of course AFFINITY coming up. I'm sure you're all aware of that. What I'd like to emphasize is that while it's not only about AFFINITY and AFFINITY is not the only important readout. It is an important readout in 2017. Don't get me wrong. I understand that people focused on AFFINITY now because it's the next upcoming important readout, but we have very, very important readouts to come for the remainder of the year. You know about Tecentriq. We have started ongoing in combination with chemotherapy for first line lung cancer with and without Avastin. Competitors of course are running a number of studies in that field, so we'll see how this plays out. Lampalizumab, huge opportunity if it reads out positively which should come in later this year and then of course we have the remaining studies in Hemophilia for Emicizumab in the non-inhibitor segment. So a lot to come in 2017 in terms of clinical results. With this, let me conclude. We expect sales to grow low to mid-single digit. EPS broadly in-line with sales and on that basis we plan to further increase the dividend in Swiss francs. Thank you very much and with this, Dan, over to you. Daniel O'Day: Good afternoon from my side as well. Thank you for coming today and your interest in Roche. I'd love to cover the pharmaceutical division. Overall, I would say just to pick up on Severin's comments. A very solid year for the portfolio, for the performance of the business. Significant readouts coming up in 2017 and we're really focused on getting these launches off to a very strong start. They're all coming at the right time, I think, to also offset some of the biosimilars that we have and I'll be focusing on a bit about a lot of those dynamics in my presentation. First of all, just going through the 2016 results, we have 3% on a constant currency exchange basis growth in the pharmaceutical division. You can see across the board a solid growth in each one of the regions. We did have an improvement in growth as well in the fourth quarter versus the third quarter as we talked about in the third quarter call - 3% in the United States, 4% in Europe, despite some pricing headwinds there; the volume growth was around 8% in Europe. Japan actually overcame the significant price decline in Avastin of 10% that happened earlier this year, overcame that with true good volume growth with Avastin, but also with other products in our [indiscernible] organization. And the international, I think back to mid-single digit growth at 4%. All the international regions contributing. China back to a growth of 6% and as I'll speak about the Herceptin, MabThera and Avastin in China are all growing more than 20% as provinces in China expand their coverage of those medicines. On the P&L line, again, I think the bottom line is that we're able to slightly improve the margin again on the pharma business at a time where we're significantly investing and it's probably worth a little bit of explanation to P&L and some of the dynamics that are going on here. Sales growth of 3%. The royalties, we had a decline year-on-year versus '15 in terms of some outlicensing royalties in particular. So that was a headwind if you like for us. On the cost of sales line, we basically have two dynamics going on there. We have as I've spoken about in the past, our increasing investment in our biologic manufacturing capacity, the closure of some of our molecule manufacturing facilities that are adding to the cost of goods lines there, but its offset because of our royalties that we pay in the cost of sales line. So essentially you get an offset between those two factors in the cost of sales line, but we are continuing to invest in our manufacturing capacity which is essential to the growth of our medicines in the future - most of which are biologics. On the M&D side, we've done a very good job at de-investing in products that are quite mature in our portfolio, even though they're still driving a lot of our growth today including that investment into our launch medicines to make sure we're very effective in getting our launches right in the beginning of launches. We know in our industry that you have to get out the door fast and strong with launches and I can assure you, we're putting significant investment behind those, but still able to keep the M&D in line with sales growth because of that offsetting effect of de-investing in other areas of our M&D business. R&D, we have significant opportunities to continue to invest in. Certainly cancer immunotherapy is becoming a larger and larger portion of our R&D line, very important to our mid to longer term future. And G&A of course, we had the benefit from the pension in the adjustment this year and essentially, between the royalties and the pension income, to a large they offset each other which allows us to continue to grow the corp in profit at 4% and 5% in Swiss francs. A good strong P&L that's quite a bit of work behind getting there. I'm going to go into a variety of these products throughout the course of my presentation, so I'll just touch on the ones that I'm not going to dig deeper into. But you see, the good growth on the oncology franchise overall, 4%. I would just say that Activase is now more than a billion Swiss francs in growing double digit. You can see there, it's 16%, a significant medicine in our portfolio that continues to deliver with line extensions as well. It is important to note, I think on this slide is a good way to look at it, that much of the new growth is coming from our newly launched products that are just beginning, I would say, to take off and to penetrate. Alone, just the growth year-on-year Tecentriq and Alecensa contributed 250 million Swiss francs to the growth this year and interestingly, you can see Avastin relatively flat - and we'll get back to that - but it's a tale of two stories between the emerging markets in the U.S. But what you can see here is that on an absolute basis, in fact Tecentriq, basically compensates for the Avastin loss in the U.S. and of course, part of that is in lung cancer and part of that is a part of our strategy. In other words, to reinvent the next generation of care in lung cancer and in many of the Avastin indications we think will come from our cancer immunotherapy program. You begin to see that playing out already in 2016. Lucentis, I think we're at a situation now where we're able to see stabilization to an extent as we go into next year. We'll be launching very shortly the pre-filled syringe and the mCNV indication with Lucentis which are two new opportunities for growth with Lucentis and of course we have longer term programs with Lucentis around delivery options as well as a bispecific antibody that will read out this year in the ophthalmology space as well as a potential follow-on product. Tarceva basically boasts with in-marketing competition and in at least the wild type segment, some additional erosion with the success of cancer immunotherapy in lung cancer and Pegasys as you know, really at the end of its life cycle in HCV. So a solid performance in terms of the new medicines, far out pacing the medicines that are older or going out patent. Severin already mentioned it, but next year, we'll need to offset the Tamiflu business - the small molecule patent is expired in the United States. Competition as you know in the U.S. for small molecules is very efficient and I can assure you that that is very efficient already in this flu season. We had a little bit another month or two of the pediatrics suspension this year, but a very small amount because the vast majority of the business is the Tamiflu tablets, or capsules I should say. We would expect of that 800 million, 500 million is in the U.S. in terms of seasonal demand and then another 100 respectively in the international, Europe and Japan, which we'll also see generic erosion throughout the course of this year. Again, our model needs to display stats and we've been successful at that with our pipeline and you certainly can see that in our guidance for next year as well. The oncology business growing at 4%, the new products really up to a good start. HER2 franchise at 8%. I'll get more into that in the next couple of slides, but you can see the Perjeta and Kadcyla continued growth and Herceptin actually also had growth with longer duration and take up in the international regions. Avastin, 18% growth in the emerging markets and 5% decline in the U.S. CD20 franchise growing at 3%. Of that, Gazyva is growing at 52%. So Gazyva continues to further penetrate. Of course in the CLL segment, we'll continue to have competition in that on a standalone basis, but it will find its place there. But we're seeing now the role in relapse refractory, indolent non-Hodgkin's lymphoma starting to take up in 2016, more in 2017 and of course we're working now to get the approval in the first line setting for non-Hodgkin's lymphoma. We expect to have that filed in the U.S. and Europe this year. Gazyva, just beginning to get its legs underneath it as we head into 2017. Tarceva with continue competition. Cotellic and Zelboraf, interesting point to note here is it's doing well in the segment that it competes in melanoma and in fact there's quite a stabilization in terms of the market share in melanoma where we see immunotherapies playing the role versus targeted therapies. So we don't see continued erosion between those two and we see continued growth there as well. I would just point out that we have now our first patient in our triple combination BRAF, a wild type program, so that's Cotellic plus Zelboraf plus Tecentriq. The first patient is in on that and we expect to start the mutant trial there that will have Tecentriq in that as well in the melanoma population shortly. Tecentriq, really a good start in bladder cancer. About 60% penetration in the market and its labeled indications in the United States really performing well, has become the standard of care certainly in second line and the portion of first line that it plays in. Since the beginning of the year, we received the PDUFA date and the prior to review for the first line bladder cancer setting. That PDUFA date is now April 30 and of course that opens up the entirety first line cyst platinum in eligible population for bladder cancer for us after that approval. And the phase III trial, the confirmatory phase III trial should also be reading out this year and presented this year to confirm Tecentriq's role in bladder cancer. Lung cancer as you know, we just started to get out the door in October. All-comers label remind you that we showed a very significant survival effect across every subset including smokers, not-smokers, squamous, non-squamous, PDL status and that's off to a good start. It's too early yet to give you market shares in that area because the data is just too fresh, but we're definitely off to a good start with a comprehensive study program that's allowing us to be competitive in that second line setting. Alecensa has really start the stage in the second line setting. Market share continues to grow well in the United States. In the first line setting in Japan, we have up to 60% market share in crizotinib-naïve patients and the breakthrough therapy designation was granted for the first line setting based upon the Japan first line study, so-called J-Alex and we'll have the confirmatory trial if you like, the global study, the Alex data available on the first half of this year and really look forward to bringing that medicine to patients in the first line setting with as you know, brain meths in a very effective way as soon as we can roll that out further in the U.S. and Europe. But it's really best in class product, obviously, without positive mutation in lung cancer and contributing significantly on the financial perspective. Now I shift just to the quarterly growth as I always do for you and you can see continued strong growth in Perjeta. Perceptin, relatively penetrated in the developed markets continuing to grow in the emerging markets and Kadcyla basically in its labeled indications has good penetration now in the U.S. certainly. Europe getting very-well penetrated, still growth in the international and Japan. AFFINITY, we're all looking forward to those results. We expect, as I mentioned at JPMorgan, we've got all the events in now, the data is being cleaned, that's a very comprehensive study. We will have a readout obviously between today because I'm not announcing the AFFINITY results today and the end of the first quarter, I can assure you that as soon as we have the results shortly after, you will have the results. So I don't have any additional information on AFFINITY. We have to see the readouts. As I've said before, the signals are good, relative to what we've seen in neoadjuvant and elsewhere, but we'll see how the AFFINITY trial reads out. Perceptin, we have good sub queue conversion in the markets that were launched. In fact, the MabThera sub queue conversion is near in the Herceptin sub queue or conversion now in the European countries where it's launched. But we have countries that are well over 50%. On average, we have around half of the business in the markets where we're launched with sub queue formulation and Perjeta as I said further increasing. Avastin, I'd just remind you that the lung cancer setting for Avastin is around 15% of the total Avastin business globally. That's the area where we see pressure in the United States. In the second line setting where it is not labeled and beginning in the fourth quarter of last year in the first line setting based upon diagnostically positive patients in the first line setting with immunotherapy treatment. So we do see about 5% decline for the year in the U.S. business. That was offset globally by the growth in international and you can see some of that growth here as well continue to uptake in ovarian and cervical as we go into 2017, but we'll continue to see pressure in the lung cancer portion of the setting in the United States as well. Overall, we will participate in some of that cannibalization of Avastin in the lung cancer setting with Tecentriq as well as I mentioned. The immunology franchise, we often forget that not more than about a quarter of our business in Roche Pharma is driven by the immunology, the ophthalmology and the infectious disease franchise. You can see the immunology business alone is above $7 billion, continuing to grow in an annualized basis, a double-digit 10%. Xolair with new indications continuing to do well despite some new entrance in general in that allergic asthma space and the pediatrics launch is going well and also providing the halo effect to the adult usage of Xolair as well. In MabThera/Rituxan on the full year basis continues to grow in the vasculitis and rheumatoid arthritis areas as well. Actemra, another good strong year growth with Actemra, really good leadership position in in the monotherapy segment. Sub queue represents around 40% of the business now and growing, so you can see both in the developed and international markets. Good growth. We continue to see opportunities in the first line monotherapy setting for Actemra as we go into 2017. And a really interesting new line extension for Actemra will be the giant cell arthritis which is the second breakthrough therapy designation for Actemra and a priority review status, and we expect to be launching that in the United States and the EU this year. As you know, steroids are the only options for patients with this condition and Actemra showed really positive results on top of the benefit of steroids. This is a significant line extension for Actemra as we go into '17 and beyond. Esbriet above 30% growth this year, continued market leadership with around 60% market leadership of the class doing well in the moderate patients. The real opportunity with Esbriet moving forward will be getting into the earlier stage patients. Really, looking to get into the mild patients and of course this disease continues to progress in patients in the earlier retreat and the greater benefit for patients and we have started programs that are directed directly at patient education, working together with the idiopathic pulmonary fibrosis patient associations as well as physician groups and we're off to a good start. Testing it out in Europe and in the United States to attempt to increase that portion of patients that can be treated with Esbriet sooner. That will be our clear focus as we head into 2017, in addition of course, maintaining or improving our share. Overall, a good solid performance on the results side. For the news flows side, it was also a very good year overall. I would just say since we last saw each other, we have the Alecensa EUCHMP positive opinion in the second line up positive setting and we also have results, the Phase II readouts both from both Venclexta and Rituxan, this so-called Phase II relapse refractory [indiscernible] study and the renal cancer results for Tecentriq and Avastin, the emotion 150, which we will see at ASCO GU coming up here in February. I look forward to sharing those results with you. In addition to that, showing the benefit in renal cell carcinoma, we think it's an important signal relative to the potential synergy of Avastin with Tecentriq for other programs we have like our lung cancer programs and others where we are testing this combination. So it's a first time to really see substantial data on that combination. Terrific news on emicizumab at the end of last year with the HAVEN 1 study reading out all positive primary and secondary end points will be presenting this to you at the Hemophilia meeting in July of this year. I think it's called ISTH. But very solid performance in terms of reduction in bleeds, both on patients that are naïve to prophylaxis, but very importantly also on the secondary end point of the intra patient comparison where you have patients that received prophylaxis and the bypassing agents and then is receiving prophylaxis with emicizumab and seeing also a significant reduction in bleeds in those areas. The safety profile continues to demonstrate what we saw in Phase I tolerability with the most common side effects being the injection site reactions. It's very consistent with what we're seeing before and very importantly, the entirety of the program is going well, so the recruitment into the non-inhibitor trial which we expect to read out at the end of this year as well as the pediatric trial and the once-a-month dosing is progressing very well. We're excited to show you this data, we're excited to talk with regulatory authorities about bringing this option to patients with Hemophilia A. Gazyva, some of you are coming off of ash of course and I'd just remind you, if some of the data here in both the relapse refractory as well as the front line setting in the PFS has a ratio, if you look at - of course this is an interim readout - but if one puts forward to what this would translate in a longer term readout because I'm often asked the question, 'How significant is Gallium in the front line setting?' We're talking about a 1.5x longer medium PFS potential with this medicine. That's the difference of nine years versus six years when you play that out over the long term, I would say that's pretty significant and it's something that we will continue to read that study out. But clearly, I think we have demonstrated a change in the standard of care in the indolent setting with Gazyva and you know that we've got continued programs in the aggressive forms of non-Hodgkin's lymphoma with a very broad hematology program that I'm not presenting to you today including medicines like Venclexta like our bispecific antibody and like our immunotherapy programs. The data in bladder cancer for Tecentriq, the Cohort 1 data is now what's being used for the PDUFA date of April 30 for the first line setting. This is just to remind you that it expands. If you see here whether IMvigor210 to 10 is Cohort 1 on the upper side here, we've already demonstrated Cohort 2. It brings us into the full first line setting and then we also have some additional trials that we'll be reading out in '17 and beyond including the confirmatory Phase III trial, the IMvigor211, which we may have in time for ASCO, but will be this year for sure in terms of the presentation of the Phase 3 and of course we continue to work with IMvigor130 and IMvigor10 as well to expand us into the earlier stage settings of the disease. All right, cancer immunotherapy program continues to evolve every quarter. I'd just remind you, we have now 10 new molecular entities beyond Tecentriq that are in clinical trials in humans and you can see we have a comprehensive approach around the cancer immunotherapy cycle from those that are in the desert area to those that are excluded, to those that are inflamed and we have more than 50 trials now going on across the Roche group in cancer immunotherapy, 40 of those are in combination. As you see here, reading out next year is seven of these NMEs across 10 different trials. So some of these NMEs are in more than one trial. You can see for yourself, the ones we expect to read out in 2017, end of 2017 and some into 2018. We certainly look forward to providing you the data on these as we can and as the appropriate meetings come up. We're hoping to make at least some of these for ASCO, others potentially later in the year and we will certainly keep you up to date because I think this is not only informative for our own programs - which obviously we're focused on - but this will move very much the science in cancer immunotherapy over the course of this year between what we're doing, between what other companies are doing. I think we're going to really understand much better the key leaders beyond Tecentriq, beyond PDL1 in terms of what are the combinations that are most likely to be the right mechanisms to go after to get to the vast majority of patients that are not responding to monotherapy alone. Our lung cancer program is incredibly comprehensive. I know there has been a lot of data on the lung cancer programs in the recent days and weeks. I would say that to just remind you, we have a very comprehensive program across a very comprehensive set of chemotherapy back bones. The first of those, the IMpower150, all of these are powered for both PFS and OS. We've taken into account that PFS may read out first. It could be the PFS and OS readout at the same time, but we've taken that into account in terms of how we power these to make sure that the PFS does read out first, that we can still get that OS end point which we believe in cancer immunotherapy and certainly in lung cancer will discriminate and differentiate medicines out there. So we can expect the first readouts till this year with the IMpower150 with and without Avastin and the chemo backbone, what you see there. Many others in 2018 as well to complement that both in non-squamous and also one in squamous. I'm sure we'll be talking about this more in the Q&A, but we have a very comprehensive and complete program in lung cancer. I just wanted to point this out because I think we're constantly evolving as a cancer immunotherapy community, I would say and certainly within Roche and we're adapting as we see how regulators are adapting to acceptance of different types and levels of data. So on the top here you'll see what we have been using and we'll continue to use, which is looking for early signals in Phase IAs. Moving those quickly to Phase IB is the most promising and then shifting from there into label-enabling trials. We'll continue to do that because as we've seen from some of the recent past of readouts of not only our programs but also competitor programs, there is no one-size fits all with this. There is no way to assure that with a particular medicine, that a particular regulatory strategy or particular trial strategy is most effective. So this is our new program called Morpheus Novel CIT platform which allows us to - if we have a very promising signal to get it into a multi-pronged kind of Phase II, potentially label enabling trial that will take advantage of common backbones, common control arms, it's an umbrella study, it will have all the diagnostics involved with it, it will allow patients to go into these arms; the first ones we're going to be starting with in 2017, will be lung cancer, as you see here pancreatic, gastric and breast, more to come it's going to be an adaptive program but I think it's one technique that we're looking at to allow us to respond even more quickly to early singles to get them into potentially label enabling trials and to potentially get approvals even sooner or determine exactly how the medicines will move. So this is one of our innovations but we have others of course in the cancer immunotherapy area that we are looking to make sure that we move as fast as we can but also come up with good comprehensive programs that are going to test - stand the test of time as well. This is the entirety of our late-stage oncology portfolio, we have now 20 Phase III, these includes both our cancer immunotherapy and our targeted medicines, you can see - we have the cancers well covered right now in our Phase III, the ones that were highlighted here are some of the most recent additions. I would just say that we now have a very comprehensive combination program in Phase III with Tecentriq, it's happened over the course of last year but they are generally in three buckets of combinations; one are in chemos in different diseased states, one is with Avastin, and one is with Cotellic so far, and more combinations to come. I mean this will clearly evolve, there will be more Phase III that we'll be starting in the near term future and we'll be updating you as the year goes on. In terms of the outlook, we have a lot of important news read out in 2017, yes AFFINITY but not only AFFINITY and I think that's what we're emphasizing here as we have lots of levers to poll as we go into 2017. So of course we have AFFINITY, we have - we have judges [ph] call some of the larger ones out, we Tecentriq in first-line lung cancer, we have also Lampalizumab, and at the end of the year we have Emicizumab and the non-inhibitor population, so - and others that you see here and some that aren't on this list that are well worth pointing out. And of course that innovation doesn't stop in 2017, it continues into beyond 2017 with a lot of the readouts in immuno-oncology program reading out also post 2017 as well. So to conclude, another busy year on the pipeline; I think we've invested smart, we've invested well on this. We're going to get regulatory readouts as you see here, Phase III readouts, you know, important for us is to look at this upswing in our launches right. So we've launched 11 medicines over the past seven years but that pace has accelerated, such that in the past a little over a year we have four new medicines that have entered the market and we have two more significant medicines coming in 2017, potentially we've got Ocrevus and we have Emicizumab in addition to really important line extensions. So our focus will be on getting those launches right, making sure we invest in those appropriately, progressing the pipeline, making sure that our precise healthcare approach, our data-enabled precise healthcare approach continues to evolve, we think that's a significant competitive advantage when it comes to making better R&D decisions and running R&D better and better patient care and obviously a continued focus on productivity and cash as well. So with that I would turn it over to Roland to cover diagnostics.

Thank you, Dan. Thank you. A very warm welcome from my side as well. I like to share the diagnostics results and as pointed out it was a very good year in terms of sales, it was a good year also in terms of the pipeline progression. It was a challenging year in terms of the bottom line and we'll come to that. Here is the sales; overall 7% for the division, now $11.5 billion in size where you also can see continued pressure on the diabetes business, minus 4%, largely in the United States and I'll come to that as well. On the other hand, when you take together the clinical diagnostics franchise, so centralized molecular and tissue diagnostics combined 9% increase, far ahead of the market, very good sales, I mean actually adding $700 million in additional sales in 2016. And when we then look at the geographic distribution, very pleased to report that we can grow in all the regions, very good global representation of our sales; the emerging markets of course growing at a faster pace, APAC was 16%, here China stands out, we've now had more than 10 years of above 20% growth in China, 22% last year in China. Latin America was 18% and also other good sales in Middle Eastern and other regions. And then if we net out the diabetes sales from the numbers that you see here, actually EMEA is growing at 4% above over the market, and then North America is growing at 8%, also ahead of the market in clinical diagnostics. Some of the drivers in terms of sales and I'll talk more about centralize and point of care in a minute. I've addressed diabetes care, when we look at the geographic distribution here, the U.S. was a very slow performance, minus 27% and I just like to remind you that this is of course due to the reimbursement cuts that were introduced long ago as three years but still have an impact on the new contracts and on translating into the private market. Molecular Diagnostics doing very well in both areas of blood screening and virology, so the large panels; HPV in virology continues to grow above the market, it was 8%. And then on the tissue diagnostics franchise, we see good growth in our core business which is advanced staining and then with the introduction of a new plant forming in primary standing we see very good growth there as well and along the lines with companion diagnostics as well. So here is the P&L and you can see the impact; 1% One percent at the bottom line impacted largely by new product launches in diabetes care on the gross profit margin, and let me walk you through the P&L in a bit more detail. The decrease in royalty income is something that we have been projecting overtime, these are largely our PCR licensees, so you'll continue to see you a deceleration on the license income. And then on the cost of sales line, it is the gross margin pressure from diabetes care. It is on the other hand also the launch activities in important new platforms; I mean immunology, in tissue diagnostics, also in molecular diagnostics. What we also see is some impact from some suppliers related to foreign exchange but then also an investment in our manufacturing capabilities, we're building a new plant, the first one for us in China, and this is slowly coming online. R&D; which is what we have flagged to you, an ongoing investment and a future innovation in our core business, and in addition to that the sequencing investments that you can see here which are mostly responsible for this increase. Let me switch gears and talk a little bit about the entire portfolio and the strategy, what do we actually pursue in terms of providing really the connected core lab to our customers. First of all, helping our customers to become as efficient as they can, as productive as they can, with the highest test accuracy of course and combining these in an ecosystem for our customers. So it starts at the bar at the top right whereas high throughput analyzers was having very performant instruments, it then goes on to connecting these instruments in the laboratory, making sure that as many tests can be performed in one integrated solution with the highest degree of automation, and what you need for that is of course also intelligent workflow solutions and IT systems that allow you to share the information and the results across. And in the future we'll also see more investments in the data management, all the data that we generate in terms of a diagnostics organization that could inform us into more and better R&D. Here is some of the examples; it's been on the launch list initially, already and Severin mentioned at the E8 or one being one really important element of our innovation. It is supporting our work area franchise which is the largest and most important in diagnostics, and you can hear see the growth across all geographies; and as we continue to introduce this platform, as we also continue to add to the menu and expand the menu. Our expectations were surpassed, it was the first six months after launch so we have about 190 instruments in the market, that number continues to grow. And another exciting platform which we're introducing for the first time, a Roche-owned instrument is in hematology where you see here follows our same design and engineering philosophy which is to automate and integrate diagnostic modalities. This is very different from what you have on the hematology market so far as it integrates three different steps; provides you with a lower sample volume, immediately a digital morphology picture of the abnormal cells so you don't have to go through several cycles in diagnosis, it gives you that information right away which is really important in the case of specialized institutions such as cancer centers who deal of course with blood diseases and who deal with high number of abnormal cells to analyze and diagnose. We introduced our high throughput molecular systems back in 2015, this is basically to report back to you we're doing very well in terms of placements in terms of installing these systems, and of course these are complex systems, these are being installed over weeks and months until the whole system is being validated. And we continue to add menu here making sure that we continue to be able to provide really complete panels to our customers that holds true for both, blood screening as well as the clinical space virology and women's health [ph]. One addition to the portfolio, in the largest area in diabetes care in traditional blood glucose monitoring and also to protect and safeguard this franchise is the Accu-Chek Guide System which is based on the universal technology, it also allows a cloud based solution which means that patients can actually monitor their own data in a better way that they can really manage the diabetes status. We have introduced that late last year in Europe, in the countries bearing the C mark and we'll introduce in the U.S. this year. So this is an overview, and reporting back on the key launches; it's been a very busy year, it's been a very good year in terms of the portfolio evolution. The one Red Cross that you see here was the relationship that we terminated actively. We have another very promising program in the sequencing which is in-house with the acquisition of Genia [ph] platform that waited a couple of years ago where we have been seeing some good progress, and for the rest two other programs that we deliberately delayed to introduce actually more features - for actually even a better diagnosis and a broader scope. So I won't go into too many details but wanted to give you a little bit of an outlook here also in terms of platforms and solutions to come for 2017. I mentioned that hematology entry was our own solution which is ongoing. Then the E81 [ph] in immunodiagnostics which is of course key franchise, the introduction of the launch in the U.S. was the FDA approval expected towards the third quarter of this year. And then in the fourth quarter we'll also introduce our own solution, our own Roche developed solution in coagulation diagnostics which again will play right into the connected core lab with the ability of course then to integrate to connect for our customers. And I'd like to finish on this overview here, I've mentioned the instruments, there is a host of assays and reagents that we're going to bring to the market. Let me just point out a few on the molecular point-of-care on the lead system, we have now a comprehensive respiratory panel, we're moving into the hospital acquired infections. You can see this here, we see different MR assay. And then what's also very exciting here is of course the entire area of liquid biopsy cell-free DNA and we'll introduce this year three different panel, some to inform on oncology therapy and one was a larger panel for surveillance purposes. And then finally, continuing to contribute to personalized healthcare with our complimentary testing for PDL1 supporting Tecentriq, these tests are available in the U.S. and will be made available in Europe in the course of this year. And with that I'd like to hand over to Alan Hippe. Thank you very much.

Thanks, Roland. Hello to everybody, it's great to see you. Good news it's the last section. Bad news, perhaps there is still 15 slides to go. I'm going to try to make it comprehensive on one hand and brief on the other hand. When you look at the numbers and I can say really - it's really characterized by the investment we have done in 2016 to prepare ourselves for 2017 and also challenges ahead and you will see it really - it's obvious and the P&L is obvious and a balance sheet also obvious in the front statement. So when you look at the highlights, I think all mentioned by Severin when it comes to business and with sales 4% up, core EPS growth 5% up, and 2% up when you exclude the past service income effect. You know we have changed the pension plans in Switzerland which gave us a positive impact on the P&L, you will see that the liabilities on the other end came nicely down, and it's really obvious in the balance sheet, so really a good step in the right direction. Cooperating profit up 4%, dividends in Swiss francs further increased to 8.20, at least that's what the board proposes to the shareholder assembly and we're all very hopeful and that's a proposal we've got through. Cash flow, the cash generation and I have a slide on that. 800 million down but very consciously, very consciously because we have increased investment into intangible assets, as well as into PP&E significantly, roughly 1 billion and I will show you that later on. So very conscious development, when you adjust for that you see a really cash flow generation at a very high level as you are used to. And then the accounts receivables story in Southern Europe further decreased, I think it's a need story and I will lead you through this in a second. On the net financial results, and I can't cut short because we have 22.3 billion gross debt and what we did is '15 to '16 is we brought the interest rate from 3.7% to 2.9% and by a major debt restructuring as you know. And in fact that gave us a saving in 2016 of 180 million Swiss francs which is mentioned here on the slide and that's certainly quite significant and helped us to boost the EPS growth. Here you see the comprehensive set of numbers, don't want to go to strains [ph] cooperating profit, core net income as you can see up 7% and core EPS up 5%, and in Swiss francs even 8%, and that's a point I can come to later on when I give you read the basis for 2017 for your projection, so really for the year and core EPS for 2016. And you also see that the IFRS net income developed quite nicely with 7% up. I think when you really look at the non-core topics on one hand a little bit more than in 2015 on the other hand, certainly the debt restructuring not included. So that gave us a little bit of an ease and that helped really to bring the IFRS net income to a 7% increase. Operating free cash flow, we'll talk about that and then you see there is even a little bit of an acceleration to free cash flow into decline if you like but this is a contribution that we have made to the pension funds, roughly it's 300 million Swiss francs went to the pension funds and that certainly was one, how should I say - one element, we had to make our pension funds more sustainable for the future. When you look at the P&L, I think my colleagues did a great job already explaining that so I cut myself short here a little bit. And you see really royalties other operating income down 200 million, Dan elaborated about that. I think I would like to put that into perspective with the PSI effect. PSI effect, positively 426 million, we lost 200 million on the royalties, so puts that a little bit into perspective when you look really at the operational performance. Cost of sales, an increase of roughly 800 million, 5%, and really investing into our manufacturing work network on one hand and on the other hand certainly the PDL patent which expired and reduced our royalty expenses quite significantly and helped us out here. [Indiscernible] focused on the new products and the launches and then R&D driven by both divisions but certainly the immunotherapies was one of the major focus points here. And then G&A, yes, well you see a positive development here, a reduction of $388 million, certainly driven by the past service income but even when you adjust by the past service income, I would say a pretty reasonable development of the G&A line. It brought us to a cooperating profit increase of roughly 900 million, representing 4% in constant currencies. Well, I think the margin is relatively stable, well slight decline on the diagnostics division, primarily driven by diabetes care. When you look really at productivity, and I think I would like to encourage everybody to take a look into our finance report and the restructuring charges over there and quite significant, basically in every year. And when you really look at productivity and what it means to us, certainly what we would like to do especially in a period that we are facing now, preparing ourselves for 2017 that we really invest where we really make a difference for patients but also where we really make a difference for our returns. And do you see really on the cost of sales line and that applies really to all activities here in manufacturing. And when you look really at the restructuring charges, roughly 1.2 billion, the major part of it is really our switch for small molecules, for the small molecules production capacity that we will bring down over the years and which will certainly help us out on the cost of sales line. And marketing and distribution; well, I think here are resources shift Dan has talked about it, really investing adequately for the new products and shifting resources for the more - from the more mature products to the new products. Research and development, speed and agility, and strict privatization, and then in general administration to shared service centers, you will find restructuring charges for that as well where we really move more activities in finance, IT, and as well in HR into this kind of structure. When you look at the core net financial results, quite a significant reduction and let me start really with the impact from currency, an improvement of 262 million, you know we had an impact of roughly 400 million in 2015. In 2016 we had an impact negatively of 124 million, so real difference here 260 million, you can explain that with Venezuela basically. You know, we had a major impact in 2015 from Venezuela, roughly 254 million when I remember it well, and I think that's the major effect here. But keep the 124 million currency impact in mind in the financial results because that's really important when we come later on to the core EPS and the right basis for your projections in 2017. 180 million interest expenses, I've talked about that. It's quite a reduction here that we have seen and then the equity securities with some smaller disposals that we have had; and then the loss on early bond redemptions minus 63 million, and that represents really that we had a larger impact of 142 million in the year 2016 compared to 79 million in the year 2015. When you look at a tax rate, remember vividly our discussions that we have had last year and pretty stable. What you can see here - when you looked really at the current income taxes, you see quite a reduction of current income taxes of roughly 400 million. And that means you take a look at the deferred taxes and what you're going to see is we had a positive impact in 2015 of roughly a billion and that has reduced to 300 million in 2016. And where is that coming from? When you look at the balance sheet diligently, you see that the provisions came down in general and certainly these provisions also have a tax impact and depending where these provisions are taken locally, as the local tax rate is applied, and that has driven that positive impact a little bit lower but I think for me important is the current income taxes decreased in absolute terms by roughly 400 million. Even when you look at the effective tax rate, I reminded everybody last year to look at the effective tax rate if you really want to have a real picture, then you see really that it also went up slightly from 24.5% in 2015 to 25.2% in 2016. So I would say it was an effective tax rate of 25% - I feel, rather comfortable with that number. When you look at the balance sheet, has that impacted by the investments that we have done, our cash and marketable securities was roughly 9 billion, it will help us out to pay the dividend. The other current assets moved up a little bit inventories receivables; our business moved on. And then the non-current assets where we're clear on one hand we had a reduction in the intangible assets and we had an increase in PPME and that's what is reflected here. Current liabilities went down by 800 million, this is a short-term debt that we reduced. Non-current liabilities went down by 900 million and this is predominantly the pension liabilities which reduced, and the other part is provisions as mentioned before. And then we have equity which well - I think developed quite nicely with more than 3 billion increase, so 34%. Equity increased net debt to total assets at 17%. Operating free cash flow, and at first sight a reduction of 800 million and you ask yourself what has really happened; and I said very consciously, we increased investments. So higher investment into PPME and that's to increase, so we are now about 4 billion if you - when you look at CapEx and that's really majorly driven by the expansion of the manufacturing networks. And then you see higher investments into intangible assets, here I think you remember the deal we have done with [indiscernible] that alone was a roughly 300 million when I remember it well. So I think we spend roughly a billion on investments into intangible assets and the 359 million represent the increase from 2015 to 2016. So I would say still underlying cash flow generation on a very high level. Yes, just one example here, accounts receivables stat in Europe and you know that story, that's an ongoing story and I think I presented that slide really five years in a row. And you see really we went from 1.8 billion end of December 2012 to now below 800 million, 57% reduction in a timeframe where our sales reduced by about 12%. So I think on one hand we are roughly stable I would even say with sales in a pretty critical and challenging region. I think that's quite a nice achievement but we also managed our receivables into the right direction which certainly means we have mitigated risk to a major extent. Group net debt; we started with December 2015 14.1 billion net debt, operating free cash flow I've talked about, then taxes and treasury, cash out and then certainly the dividend and then currency and others minus 1.2 and that leads us to a 13.2 billion net debt; which means in a period where we really have had quite cash outflow from investments we were still able to reduce our net debt to now a level of 17%, we've always said we want to be between zero and 15% net debt on total assets and there is still a step to go. Currency; Severin elaborated about that already, Swiss franc has weakened against the U.S. dollar and Japanese yen and the euro in average into 2016. And it's strengthened against Latam, and Latam is Venezuelan and Argentina predominantly, and other Europe - important to mention here is the British pound. So that's the kind of quite - how should I say, quite a change in what we have seen the previous years. When you look really at to projections that we have had in 2016 I would say we were basically spot on and you know this is a very mechanical projections that we're doing. At every reporting point here we keep the currency rate stable and it keep everything stable and then project what's going to happen at the end of a quarter or at the end of the full year. And the same we're doing for 2017 based on the currency rates end of 2016, and if we do this we see basically a positive impact coming between zero and two percentage points yet to sales cooperating profit as well as the core EPS. Well, we've all seen that the volatility in currencies is huge at the moment; so I think I don't know whether this mechanical view really leads to a lot but at least we wanted to give you that because we're always doing but I think you have to take that with a grain of salt because I think one thing which will be with us in 2017 and that's for sure is volatility. As Severin said everything about the guidance I think that we are certainly happy to discuss with you and where we - I think is quite comprehensive and has baked in quite some risks. Let me give you the basis for 2016 and I've seen somehow should I say it; you have some question marks in some of the phases but you know last year it became apparent that we adjust here when we have the core EPS that we adjusts to find the currency impact in the financial result and that was one of the points - questions I rose last year in this meeting here. Core EPS reported as you've seen in Swiss francs is 14.53, that's the basis. And then the 2016 currency loss in the financial result as I've said 124 million after-taxes 116 million; when you adjust for that this is 0.14 Swiss francs per share, so you have to adjust your 14.53 plus 0.14, and that leads to the basis for the core EPS growth outlook to 14.67. We have a slide in the presentation, in the backups to make it clear again but I wanted to take the opportunity to really lead you to the right basis to do your projections. And with that, I think we are very happy to take your question. Thanks for your attention. A - Severin Schwan: Good. So let's get right into your questions. Sachin, you want to start off? Do we need a mike, yes, can we have the mike here in the second row please. Good.

Thanks. Sachin Jain from Bank of America, a couple of questions please. So firstly, just a kick off on the conservatism within guidance; does it seems to be betrayed. Just to be careful, AFFINITY work, should we expect a guidance upgrade or should we think about it more as the shift from the bottom end of the - going to the top end; I'm assuming you add back in the new management and assume some off-label use if AFFINITY works? And second question is just to dig in a bit more detail than around the lung cancer assumptions; you know, one of your assumptions around first-line chemo combo from Merck and its impact to Avastin, should we expect that to accelerate? And it's a big impact Tecentriq second-line; how do you think my second-line market shrinkage in that setting? And then a final question is on Ocrevus, as we head to the PDUFA, I just wanted to know if you could discuss any expectations or you should be thinking about regarding the label, and specifically what are your expectations regarding PML, labeling and are there any variables we should think about in terms of PPA mass labeling? The reason for last question is your competitor cited they wanted to see the label before starting a PPMS Phase III study. Thanks.

So let me take the first question on the guidance. I mean as usual, when we think about the guidance we look at it holistically and we look at all the opportunities which we obviously have with the new launches, clinical readouts; and on the other hand we look at the risks, we will have some erosion from biosimilars, we will have generics coming in for Tamiflu of course, and they had a downside scenarios on the pipeline side. And if you put all of that together, then we are confident to grow the business in the range we have guided here for. So I hesitate to say, well, you pick out this element and then you can put something on top of it; it is at the end of the day holistic guidance which takes account of all the various opportunities and risks which we face as we go into 2017. Daniel O'Day: Thanks and Sachin, just picking up on your question. So I mean - likewise, I think there is a bit of ambiguity around exactly how the first-line lung cancer chemo-setting will play out this coming year. I mean, I think great news for patients that we might have an option earlier for the front-line lung cancer study based upon really overall survival data or overall response rate and PFS. We'll see how that plays out with the regular three authorities but I guess presuming success for that I think in 2017; I think we see less of a potential impact on the second-line lung cancer study just because it takes some time for patients to progress in terms of Tecentriq. I think where we see the bigger impact potentially is in Avastin in the frontline setting in terms of 2017 sales because that you know, is an area where Avastin is labeled and where Avastin is used. And today of course in the diagnostically enriched population, we're talking around 30% or so of the frontline setting; and depending on the label and depending on how things work out, obviously that 30% could potentially increase in the first-line setting to areas where Avastin has kind of a stronghold today in combination chemotherapy. Of course, maybe just a comment on that because I know we'll have probably multiple questions on this - I mean, we always strive to be first and best, that's what we would like to be in these areas. We'll see how that plays out in the first-line lung. Well, I believe that cancer immunotherapy is the best will win, we're not far behind. First of all, in terms of our readouts and in our study that's fully enrolled now, the EMPOWER 150 does have both PFS and OS as nil point, we'll see how that reads out; it's event driven, we'll see how that plays out, whether that plays out at the same time PFS and OS or two different times. But there is a lot of unknown in this space and I also think that we compete well when we have very good data, I think we've seen that in the second-line lung cancer setting with OAK [ph], I mean we're beginning to see that I should say but certainly we have nothing to shy away from in terms of our OAK [ph] study and being third to enter into the second-line lung cancer setting. So I think we always have to balances kind of - you know, first to market and best in terms of the space. But specifically on the guidance because I know that's what you're geared towards. I think the real impact is on Avastin for next year and less so to Tecentriq. And then finally on Ocrevus, yes we're still working well with the FDA for the PDUFA date at the end of March. Just to put this into context, this was perhaps I think that the largest file we've ever submitted to the FDA in terms of the data points, I mean we have data from RA from - extensive data on our RMS and PPPMS to the neurology division of the FDA which is working very well with us. But this is the first breakthrough therapy designation priority review the product the division has reviewed when they set the clock. We were all ambitious to have two indications in a very short period of time at the end of December. Based upon some requests for some additional data on the routine manufacturing which we've submitted and is now being reviewed; we are all working toward PDUFA date at the end of March. We expect - you know, and it's too early obviously to comment on the labeling but at this stage everything suggests that we have the potential to receive an our mass RMS and a PPMS label; and you know in the United States which is different in Europe, the secondary - the SSPMS is also within the context of the remitting label in the United States. So that's what I can tell you so far, I can't tell you what I don't know because review is still ongoing but we'll certainly update you as we get nearer to the PDUFA date.

Alright, thank you, Dan. Perfect question over to answer.

Thank you. It's Andrew Baum from Citi. Three questions; firstly under the name administration in the U.S. proposal of tax reform. Your tax rate certainly isn't one of the lowest in the group. I understand that tax planning is a complicated issue but given your manufacturing within the U.S. for the U.S. market - to what extent you think could benefit given what's being discussed. Second, in anticipation of biosimilar entries into the U.S. I'm anticipating the focus of the response is initially to be on U.S. oncology or some of the marquee names for [indiscernible] and so on to validate the concept of biosimilars, is that consistent with your expectations or do you think they're going to be different hospitals that they will be targeting 340 B's and so on. And then finally, you've obviously to me it seems at least there's been a step change within the immune-oncology, you're now doing basket trials despite - this is something that was discussed a long time ago and Roche has been behind some of your peers in exploring this particular setting, you did the biotech deal with a pretty significant upfront in September. So something changed within the organization which is accelerated, added additional urgency within the group and is that a clinical or is that a business related issue.

Okay, thank you. Perhaps I can just give the first comment on the dynamics in cancer immunotherapies. I think it's probably both, as a scientific element to that because as we see data reading out we get more and more excited and the more excited you get driven by our internal data, driven by external data, the more confidence you build and then more aggressive you go into it and there certainly been a dynamic in this sense that we have put more resources overtime behind cancer immunotherapies and today it's really a significant investment as you have seen. We have now 10 new molecular entities in clinical development, more coming through and we have over 50 programs running. So that tells you what has happened behind the scenes over the last couple of years. So there is a scientific element here but I would say that is also a commercial business element here because it's a highly competitive field and competitors take a lot of risk. So at the moment, somebody sees a signal they go ahead with pivotal trials. So you have to adapt your strategy because if you're alone you might gain the respect risks and you might optimize the benefit risk in a different way than if everybody is taking full of risk because you might do the right thing but you are anyway too late. So therefore, you have to think more aggressively or how can you accelerate trials; how can you cannot beat your new concepts. What I should also say is interesting to observe if I look at the evolution of how we deal with the challenge is how we work together internally in the group because you know, traditionally, this is broadly the case today. Our period organization in Basel and the GOI Genentech organization in San Francisco are completely independent in the decision-making but now here in cancer immunotherapies you have the potential of combining compounds. So what we also had to do is how work - how do we work together between early stage in San Francisco early stage and in Basel and later stage organization, how do we optimize our fortress [ph]. So yes, there was a lot going on over the last couple of years and still we're only at the very beginning. And we will learn a lot actually this year as we get clinically readouts, we will learn from our own data, we will learn from competitors data; so we'll see how it play or stand. Anything to add on that before we move on to Alan - you had the tax question right, and Dan, if you could also cover the biosimilars? Daniel O'Day: Okay, we're going backwards, so I will complete with this question. I think the - some of your statements I would take some issue with and add too. I think in terms of basket trials in general, I would say that we were some of the first people into basket trials in terms of the targeted agents. And in fact, some of our Phase Ib's are in effect basket trials. I think what we're showing here with Morpheus and I'm not sure if I explained it completely well is that, this is kind of a new evolution I think of a basket trial which to a certain extent required a critical mass of kind of internal medicines to get to the stage which we're at now so this is certain evolution of the portfolio in following the science. But you know this concept of moving fast and then confirming later is not something that's new to us; I mean you saw it with the popular in oak data and second-line lung turns out because of the review process at that time that the oak did, it was confirmatory before the review process but initially as you remember, I mean we were looking at a popular approval - accelerated approval with oak confirmatory. I think we - you know, there is no one size fits all that I've seen in immunotherapy, particularly if you look at the data in the past several weeks. There are times where a so-called label enabling trial turned out to be not sufficient. We're dealing also with regulators that are changing their opinion, I don't think if any of us receiving [indiscernible] a few months ago that we would have expected necessarily that a 60-patient trial would be sufficient for accelerated approval. I'm happy about that if that's the case that accelerates things and means we can do things differently but it doesn't mean that every immunotherapy compound we have now is going to necessarily benefit from a 60-patient trial and readouts and the regulators see it the same way. So I think what's most important and that's what I was trying to show in this slide is that we have multiple options depending on the disease state, the medical need, how we think regulators will respond to it, the type of activity we're getting, and try to do this both, first and best strategy alongside each other. So you know, I haven't seen anybody else do what we're doing with Morpheus with the same type of design; I've seen other types of basket trials and we can get into this a little bit more but I do think that we need to adapt, we need to be flexible, we need to learn from the outside and inside and adjust our internal mechanisms as we go. On the biosimilar side, yes, it's a bit early - it really is early to talk about the U.S. strategy not just because we don't expect biosimilars to enter there but because we're seeing new regulations be published; so you're probably all seeing the draft guidance if you like on biosimilars which I think was generally interpreted as - you know, in terms of interchangeability as a positive for the originators. I mean it starts to outline in a very broad way, what will be required in terms of immunogenicity and other factors there. So I think there is a whole topic around what's it going to take to get a medicine approved in the U.S. and what type of labels will those medicines have. And then to your point Andrew, the question is what segments will biosimilars enter in first. So I mean hard to tell - I don't necessarily disagree with your assessment as of now that they would enter in more into the private pay setting, there are potentially as it exists today with a 340B, some benefits to originators in the 340B segment but then again, by the time biosimilars enter, we'll also see how the 340B program evolves over that period of time. But as it's structured today, I would argue probably the private setting has more inducements in the public setting for biosimilars in the U.S. if I had to make your call today.

Alan, on the tax?

Yes, and Andrew I know that for so many - I don't want to speculate here and I would like to start to answer the question about what's in there? And when it comes to tax reform in the U.S., here for us; let me first say, I think we are great citizen in the U.S. I said we have roughly 40% of our sales in the U.S., we have 40% of our operating costs in the U.S. We have invested billions there, we have significant manufacturing in the U.S. as described today where we invest in. And we're also a pretty significant tax payer, yes. So I think really when you look at it, I see - I remain cautious because I would like to see does it works out at the very end? And perhaps I can draw a parallel to what happened in Switzerland; you know, that we are discussing tax reform in Switzerland and there will be a vote out there. And do you know what the impact will be for Roche, in fact the impact will be that we paid a same or a little bit more and why is that because, well, a lot of loopholes, a lot of - how should I say it benefits to certain industries were scratched out and it all was brought down; how should I say to - to a transparent and standardized system. And I think when you think about the U.S., I think that's perhaps something which might happen there as well; so I would wonder if we - just - the tax rates will go down and nothing else happen. So I remain cautious here and I wouldn't indicate anything at this point what could be of benefit to us.

Okay. Can we move a bit to the middle here. Thank you.

Thanks very much, Jeff Holford from Jefferies. Just first of all, if AFFINITY does not work, should we expect that there will be some sort of update from management as to how they might manage earnings in the mid-term such as cost focus or potentially share repurchases? Second, you know you just touched on the importance of potentially overall survival as part of your lung cancer program given that don't have one for their chemo combo as yet. Can you just remind us about the key studies that you have in lung cancer, the level of crossover that you might know about is potentially going to happen in those studies and how we should think about that especially given the result of chemo 21G? And then just last - you know, Bristol [ph] very specifically referred to Tecentriq as the driver of some of their loss of share in second-line lung cancer; can you just talk about some of the switch factors that you're seeing there being preferential, I think they have mentioned dosing schedule but is there anything else that you're seeing about the Tecentriq profile that's helping in the United States? Thank you.

Okay, so let me start with your question regarding AFFINITY; I would say we have to work on our cost structures no matter what irrespective of whether AFFINITY reads positively or negatively we have to constantly work on our productivity. And you know, it's good practice to start with but I tell you with all these launches which are underway, we really need the money to be invested into this new opportunities and you know there is an enormous pressure on the R&D side with all these additional investments we have into cancer immunotherapy. I can tell you there is a lot of intrinsic motivation to look at productivity across the value chain no matter what. Again, what we wanted to indicate here is that irrespective of the outcome we stand behind this guidance and that should give you a certain frame. If I look at it mid-longer term, of course overtime you always have to adjust your cost structure to the top line, I mean you cannot spend more than you earn. So in this sense, there is always a correlation and the other way round if you grow with an organization, if you increase your topline, you also need more ammunition and you need more opportunities to feed the animal as it gets bigger and then you tend to invest more into research and development etcetera; you need more to launch the various products, so there is always a correlation. I wouldn't make it a digital correlation, I think we have to constantly work on our cost structure and we also have to be opportunistic along opportunities and not be slaves to some ratios but in the long-term of course there is always a correlation here. Right, Dan? Daniel O'Day: Alright Jeff, so thanks, I hope I understood your question well but - particularly lung cancer trials that we have, I mean - you know, we started the program and during the program we also amended it to include overall survival as a core primary endpoint, so these are true co-primary endpoints. The trials have thus - you know, some of them increased in terms of the number to satisfy the statistical significance also anticipating crossover. So this has been built into our trial design, this is a little bit of what I said about - first and potentially best, and that is - there is a trade-off of course between trial design recruitment, what you're going for end points and speed. And I think we're all making judgments as we go along here but we want to be fast but we also want to be thorough with these programs, particular because I think you know we'll only get a chance to show overall survival in some of these disease states once and early because once the standard of care changes, it's very hard to do trials to show overall survival. So we're happy that the EMPOWER 150 is enrolled now and clearly we're looking forward to that read out and the other trials are enrolling well prior to the standard of care and change, I think that's important. And you know, I can't - I can't forecast whether the end points will come in at the same time or different times but our statisticians have clearly factored in the fact that PFS is likely to read out first and NOS will read out a little bit later and there will be some crossover; so that's all taken into account in terms of the trial design as well. And I would just point out here that as we look at the chemo backbones that are most frequently used, it changes from region around the world but in the United States, clearly, the most frequent chemo backbone in the frontline setting is a the corbo-platinum plus paclitaxel with or without Avastin, in fact with Avastin is much larger than without Avastin. And then the third if you like chemo backbone in United States used today is carbo-platinum Pemetrexed. So you know, we've also thought through that in terms of having the chemo backbones that either are most likely to be most comfortable use with different physicians and who knows, I mean we could get different types of readouts with those chemo backbones. You know, we're informed obviously by our Phase Ib trials but we have to demonstrate that now in the full Phase IIIs. And then finally, again, it's a bit early to give you and I would love to give you more color on the second-line lung progress other than we think it's going very well; I mean the qualitative feedback we're getting is that the data speaks for itself, it's a very comprehensive data which I think is first and foremost important. You know we have a Genentech sales force and medical force that has incredible relationships with oncology in the U.S., I'm sure that's helping us. Yes, we've heard feedback that the Q3 we go seeing is also meaningful to patients and physicians. I just hope I can give you a little bit more data the first quarter, we just - we're just at a time period of the launch now where we just don't have the data to be more quantitative than that. So I'll have to give you an update of the first quarter.

Okay, thank you. We can take one more question perhaps here in the second half before we break out into the other sessions. Please.

Thanks very much. Richard Vosser from JP Morgan. Just one clarification question first, on the EMPOWER 150 readout, I think there was some suggestion this morning that our SMPFS would come together, you're suggesting I think now that PFS might be first, so are we going to have an interim of OS then or is it just PFS? Second question on S910 [ph]; could you update us on the ongoing safety of the product in terms of managing breakthrough bleeds in the extensions, also in the non-inhibitor patients and also the drug safety relative to bypassing agents? And then finally just on [indiscernible], we're going to have a couple of other IL-6s I think come into market, inflammation is quite a high cost burden, so just your thoughts on how we should think about formulary discussions, pricing pressure, not for '17 but for '18? Thanks very much.

Thank you, Richard. Those are meaty [ph] questions but - but I mean on the EMPOWER 150, we just don't know whether the readout identically or not yet PFS roll out. So - I mean, I don't know what more I can tell you that will depend on the maturity. I mean PF - whatever it's called primary, so when one is mature it reads out and triggers an ability for us to look at the data. At that time we may very well have a trend for OS, we may have PFS and OS both being significant at that stage. I can't guess that and that maybe different from trial to trial depending on the synergistic effect with the chemo background. So I can't give you more than that. So - but we've always - if we confuse you and gave you some confirmation that both would read out the same time, we didn't mean to do that. We just don't know at this stage. On S910 [ph], yes not too many additional updates since the JP Morgan but I'll repeat those; so in the inhibitor - the Haven One [ph] trial, I just want to repeat we had four patients that experienced adverse events when they used bypassing agent in combination with S910 [ph] at a relatively high dose, I would say. And to report all four of those patients have resolved, two are back on study medicine; and we haven't had additional indications of that since we put out the dear doctor letter indicating basically to make sure that if a bypassing is used in a bleed that it's used in a proper fashion and in a proper dose and proper escalation. So we haven't seen more of that in the inhibitor population. We haven't seen anything untoward so far in the non-inhibitor population, I mean it's still recruiting, and going well. So we haven't seen anything that would be more concerning to us on the S910 [ph] program that what we reported out. And then finally on the IL-6, I mean what we've seen so far and I know there is more IL-6 to read out is nothing we can see that provides significant efficacy or safety advantages to our Actemra. So far what we see is competition if you like coming into play over the course of the mid-term like you said with relatively me-too agents which of course, you know, is not - it's not nothing, we have to pay attention to that but we continue to see both our first mover advantage in this space and the experience in profile of Actemra to be competitive with what we've seen so far in late-stage trials from the IL-6. Thank you very much. Carl, we are breaking out in three different rooms, right. Alan and myself will the stay here. Dan is which room, do you know, Avenue. And Roland will be in Form-1. Thank you very much. See you later.