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

Sachin Jain - Bank of America Merrill Lynch Richard Vosser - JP Morgan Vincent Meunier - Morgan Stanley Steve Scala - Cowen Keyur Parekh - Goldman Sachs Andrew Baum - Citi Jo Walton - Credit Suisse David Evans - Kepler Cheuvreux Jeffrey Holford - Jefferies Tim Anderson - Bernstein Stefan Schneider - Vontobel

Ladies and gentlemen, good afternoon. Welcome to the Roche's Nine Months Sales 2016 Conference Call and Live Webcast. I'm Shayi, the Chorus Call operator. I would like to remind you that all participants will be in listen-only mode and the conference is being recorded. After the presentation, there will be a Q&A session. [Operator Instructions]. The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to turn over to Dr. Severin Schwan, Chief Executive Officer. Please go ahead sir.

Good afternoon everybody and welcome to our Q3 sales briefing. As you have seen this morning, we had good sales growth in both divisions and try to meet our annual target. On the portfolio side, let me highlight of course the reason to pull loss of Tecentriq. We also promised you to keep you updated on the expected read-out of APHINITY. APHINITY as the events are coming in. And if you know the originally guided you and expected that we have the good chance to get the data still by end of this year with the possibility that it slips into 2017 I said it's event driven. We now believe that it will not read out till in 2016. So we have updated our guidance for Q1 2017. If you turn to Slide 6 again good sales growth, pharmaceuticals up by 4%, constant rates pretty much driven by oncology and immunology diagnostics division up a solid 7% clearly outperforming the market, again very much driven by our immunodiagnostics portfolio as in the past. Turning to Slide 7 you'll see the quarterly development. Now we actually had a slower quarter in the U.S. I should point out that we expected to moderate in the fourth quarter, so really we believe that we keep the underlying momentum for the full year. Again on Page 8 the splits by region, all regions growing. If we turn to Slide 9 I'm really excited about launching five new medicines in the timeframe of only 12 months. This is really unprecedented in the history of Roche and probably across the industry. And the portfolio progress is I believe not only a matter of quantity but also quality and commercial, potential. And turning to Slide 10 you also see this reflected and the innovation and the potential of our portfolio in a number of breakthrough therapy designations which was received by the FDA five alone in the current year. Turning to Slide 11, the pipeline is progressing well. What is actually not captured on this slide is the Bolus of early development readout, we expect in cancer immunotherapy as we go into 2017 and 2018. And I mean also good progress on the diagnostics side, I'd like to highlight in particularly E801 which we launched in the meantime, our high throughput instrument in immunodiagnostics which is doing very well off to a very good start so far. Page 12 Tecentriq data, Dan will certainly cover this in more detail in his presentation, you of course seen the data which was occasionally presently presented at ESMO in Copenhagen. Let me just emphasize again that we received a very broad label, believe we are very well-positioned from a competitive point of view with very good data literally across all the various patient groups. To conclude, we confirm our outlook for the full year low-to-mid-single-digit sales growth. We expect EPS growth to be ahead of sales and on this basis also be able to further increase dividend in Swiss francs. And with this, I would like to hand over to Dan. Daniel O'Day: Thanks a lot, good afternoon, good morning from my side. I will pick right up on Slide 16 good growth, good solid growth in pharma division at 4% in constant currencies, United States driven by oncology, by immunology. There was a bit of softness as we mentioned on a couple of the oncology products, I will come back to that, but we expect that to moderate in quarter four and to really to continue to see good strong growth in the U.S. moving forward also the new launches. In the EU, plus 5% growth really led by Perjeta, by Actemra, by MabThera. So really good strong growth across several key European markets. To Japan I mean despite a significant mandated price cut with Avastin in April this year showing that they could displace that through good volume growth, mid-single-digit growth throughout the portfolio. And finally the international region at 4% as usual a bit of volatility in different parts of the world, I think a nice -- one of the nice bright spots is China growing at 8% which is driven by the underlying oncology products and you can see the currency impact turn positive here in the third quarter. So moving to Slide 17 always the overall summary slide I will talk about most of the products coming up in the slide follow that it's the first time on this slide we have added here Tecentriq and Alecensa together. So far in relatively short period of time, they contributed 150 million Swiss francs of additional sales to the third quarter. Let me just touch on a couple of other products on this slide first of all Activase again very nice growth generating over 800 million Swiss francs year-to-date with a plus 16% increase. And then on the declining side, year-on-year we show a slight decline by Tamiflu. As you know with Pegasys and Tarceva, they remain impacted by the effects of competition. Lucentis just a comment here, you know we've been receiving some in class competition, we see patient share really stabilizing now in the third quarter, down only 1% in the quarter. So we would expect this decline to really moderate at this stage and really have some nice news on the approval of our pre-filled syringe which happened also this week and also priority review granted for NCNV both what should help Lucentis as we go into the fourth quarter and into particularly into 2017. So let me drill down a little bit on some of the other major products going to Slide 18. Now these by turning your attention to the oncology portfolio with an overall 4% growth HER2 franchise growing at 9% really several important launches here including Cotellic, Alecensa, Tecentriq. I'm going to touch on HER2 and Avastin franchises in the following slide. So let me just cover some of the other important products in oncology here starting with the CD20 franchise. CD20 franchise grew year-to-date 4%, MabThera/Rituxan in oncology by 2%, Gazyva grew year-to-date 54% in the U.S. EU, nice to see the Rituxan refractory in non-Hodgkin's lymphoma setting start to off to a strong start approximately 20% share in new patient settings and then marketplace good early launch indicators. And there is a lot of stronger data that you will see in ASH including updated data on GADOLIN and overall survival as well as GALLIUM in the first line setting of course. Cotellic and Zelboraf year-to-date 21% growth back to growth after the approval of the combination of the end of last year. We see further first line market share gains in the U.S. in quarter three up to around 19% from where we started in the first quarter around 66% or so. And in the EU we're seeing shares of close to 12% for the combo and interestingly the immunotherapy market share seems to have stabilized in melanoma in the first and second line setting at this stage. We also presented the first data on our triple combination for Cotellic, Zelboraf, Tecentriq in BRAF positive patients at ESMO and we expect that Phase III trial to start in 2017. Another real bright spot is Alecensa for the positive presented on failed patients 174% increase U.S. launch off to a very strong start there, year-to-date sales of $47 million and second line market share already in a relatively short period of time around 35%. Continued strong momentum in Japan where we obviously launched some time ago and the EU approval in second line for Alecensa is expected later this year. Also the terrific news in the quarter, we got the second breakthrough therapy designation for Alecensa from the FDA and the first line setting based upon the J-ALEX data from Japan and I will get back to that also in a later slide. Last but not least certainly and we will pick this up in the innovation section, Tecentriq reported CHF 77 million of sales after only three months, bladder cancer patients share for the on-label segment exceeding 40%. As of today more than 3,500 patients have had the opportunity to benefit from Tecentriq and obviously we're just under early stages of the Tecentriq launch. Really pleased about the approval we got just two days ago in the second line lung setting, which I will get back to. So moving on to Slide 19 now and shifting from year-to-date growth to quarterly growth, quarterly growth quarter-on-quarter for quarter this year versus quarter last year, the HER2 franchise 8%, Perjeta growing at 24%, in the EU 42% obviously having launched a bit later driven by neoadjuvant and metastatic breast cancer setting and in the U.S., we continue to see further penetration with an 8% growth in the metastatic breast cancer setting. Herceptin is 5% U.S. essentially flat and this was really due to some effects on the higher reserves relative to 340B, so it's a bit of updating the fact between different quarters for this but as we know Herceptin also has a very high penetration in the U.S. certainly in the metastatic setting. In the EU we see Herceptin sub queue sales now approaching 50% of sale shares with a 4% growth, the international growth plus 10% really strong growth, good double-digit growth in China as well. And Kadcyla at plus 5% well established in the U.S. and we have stable shares basically at this stage in U.S. and Europe because of the high penetration level, good growth in Japan at 4% and now obviously in the international regions a 44% growth rate, we're just beginning to be launched now at this stage. Just to iterate we are obviously monitoring the number of events for the AFFINITY trial and based upon our most recent assessments, we now expect the trial to read out in the first quarter of 2017. On Slide 20 digging a little bit deeper on Avastin the quarter three sales growth is minus 3% as you will see impacted by the U.S. and Japan, good strong growth for Avastin overall for the year at 18% in the international region, in EU relatively flat a tail of two stories I mean good growth in Germany with some of the new indications, but we also had a impact in the UK delisting of some indications as well so that moderated a bit that growth in the EU. And then the U.S. is minus 9% quarter three sales were down due to some softness on the second line lung cancer setting which is a relatively small indication for Avastin and also some increasing reserves to 340B. But I think it's much more appropriate to look at the year-to-date figures for Avastin in the U.S. as we expect this quarterly effect to moderate Avastin in the U.S. in the fourth quarter as well. And Japan despite the 10.9% price cut is down 6% so they are offsetting with some volume and obviously as time goes up price effect has less of an impact on the totality of the sales in Japan. So moving on to Slide 21 the immunology franchise quarter three immunology franchise growth of 10%, Xolair 13% really being driven by both the allergic asthma indication as well as relatively strong CIU growth really around half and half driving those two levels of growth. Really pleased to see the FDA approved Xolair for patients from six to 11 years from moderate to severe asthma. It's the only biologic approved for pediatrics and certainly we think that we are continuing to drive the growth of Xolair as we go into the end of this year and next year. Actemra plus 15%, 38% of the growth of the total sales for Actemra coming with sub queue indication where we continue to see good strong growth in the EU and the U.S. and in the EU, Actemra has now become the first and the number one medicine in the RA market in the first line monotherapy setting. So we continue to see good growth of Actemra. And our second breakthrough therapy designation for the quarter came with Actemra in the giant-cell arthritis indication where we have also received the breakthrough therapy designation there. You will see the top-line results from this -- the patient group at the ACR in November, it's actually a part of the opening plenary session. So some significant data there that I think you will be interested to see the detail on. And then finally, MabThera/Rituxan in the immunology indications growing at 4% driven predominantly from the vasculitis indications that we've had a benefit of in also previous quarters. On Slide 22, Esbriet sales development continues to be well on track plus 35% for the quarter in the U.S. plus 38% in the -- sorry 38% in the U.S., 33% in the EU. Year-to-date sales were up 45%. We have established market leadership in the U.S. and in the EU as market shares above 60%. In the U.S. we have a good strong underlying growth, the run rate of new patients is well on track, the growth continues to be driven in the moderate and progressive patient segments, and we are having basically the first entry into the mild patient segments where in the mild patient segment the treatment rate today is only around 30% and that will be a key part of our strategy to increase that mild patient treatment paradigm. And then in the EU we have the incremental revenue is driven by greater penetration into the diagnosed patient population as well as launching into new countries. So the outlook for the remaining part of the year with Esbriet, we work through the bullets that we talked about in previous quarters, so we are now seeing more of a steady state growth moving forward a growth that's going to be driven by longer treatment times by increasing the milder patients segments and those are things that we're working on with the patient and physician community for this devastating disease to get it treated also earlier. So shifting to the innovation section 24 now shows that we have 14 breakthrough therapy designations, two added as I've indicated, one with Actemra and one with Alecensa. We just mentioned that in addition to the J-ALEX data and the breakthrough therapy designation on Alecensa, we expect the global Phase III ALEX study to read out in the early part of 2017. So let us talk a little bit about cancer immunotherapies Slide 25 I get a focus on Tecentriq first. Severance showed the overall Kaplan Meier curve. Let me drill down a little bit into this data, some of you have heard this at ESMO but I think it's worth repeating. You know the OAK really placed Tecentriq with an all-comer label in a very good position in the second line setting, the data were consistent at what we have seen previously observed in the other randomized trials that we had certainly the Phase II Poplar study and the strong OS benefit demonstrated an OAK, included a strong benefit in low and no PD-L1 expressing patients. This along with every three week dosing makes it I think very competitive in the second line setting this all-comers label with convenient dosing. But for the first time, we now have a PD-L1 immunotherapy approved in lung cancer and as you have heard at ESMO as well, we are beginning to see a body of data that may suggest an important role on the different mechanisms that we see within PD-1 and PD-L1 compared to the other cancer immunotherapies on the market today. So some evidence of that is to be seen consistent OS benefit across of variety of different subgroups here including non-smokers with hazard ratio of 0.71, patients with CNS brain metastases with hazard ratios of 0.54 and the data clearly show a survival benefit across both squamous and non-squamous histology. On Slide 26, these are some of my favorite graph to look out of the OAK data. The overall curve compares very well to other trials in similar settings, including the tail and I remind you that the primary analysis that you're looking at here on these slides on Slide 26 comprise the first 850 patients only. The overall survival data maturity for the entirety of the 1,250 patients which is the defined ITT population is not yet reached, we'll continue to follow on that. As you know, OAK is an event driven study, so the timing of the secondary analysis is not yet known but when it's available we'll certainly update you on that as well. So a couple of things as you look at these graph I mean it's important to note that all the curves separate around three-month point and do not cross the chemo arm. And although crossover was committed 5% of the patients in the Tecentriq arm, 17% of patients in the docetaxel arm received subsequent cancer immunotherapy. So and despite this actually Tecentriq demonstrated statistically significant and clinically meaningful overall survival benefit. If you look at the upper half of these slides, you see the slides for the PD-L1 negative patients with true PD-L1 negative patients TC0, IC0, and as stated before both population benefits irrespectively of the expression levels with hazard ratios of 0.74 and 0.75 respectively. So it's the first time we've really seen a very clear benefit for low and no expresses when treated with the PD-L1. And in the lower half you clearly see that there is no difference to the histology non-squamous or squamous patients they both received a hazard ratio of 1.73. So good strong data not only for lung cancer, but I think also establishing Tecentriq, it's mechanism and how that may play out in other disease types as well. On Slide 27, I know this is and I chart this table summarizes our current cancer immunotherapy program it's quite challenging to keep up. But let me quickly summarize a couple of changes since the half year. We've added several Phase I and Phase II studies. We have two additional Phase 1 Tecentriq combination studies in AML, two Phase I Tecentriq combination studies in solid tumors with partner compounds. A Phase II study of Tecentriq post-Kadcyla and HER2-positive breast cancer and a Phase I study of Tecentriq Cotellic and Avastin second line colorectal cancer. In addition you see a second green tick here with the approval in second and third line lung cancer. And I also want to use this opportunity to highlight again the ongoing efforts in lung cancer. I think we are in a very strong position. We have one of the largest immunotherapy programs with lung cancer with seven Phase III studies of Tecentriq in first line lung or the adjuvant studies are in the early stage. And this include the five studies for evaluating Tecentriq as a first line therapy in combination with platinum based chemo and Avastin or platinum based chemo including Pemetrexed for people with non-squamous and squamous lung cancer. Moving onto Slide 28, looking at the broader immunotherapy portfolio. This slide tries to summarize that we expect first read-outs on our early CIT assets within the coming 14 months between now and end of 2017. The cancer immunotherapy portfolio now includes more than 20 investigational candidates, 10 of what should be evaluated in clinical trials, and one of those obviously Tecentriq already in the market. As you can see, we expect the first read-outs for all our CIT molecules between now and end of 2017, including several data from immune dublets such as the CD40 emactuzumab and Tecentriq combinations with the IL2 OX40 with the biospecifics CEA/CD3 and the IDOi and the TIGIT. So not only will this data tell somebody about the individual molecules obviously, but I think they will be very informing on the scientific rationale for selecting these combinations that are ongoing programs in selecting different combinations to help us better understand what's going on with the immune cycle which is going to be fundamental for the field and certainly for our very large portfolio. I also want to take this opportunity you may have noticed that we took off vanucizumab in the monotherapy setting of this list as of the quarter three and that's to the read-out of the Phase II study the McKay study and the fact that the primary endpoint of that study was not met. We set the bar very high. We have always but by testing vanucizumab against Avastin the first line colorectal cancer study we set the bar very high and did not achieve that bar. But we do also have vanucizumab in ongoing studies with other combinations that are enrolled now and will continue those studies to see how read-out is of those as well. But again I remind you that our Avastin strategy as we approach biosimilars was strengthened by the significant number of immunotherapy programs and targeted therapy programs we have with and without Avastin across the seven diseases in which Avastin is approved today, including things like lung cancer that we've already seen some good progress with Tecentriq. Now leading oncology on Slide 29, just a quick update on Ocrevus. We had a good share of voice at the ECTRIMS Conference in London in September with an oral presentation in posters that should new endpoint which you see on the slide for RMS and PPMS. So these are analysis that are focused on the previously released clinical trials that really look at how are we looking at no evidence of disease activity or no evidence of progression in RMS and PPMS respectively. And the analysis showed in RMS even after we baseline in a week 24, we had a 72% of the Ocrevus treated patients achieved no evidence of disease activity between week 24 and 96. And in the case of the Oratorio study in the PPMS, the new composite endpoint of NEP or no evidence of disease progression, the analysis showed that there was a 47% relative increase in Ocrevus treated patient who showed no evidence of disease progression essentially indicating those patients remained stable. And on the regulatory side the reviews with the FDA and EMA for RMS and PPMS are progressing well and we have the PDUFA date in the United States at the end of the December everything continues to progress towards that objective with RMS and PPMS in the U.S. On Slide 30, a quick update on emicizumab in hemophilia A. New data was presented in July 27 at the World Meeting on Hemophilia and what you see here is after longer follow-up than previously reported so we're now up generally beyond the two-year mark, 21 to 33 months. We see emicizumab continues to demonstrate promising prophylactic efficacy in adults and adolescents with hemophilia A, both with and without factor VIII inhibitors. The safety profile in emicizumab was also consistent with our previously observed in the earlier follow-up and which had results of this longer follow-up continued to confirm from the data that we saw at the 12 weeks of treatment mark that was published in the New England Journal Medicine sometime ago. We have also made some very good progress in the development program. In addition to the inhibitor program being fully recruited and expecting that to read-out at the end of this year, we have started the trials on non-inhibitor population and also on the pediatric trial and we would expect the Q4 week dosing study to begin before the end of the year. So moving to Slide 32, the picture is basically identical to the half year update, let me just give you a couple of highlights here. Tecentriq approved obviously second and third line by lung with an all-comers label, the two new breakthrough therapy designations, and the emicizumab top-line data coming in quarter four. Quick update on Lampalizumab, patient enrollment is on track nearly completed and we expect the Phase III results for the CHROMA and the SPECTRI studies on the 12-month endpoints to read-out in 2017. And then the Tecentriq plus chemo come in the first line setting, no change here, we expect the first read-out of the Emotion 150 study that's the Tecentriq chemo plus Avastin it's still possible in 2017 based upon the PSF endpoint. Slide 33 ASH is going to be a good strong meeting this year for our strong presence over 61 abstracts, one plenary session, 21 overall including being in the Presidential Session this year. We'll host an Investor Analyst events on December 5, Monday of that week and we prepare to have all of our key leaders there for the hematology portfolio to answer any questions you may have at that stage on the portfolio and the key topics for the ASH this year will be obviously the GALLIUM's first line read-out, follow-up data on the GADOLIN in terms of extended OS data, and then the GOYA study as well in understanding how the GOYA read-out in detail occurred. We will have also updates on our antibody drug conjugate Polatuzumab vedotin in Phase II data and the relapsed refractory Follicular lymphoma setting, the ROMULUS study and also Venclexta our Phase II data and the relapsed refractory follicular lymphoma study as well. And then finally on Slide 34 to give you some changes for three we've ticked down to essentially in the second and third Phase III results of the OAK, just the AFFINITY read-out for more clarity for the first quarter of 2017 and also the Phase II results in topic dermatitis for Lampalizumab were presented at the European Academy of Dermatology and we will be taking further decisions on that program in the near-term. So thanks for that. I will turn it over to Roland for diagnostics.

Thank you, Dan. Good afternoon, good morning from my side I'll go straight into the sales results for diagnostics for the first nine months, strong growth overall with 7% increase in sales and excellent growth even if you look at the clinical diagnostics market and sales overall was 9%. We had some softness in diabetes care with a minus 2% for the three quarters. I think the reasons are well known it's based on the U.S. pricing and a continued spill over into the private sector. We have also seen the second round of the so-called competitive bidding process which led to a reduction of 20% on the blood glucose monitoring strips. On the positive side, the third quarter was positive in diabetes care with plus 3%, some base effect here but then also very good sales particularly in Latin America. But overall continued strong growth in the clinical diagnostics business for professional diagnostics as well as molecular diagnostics and as well as for tissue diagnostics. Strong growth also moving to Page 37 across all regions. I would like to point out the emerging markets and in particular the E7 countries up 21%, some very strong markets amongst them Mexico up 37%, and good growth in India and then continued very strong growth in China with 22% year-over-year. I think the global presence that we have here is allowing us to compensate for some weaknesses in some markets for instance in cases of political instability or other. The last point I would like to make on the slide is North America which is up 4%, if we do exclude the aforementioned diabetes care impact, sales in the clinical diagnostics are up 8% in North America due to a very strong result. Going on to page 38 in some of the drivers on the sales line in particularly immunodiagnostics as you can see here in the top-line, this is now immunodiagnostics growing 14%, franchise now accounts for over 2 billion sales on the manual basis and if you take that together with clinical chemistry, so our serum work area even 3.5 billion annual sales. So this franchise is doing very well. In molecular diagnostics virology up 11%, including HPV up 13% and then in tissue diagnostics the core business advanced staining up 9% and in primary staining benefiting from the recent launch of our primary staining automated platform, the HE 600 up 19% or 18%. Moving to Page 39 and some updates on portfolio development and launches. The far most important launch for us is the launch of the immunoassay analyzer cobas E801. This is a module that is the latest addition to the cobas 8000 family which caters to high suite of laboratories. Here we see a picture which is actually this cobas 8000 family of products and solutions. The 801 being the key module here and as I mentioned earlier 14% up for the nine months, immunodiagnostics is the main growth driver, and with this new module which can be fitted into installation seamlessly and in the modular basis, we have again a very dedicated solution for high suite of laboratories. The module itself offers double testing capacity on this footprint, can be operated by one person as an operator only, uses less disposal, less blood tubes, and then many other features. So in essence what it does offer the laboratories is efficiency, productivity gains and allows the laboratories to operate more profitably and this is why we're very encouraged also by the launch. We have 80 instruments placed in the first three months after launch, so a very good initial launch phase indeed. Shifting to molecular diagnostics on Page 40. We observe similar trends in molecular diagnostics than in the central laboratory. So the need for automation, comprehensive solutions, and high walk away time for the operators, and then overall modular systems and this is what we have. We have large panels of test that can run across several instruments, families that cater to the different customer segments, so from low to mid to high throughput and that continues to drive the sales as you can see here increased HCV sales due to new treatment options, increased HBV sales due to particularly the prevalence in Asia Pacific and an increased HIV sales in part to a global access program that we run out of South Africa for all of Sub-Saharan Africa. And that would not remise to mention, we continue to increase HPV sales up 13% for the nine months particularly driven in the U.S. by the FDA approval for primary screening. Then moving to page 41, I would like to give you an update on Zika which initially was actually just an opportunistic approach. I'm trying to be fast in responding with the broad solution to a looming health crisis and in the mean time has actually moved into an opportunity to truly differentiate into grow sales especially since the FDA mandated a testing in the blood screening and this is what you can see on the left side of the slide. We do have since March 30 under an investigation with new drug application, the cobas Zika test for blood screening. To-date, we have been able to test more than 300,000 blood donations. We have been able to remove from distribution over 300 positive donations, and then we have on the right side of the slide, the clinical diagnostics also nucleic acid based testing for qualitative detection of the Zika virus with two solutions and we expect to have a serology, so an antibody test available by next year. So this is truly been a nice addition to our overall leading portfolio in molecular test. Moving to 42 to give you an update on the point of care molecular diagnostics with the test approved for combined flu A/B & RSV. This is important; it gives us lab like quality for point of care PCR results in just about 20 minutes. And in particular the approval for Influenza A/B &RSV it is the fastest real time PCR test with a CLIA waiver. So it allows us to be to go into non-laboratory testings and it is for the detection and disclamation of Flu from RSV and this is a very particular importance to make appropriate treatment decisions for immuno compromised patients such as pediatric patients and the elderly. So with this, we now have a complete respiratory infectious panel on one hand and we have been able to over the last nine months ramp up the manufacturing process, so we are indeed ready for the upcoming Flu season in the Northern Hemisphere. And this brings me to the last slide which is 43 just an overview of the key launches for 2016, as mentioned the cobas 8000 E801 is by far most important for us, I've mentioned the cobas Liat. And I would like to just point out at the bottom of the page, the companion diagnostics for PD-L1 both for bladder cancer and non-small cell lung cancer for Tecentriq as a complementary test with which we continue to measure expression on both cancer and immune cell and of course we can continue to support the cancer immunotherapy franchise in the pharma side which is a nice addition to the diagnostics portfolio together. So with that, thank you very much and I will hand over to Alan.

Thank you, Roland. Yes, just a couple of comments on currently impact on sales and then a projection on the year-end on sales, the core operating profit, and core EPS growth. So I move to Slide 46 and so let me start with current impact on sales year-to-date September left hand side you see the sales increase in constant currencies year-to-date September 2016 versus year-to-date September 2015 4.3%. And this one on the right hand side where you see the increase in Swiss Francs and you see the difference between the two is 1.3 percentage points. In between the bridge you see on one hand the currency which in fact we tend to get Swiss Francs, so really the U.S. dollar or the Japanese Yen or the Euro and then in other currencies that which against the Swiss Francs. And I think on other you see the British Pound and to a certain extent after the Brexit so that contributed a little bit and then you see larger impact coming from Latin, where that's the Argentinean peso, the Venezuelan bolivar, as well as the Brazilian real, which impacted if you like the bridge in a negative way. Overall the 1.3 percentage point difference represents a positive impact of 457 million Swiss Francs. And with that let us go to Slide 47 where we delivered yet, a kind of an expectation of how the whole thing would like yet in 2016 and you know that this is based on the assumption that all the currency base on September 30 remain stable until year-end. And what you see is basically, is a very consistent picture, now that we have delivered in Q1, 2016. And interestingly the British Pound impact that's pretty much mitigated by the Japanese Yen. So what do you think and let me start with the left hand side is a relative strengthening of the Swiss Francs against the U.S. dollar and the Euro and as you can see the U.S. dollar and the Euro strengthened last year second to Swiss Francs. So this reduces the positive effect on sales, core, and core operating profit at year-end. And let me outline now, around the potential on currency impacts on sales, core operating profit and core EPS at year-end. If you see on sales plus one percentage points, you see on core operating profit basically no impact and if you see core EPS at year-end, the projected impact of plus 2 percentage points. And with that, let's go to Slide 48 [indiscernible] represents that. And with that, we are happy to take your questions.

Can we take the first question, please?

We will now begin the question-and-answer session. [Operator Instructions]. The first question is from Sachin Jain, Bank of America Merrill Lynch. Please go ahead.

Thanks for taking my questions. First one is a clarification for Dan, I think in your introductory comments you mentioned EMPOWER 150 the chemo combo Phase III was possible in 2017. I just wanted to clarify was any delay versus my perception from the middle of the year and the data end of the year, how would you frame timelines related to Merck and Bristol Phase III studies with that chemo combination. And secondly again on chemo combo there has been some discussion post-ASMO after Merck Phase II data it was published potentially companion listed and driving off label usage, can you provide your perspective on that and do you have any similar intentions yourselves from your Phase I/II data? And then final question just on the EUS performance, you mentioned that pressure should abate into the tail end of the year, could you clarify what gives you comfort on that, it is just 340B if so what are the driving factors there or are there any other factors we should be aware of? Thank you. Daniel O'Day: Okay, thanks, Sachin. So yes just I want to make sure my wording is precise. There is no change to our expectation for EMPOWER 150. I mean we are still expecting that trial to read-out in the second half of 2017 that's the arm with Avastin. From a PFS perspective and from a timing perspective competitively, I mean we think the trials of Merck and BMS, some are obviously the CTLA-4 combo and some are now the chemo combo frontline would be reading out, quarters I think on either side, and not significantly differently, around the same time is what we expect between the middle of next year to the end of next year kind of a thing. So that's I think it will be a very rich year of read-out in the first line setting and provide data, I think help us all answer the questions around how do we begin the segment first line patients particularly beyond those that highly express PD-L1 in the monotherapy segment that we've already seen from Merck so. I'm sure, you are and I'm certainly looking for those read-outs, we continue to be very positive about our frontline studies and may be they will transition into the -- your second part of your question which is how do we interpret the Merck data that was presented at ESMO obviously it's a different molecule, it's a different backbone, it's a different study but what I would say is that it tends to have another proof point I guess to affirm the role that chemotherapy combined with immunotherapy in the frontline setting could play for patients moving forward. Our first -- our Phase I data continues to mature as you know we will certainly update you the next time we would update you is when the OS data matures in our Phase I frontline chemo combo obviously that will be event driven, so we will see when that happens but we will certainly update you on that. And I guess what I would say is just continue to encourage us the data we saw at ESMO that we are on the right track for certainly some proportion of patients in the first line setting with the chemo combo. And again, I think we have the broadest program with the chemo combo programs with Avastin with and without chemo, with Pemetrexed, the monotherapy, so a very broad base of studies as well and including obviously squamous and non- squamous as well in Adjuvant study. Happy to go in more detail another questions on that, let me just finalize that the third question that you had there around the pressure abating in quarter four. Yes, I think as you know I mean just articulate what the 340B program is in the United States. I mean it's a program that provides discount to hospitals that qualify for basically increased discount and then two things that are happening there, one is that we see a shift in private oncology practice being acquired by hospitals that are 340B qualified and we also see institutions getting better and better at increasing their ability to claim the discounts. Having said that, I just remind you this is a bit of a gradual effect in fact as of the second quarter 340B sales account for around 18% of the total volume of Genentech products to slight increase over 2015. So if you can imagine this isn't evolutionary not a revolutionary process. However when you look at it, quarter-to-quarter you can have reconciliation issues, you can have certain things coming into the quarter that may amplify these situations and that's why I say that is on the two products that we see perhaps the greater effect of so that's Avastin and MabThera. Our forecast indicates that we will see lessening of that effect, there is a slight sales days difference between the two quarters as well and that's why you know the underlying demand to those products, we feel good about and that's why we're quite confident that we will see moderation as we go into the fourth quarter for those products. And then obviously increasing sales of Tecentriq, Alecensa potentially the first sales we might see on Ocrevus, so you know you are going to see, you're going to see the new products also ramping up and moderation of what we saw in the fourth quarter for the key products. So I hope that gives you some more color on my confidence.

The next question is from Richard Vosser, JP Morgan. Please go ahead.

Hi thanks for taking my questions. Couple of questions on AFFINITY if possible please, so given the duration of the trial that's been going on for with the four-year survival is probably going to be somewhere in the low 90s potentially for the trial. So based on this and your experience which Herceptin does this change anything about your view on AFFINITY success? And then secondly the 379 planned event, if you could give us just a little bit more color where we are on those events if possible Q1 is quite a long time, are we thinking around the turn of the year for this trial read-out. And then finally just back on Tecentriq, just on the combination with Avastin in renal cancer I realize it's event driven the Phase II but just any color you can give us on that trial when we might see something if you're seeing anything, I think it's potentially interesting read-out for that. Thanks.

Yes, thanks a lot Richard, there are couple of things. Obviously we are blinded to the AFFINITY trial, we are giving you the updates that we know nothing has changed from our previous quarter discussions on AFFINITY or how we perceive the likelihood of success there given the data and neoadjuvent and also the metastatic setting, it's just simply an issue of the events coming in, I mean sufficed to say that the study is powered to show a significant effect in the disease free survival and nothing in our opinion has changed the progress to on that trial to be able to look at if you could read that positive achieving that significance of difference. So relative to the timeline of the 379 events, I can't really update you any further on that once we get the last event for the period of time that we analyze the data, clean the data and then the data reads-out, those are all taken into account when we look at both the progression of events and also the required time to get the initial read-out obviously we will be issuing the press release, when we get the read-out. At this stage, we just based upon what we know; we saw the likelihood of that coming into December to be really not there anymore. So we moved it into the first quarter of next year. And that's all I can really say on that right now but we continue to be and look forward to that event reading out. And then finally on Tecentriq plus Avastin in renal cell carcinoma the Phase II trial we should be getting results in-house still this year, so the end of 2016 and most likely that would then target an ASCO GU presentation in February and that would be the -- the meeting I think we would be targeting for release of that data as well.

Next question is coming from Vincent Meunier, Morgan Stanley. Please go ahead.

Thank you gentlemen for taking my questions. The first one is on Tecentriq and a bit of follow-up, is the crossover in EMPOWER 150 a load and could an early crossover explain the lack of separation of the curves for overall survival in keynote 21G and that absent very early in the course of the trial. And I have another question on Avastin; I mean is it fair enough to say that the drug is ex-gross with the U.S. decline being offset by Europe and international and what about your 7 billion sales targets you're mentioning, I mean a year ago? And next question is on the biosimilars, can you confirm the timing for European launch in second part of 2017 for Herceptin and also for Rituxan and then '18 in the U.S.? Daniel O’Day: Good, yes. Right, good so let me catch up with your question, so Tecentriq no in the EMOTION 150 trial crossover is not allowed. I would not try to interrupt that the chemo trial I think that's a better question for another company but I think our trial design does not permit the crossover in that particular trial. So Avastin growth in the EU. Vincent can I ask you just one more time state what your question was on that?

Yes I mean based on the trends we see now for Avastin, we see that in the U.S, it's ex-gross it's declining, there is a bit of growth in Europe, there is a bit of growth in international, is it fair to assume that overall the drug now is stabilizing and getting to your target of 7 billion? Daniel O'Day: Okay thanks Vincent. Yes I mean I think it's always hard to predict the future but I would say clearly in the United States we're at a high level of penetration, a high level of maturity. I think we see it the stabilizing for sure in the United States, in Europe the 2% growth as I said is kind of a impact from the optic communications and some access in pricing pressure in couple of markets. I think it's also very well, well saturated in Europe as well. Japan still has some demand growth going on. I mean you can see Japan minus 1% for the first three quarters despite the 10% decline. And then international, I think has a lot of opportunity and with 19% growth still an opportunity. So the question of course is how does Japan or international in particular offset some of the stabilizing growth in the United States and Europe, certainly we don't expect significant growth on the Avastin franchises as we move forward. Having said that I think the dynamics I'll have to update you on for 2017 at our next call overall but we continue to launch new indications, achieve reimbursement for Avastin outside of EU and U.S. and continue to drive the growth in the markets that we can. And then finally biosimilars absolutely nothing has changed over it's exactly on track with what we heard about the filings in Europe with Herceptin and Rituxan. We would expect the first biosimilars for those enter in the second half of 2017 at that stage for Herceptin and Rituxan in Europe and then outside of Europe for and also with Avastin, we expect those entrants to start only after 2017 and between 2017 and the end of the decade.

The next question is from Steve Scala from Cowen. Please go ahead.

Thank you so much. And Dan a higher level question but how do you think the lung market will play out for IO over the next one to two years. What portion of the second line market do you think Roche will garner some peg it at 20% share, what is your reaction to that? And how much will the second line market shrink because the first line treatment IO treatment is available and first line IO progresses probably won't get IO in the second line? And then the second question is what are ongoing discussions with FDA telling you about the likely of Ocrevus label particularly relative to safety. Thank you. Daniel O'Day: Thanks Steve. So yes I mean I guess it has been interesting to see the data in the lung setting in general I'm sure for you and for us as well in the past three or four months. I think it affirms in our strategy which is the following first of all that that we saw -- we see strong effect Tecentriq as a monotherapy in the second line lung cancer setting in both diagnostic positive and vial diagnostic negative segments. We also saw as a result of I think some of the competitive data in the frontline setting that it will take a variety of approaches to be successful for patients in the frontline setting. And our understanding of the science continues to evolve, our understanding of biomarkers continues to evolve, and very worthy our combination strategy continues to evolve both for those read-outs that we have in Phase III right now or expected from Phase III but also our blog on immune oncology portfolio which as you know we learn a lot about as we go into next year about what are the types of immune dublets might be successful what are the types of five specifics maybe supported. Now specifically to your question how I see lung playing out. I think we will be very competitive in the second line setting. Bottom-line is that we are one or two companies that are all-comers label, where the only company to show the effect across so many different subtypes I think so convincingly so that both squamous, non-squamous, non-smokers, brain mets, and true PD-L1 negatives. So I think obviously we're third to market, but second within all-comers label and I think a strong efficacy profile, good convenience profile with every three week dosing and a very strong presence in share of voice on oncology in the United States with our Genentech colleagues. So I'm not going to give the exact market share, but we are certainly looking to have a play there. And then relative to the first line setting I think what's important to point out now is that, what we've seen so far from the data and the Phase III is that we are really talking about diagnostically rich populations. So 25% to 30% the population of what the percentage of that will benefit from a monotherapy or not even the entire being the 25% and 35%. So what does that mean? That means still the vast majority of patients for the foreseeable future until we see how standard of care will change in the frontline setting, unfortunately will continue to progress to the second line. And therefore, the second line setting I think will certainly have a small adjustments but I don't think a huge adjustment over the course of the next 18 months to two years. And then of course we'll have to reevaluate the market once we get the frontline read-outs that we expect to have next year as well. So I think that those are some of my thoughts at this stage and we're data deprived in the frontline setting next year we will be data enriched and we'll see how we progress. But I like our strategy of looking at locations in the frontline setting with a combination therapy and a variety of different combinations together the broader percentage of patients and just those that are PD-L1 positive. Oh, Ocrevus, I'm sorry. Ocrevus, we -- yes, the review processes is going well. I mean I think the profile has spoken for itself again reinforced it at Ectrims with some additional information from both an efficacy and a safety profile. So everything is progressing with our discussions with the agency and both the RMS and PPMS programs are having good discussions and it's still our expectation that we -- that we get a response in -- around PDUFA date, I would say.

The next question is from Keyur Parekh, Goldman Sachs. Please go ahead.

Good afternoon and thank you for taking my questions. I have two, if I may please. The first one Severin just looking at your thoughts on the broader political environment in the U.S. kind of where do you see pricing going over the next couple of years, what do you make of the various kind of initiatives that are being popped around by the CMS, would be great to hear some of your thoughts on that? And then secondly, Dan just to confirm you did you hear say that there is no crossover in EMPOWER 150?

That's correct. Okay. So let me take off with the first question on U.S. pricing. I think what I would emphasize is how focused as a company on truly innovate very specialized medicines. And what we see that the sound of this, our portfolio is much less exposed to pricing pressure. We certainly have not seen the pricing pressures some of our peers have seen and it's really the nature of our portfolio. So will something change after the election? I don't see a fundamental change whatever the outcome of the election we will together with the government, we will work together with various institutions to improve healthcare delivery in the United States and again on a higher level, if you bring real true innovation to the market in the United States, if you prolong life, if you improve the quality of live in a meaningful way, I'm personally convinced that it is in particular to U.S. which revolves this kind of innovation and also allows to continue with our premium license package as we go forward. Dan you want to try and take? Daniel O'Day: Yes just to take it to the next level of detail I agree with everything Severin said, I think just remind you that in terms of potential changes in the U.S. pricing environment and today the Genentech's portfolio in the United States is the majority, the vast majority is Medicare Part B. And as you know we already have if you like a certain price regulation in Medicare Part B in terms of the infused products and the price increases that we take. So and they are quite smaller, they are restricted for the reasons that we all know from a structural perspective. So first of all, I think we have -- we're launching strong products and strong new products in the Medicare Part B setting. We are pricing them I think appropriately for broad distribution and access and our past model and our future model doesn't rely upon and doesn't suggest that we would be taking large price increases, so on the entirety of our portfolio. So I think that's important to note. Will there be changes in the pricing system in the U.S.? Certainly we are in discussions with CMS around a variety of those. Certainly if you know on the Medicare Part B pilot program, we as well as Congress and others have provided input on the first proposal that still being digested by CMS and we think likely as we go into next year there may still be a pilot. The question is what will that pilot look like and I think it will take into account many of the comments that have been expressed including the concern around increased side of care shift which could create actually greater cost for the U.S. healthcare system by shifting patients into them to more expensive settings like the hospital settings. So we're convinced that those types of things will be taken into account and we're working, we are very much likely to engage in further dialogue on the second part of the Part B pilot which is the value based pricing proposal, way to look at increasing the opportunity to look at combination based pricing, those types of things in the U.S. which we think are going to be fundamental as we go into the new area of immunooncology and all the combination trials will be read-out in the years to come. So we're encouraged that CMS is engaged and speaking about those things and also I think that the impact for our innovative product line will be diminished overall in the coming years.

Dan, can I just follow-up for a clarification there. Daniel O'Day: Sure.

You said historically it's been limited price increases. Can you put that in context, are we talking kind of 2%, 3% price increases across your portfolio in the U.S. on an annualized basis are we talking lower or higher than that? Daniel O'Day: That's right; it's in about that range. You can only do a certain percent about every six months otherwise you have an impact on putting physicians under water with medicines that they have at their infusion centers. So that's right, it's around that price increase on the entirety of the portfolio.

Next question is from Andrew Baum from Citi. Please go ahead.

Good afternoon. Couple of questions. Firstly for Dan picking up on the comments on the size of second line lung market and the impact of the anticipated first line approval for KEYTRUDA on the back of keynote 24. That conversation is highlighting that the likely usage of KEYTRUDA is going to be significantly greater than that 25% to 30% identified in the trial, assuming that's right then presumably rather than a smaller dislocation there is going to be some contraction of the second line market. I just like to hear whether that's consistent with Roche internal expectations in terms of how this dynamic may shift, the size of second line market as even patients stay on PD-1 until progression? And then second just turning to AFFINITY, could you remind us of the enrichment for node positive patients in AFFINITY in particular patients who have more than three nodes within that trial, just so to help to frame given the updates and another datasets in the backdrop, thank you. Daniel O'Day: Sure Andrew, I will do my best to take you little bit again to the cancer setting. I want to make sure I was clear, I think that this location that you speak about in the second line setting, I'm not suggesting it won't occur over time but I do think there will be an evolution to that and I don't think it will be dislocated entirely or to a large, large extent by just a monotherapy indication of frontline setting. So I don't think I could get into a debate around exactly where KEYTRUDA may or may not be used in terms of PD-L1 positivity. But I think it is -- I think that's important now. I think the other thing that I think remains to be seen is what will happen to progressors immunotherapy in the frontline setting, what treatments will they receive in the second line setting. If a physician decides to extent or to switch immunotherapies then we think there is at least some logics to suggest that if you been treated on a PD-1 in frontline setting then you might be treated with the PD-L1 in the second line setting. No I missed, we will see exactly how that plays out but I think not just the mechanism but also the strength of our data in the second line setting at least would provide options for patients particularly depending on how quickly they may progress after frontline immunotherapy model setting. So I clearly, I don't want you to think at all that I don't think if I'm talking about two to three years that we won't see and I hope we do by the way for the sake of patients that we see a decline in the second line setting because we're getting so many patients in the frontline setting with long durable responses but today I don't see that data existing to suggest if that's going to change in the next year to 18 months. So that's my opinion. And secondly on the AFFINITY, we have never completely disclosed the exact balance on the node negative, node positive. That we feel very good, our statisticians feel good about the balance of node negative, node positive. We adjusted the recruitment back in moment of 2013 to make sure that we had the right balance of node negative, node positive patients so for the trial and we remain confident that that balance is right and now we have to wait for results.

Next question is from Jo Walton, Credit Suisse. Please go ahead.

Thank you. Just a couple of clarifications, going back to Sachin's original question, I wonder whether you feel there is an early market entry strategy via a compendia listing either for Merck with the KEYTRUDA data that we saw at ESMO or for yourselves with the early data that you have, so it's just a question as to see when some of this data could come through. And secondly just on Avastin situation in the U.S. we've seen two quarters of stable sales then it was down 9% in the quarter -- third quarter though, clear in dollars, clearly you get quarterly volatility, so can I just be sure that all of that decline is to do with these one-off 340B rebate adjustments and then we're not already seeing an actual decline and if we are seeing it, it presumably it is in the colorectal market where patients are just using IO instead of Avastin, so just to get a sense of how quickly we're going to be seeing a decline there? And the final question sorry on the biospecific which you have said failed in the study directly comparing it with Avastin, wonder if you could tell us just a little bit more about why you think that might be and whether that's going to change your ability to do more studies going forward because you can't show that it's -- there is a reasonable assumption that this can be better than Avastin, will it just impact your ability to start more studies?

Yes, thanks Jo. Let me try to provide some clarity and please come back in it, I will try to provide as much clarity as I can. So first of all on the Compendia listing and what we had anticipated, I mean I just I don't think I can guess on that, I mean I guess what I would say is that we're seeing now Compendia listing for Merck in the frontline cancer setting but that was based upon Phase III data in the monotherapy setting in the frontline cancer setting and likely their approval will come, I would expect not too far after the Compendia listing. So it's just too challenging for me to try to guess on what might happen in the frontline chemo combo setting, I think NCC and Compendia's are data driven and they want to see very compelling data but it's up to them to determine what they see as very compelling data. I don't think -- I think we're in a very good position in the frontline lung cancer setting our Phase III trial is recruiting our Phase III program is recruiting well. And I'm comfortable that we will be reading out in a very competitive timeline in the frontline setting with the Phase III trial. So on the Avastin in the U.S., I mean yes it's two basically two different factors it is some of the increase in 340B the lumpiness in the 340B but it is some competitive pressure in but it is we don't see it in colorectal, I think we predominantly see it in second line lung, where you might expect it to be challenge because of the immunotherapy entrants. So that's basically what we see. Now over time, I think as immunotherapies show success in other aspects of Avastin business, we may see more of that but I don't see -- I don't see a drastic competitive change right now for Avastin but some of that pressure that we've seen in quarter four is also due to some of I would call the smaller indications of Avastin starting to get some competition in. So that's the additional color I would add there. And then finally on the biospecific, yes, I just want to be clear on that, I mean first of all we knew the bar was high there, I think we flag that to you before that as much as we like to be optimistic, we tried many, many times to improve a part of Avastin. I think Avastin has achieved a certain sweet spot in terms of its ability to reduce the blood flow through a tumor to allow the tumor to get start but have enough blood flow to allow biologics in chemotherapies to hit the tumor. And I think trying to replicate that sweet spot, our scientists would say it's been very difficult. So even though we had a stab at it, one more time with the biospecific antibody, our expectations were not very high there, we set the bar high, and it failed. I just want to be here that we're not suggesting that we will be beginning new studies with this medicine at this stage, there is a couple of combination trials that are going on right now that we will carry through to read-out, you will see in the case study, yes, I don't know in the future meeting we don't know yet exactly where it will be but when you see it, I mean I think you will also see that, we don't intend to now build a large program on that backlog but we would like to see the remaining data that has already started and to see how that reads-out as well. So I hope that gives -- I mean I hope that puts that into the proper context in terms of expectation setting.

Next question is from David Evans, Kepler Cheuvreux. Please go ahead.

Thanks for taking my question. Yes so with respect to your PD-L1 chemo combination studies in first line lung, can I just check on timings and endpoints, that is only the EMPOWER 150 study which you will be reading out in 2017 and all of the others will be at 2018 PFS read-out, it also seems to me that all of the studies have been converted to jowl PFS and OS endpoint. And just so just to get your latest views on is it still feasible to file these combinations purely based on PFS data or do you expect do you will have to wait for OS data in order to file any or all of those? Thanks. Daniel O'Day: Thanks David for the follow-up question. So yes just to clarify we have got seven trials ongoing right now in first line or adjuvant setting. We expect one to read-out in 2017 that's the EMPOWER 150 we expect the remaining ones to read-out in the first line setting in 2018 at different times and the adjuvant trial of course would be after 2018. So those are the timelines. They all have dual endpoints core primary endpoints, PFS and OS which means the trial is positive on a positive read-out on either one. We continue to believe that OS is the strongest endpoint, the most reliable endpoint that's why we are so encouraged by the durability and response that we're getting in things like the OAK trial that length and tail of those curves and we're also -- I mean on a separate point, we are also excited to see the separation of three months which shows a strong immunologic effect. So bottom-line is that they all have those co-primary endpoints, we expect obviously the PFS would read-out first with OS following and because of the trial design yes, the answer is yes, we could file on a positive PFS result. And so we will see how that plays out. But again we think probably the true measure of benefit from patients is eventually OS and that's what we're continuing to target as the true evidence, the differentiator for our products versus other products as well in the whole immunotherapy space.

Next question is from Jeffrey Holford from Jefferies. Please go ahead.

Hi, thanks first for taking my questions. So first one is just following all information we gathered at ESMO particularly checkmate 26 results, does any of this impact your thought around your clinical trial design or recruitment plans and in particular have you become more concerned potentially around the quality of patients being recruited into studies, just wondering if we should be focusing more closely on patients who prior your therapy potentially that might have been a factor in some of the adverse outcomes of some of those studies compared to what was expected. And then the second part of my questions are really focused around chemo IO, had you encouraged by chemo 21G result. But I was just wondering in the context of the lack of complete responses treatment trial, is that somewhat disappointing, do you have any experiences of complete responses in any of your Tecentriq chemo trials in lung cancer? Thank you. Daniel O'Day: Thanks, Jeff. So let me first just on the ESMO, but I think what I would say is in this space we are constantly being educated by our own data and by other data and we take that into account in terms of our trial designs, we also create in general flexible trial designs, I think you saw a perfect example of that with the Birch, Popular OAK setup, if you like where we were able to get important read-outs from those two trials to be able to -- to be able to make adaption to some of the decisional analysis plan, you've also seen us adjust the endpoints in our first line lung cancer trials from to our core primary endpoints, so we will be continue to be informed. I don't think there is anything in particular per se that was that came out of the particular data we saw at ESMO that is having us rush to do, changes to our trial designs in anyway and we're not seeing any adjustments to our recruitment phase, I mean that continues. So I think there was some really important data at ESMO, some of it was interpretable and some of it I think the entire community is still trying to interpret, let's put it that way and so I think we need more data to understand some of the results. Having said that when we look at our frontline lung studies and we look at our diagnostically enriched population, how we set that trial with monotherapy, and we look at our chemo combo, so far we are not seeing anything that we saw at ESMO that would have -- we would have less confidence in our trial protocol, let me just put it that way. Now you asked about the whether we are seeing complete resources in our Phase I chemo combo study. We have seen complete responses in our stage 1 chemo combo study in fact that was presented at the Indian National Association, the study of Lung cancer that was in February of 2015 you saw some of that data accordingly. I don't really want to speak on the competitor data because I think that's for you to determine relative to PFS and OS and what you saw in that trial. But I think we've shown you a decent amount of data on our Phase I trial. We will show you the data again when it matures on our last. We continue to be encouraged by what we see in the extension of our Phase I trial and continue to be encouraged that is if you like a proof of concept for our frontline Phase III trials.

So just as a follow-up, do you have some idea of the timeline to see the OS data of the extension of the Phase I? Daniel O'Day: Not exactly because it's event driven, so I think we're hoping that it becomes let's say in the first half of next year, there is two things that have happened. First is the data has to read-out become positive and then we have to present it at the meeting. And so the question is you would you can get the meeting that would be appropriate to present a lung trial as it there are several of course that's a good news now throughout the year, there are several, you don't have to wait for ASCO, you've got a variety of different opportunities. But those two things have to happen first you have to read it out and then we have to see it. So I can't make a better guess than that at this stage.

Next question is from Tim Anderson, Bernstein. Please go ahead.

Thank you. Few questions. Just M&A can you confirm that bolt-ons are really primarily what are you continuing to look at. Avastin I'm wondering what biosimilar mitigation strategy is from here are we likely to expect all advanced sub queue formulations at some point like you have done with your other products. And then a question on all kind of space, it's surprised me that you guys have been entertaining advancing two different models through Phase III, I'm wondering if the upcoming read-out of [indiscernible] and all gauging factor for you in thinking about the development of your two drugs and what we might know that down to one drug, appreciating that all the monoclonals have slightly different mechanism of action?

Thank you. On the M&A thing, there is actually no change at all; we continue to focus on bolt-on acquisitions to complement our portfolio with products and technologies. So no change there. Daniel O'Day: Yes, Tim, thanks for the question. So I think nothing has changed with our Avastin strategy in terms of how to navigate the biosimilar landscape. I mean as I said I think vanucizumab was one of many products that have established superiority to Avastin. But having said that, our mitigation strategy around Avastin although different from what we are doing necessarily with respect to MabThera where we are looking at more one-to-one replacement strategy. Despite our advancement if you like to Avastin is really going to be based upon the combination strategy and the immunotherapy strategy across the seven different disease stage of where it lies today. And some of those are going to be still in combination with Avastin and some may be without Avastin combination, there is no sub queue strategy for Avastin and it's not something that we are pursuing at this stage. But I just remind you, we have got now in immunotherapy we have got 50 clinical trials going on right now, 40 of those are combination trials and we have a very significant program behind colon, lung, renal cell carcinoma all of which are areas where Avastin is today also in breast, so this is really our overall approach is to if you like replace Avastin with immunotherapy, different combinations with or without Avastin as we move forward. Oh, Alzheimer's thank you, yes so just a couple of updates on the Alzheimer's program, I mean we continue to recruit in our first [indiscernible] trial we will be moving forward with the second trial and we are still in dose escalation for [indiscernible]. So this is as you can imagine a very significant sized market, if somebody shows efficacy in the Alzheimer's and we have to tell you that, we have two very different medicines for Ocrelizumab, the humanized monoclonal antibody that is for all forms of beta but particular preference for things, I mean it is a different mechanism of action in Gazyva, Gazyva is fully humanized monoclonal antibody that's really focuses on the removal of beta Amyloid plaques. There are precedents out there in the environment, some of those will read-out and could help us and inform us obviously we expect them to read-out at towards the end of this year. We will continue to take in data we are looking at ourselves in the previous trial on obinutuzumab and tocilizumab as well as competitive data and ongoing data from our trial to make the very best decision. I have to remind you, we have got obinutuzumab that is IV and tocilizumab is sub queue, all these factors are taken into account in terms of how we will approach for Alzheimer's strategy. And I think we have two very good medicines to look at in the months and years to come to be able to determine if we can make a big impact on this huge unmet medical needs. So we will be updating you more, I think as we get more information ourselves end of this year into next year on our strategy but we continue to believe having two of these medicines is to our advantage.

Next question is from Louis Hector [ph] from Exxon. Please go ahead.

Hello. Thank you for taking my questions. I have a few but hope they are very quick, so can you quantify the accurate impact of this 340B during Q3. Can you remind us what percentage of Avastin sales are in lung cancer in your first and second line combined. Can we rule out at this stage an FDA panel for Ocrevus and was that any stocking in present Q3 number, thank you. Daniel O'Day: Great, like this is like lightening around now, so I can't quantify more of the 340B impact for you, I mean I think it's I am giving you as much as I can give you on that relative to 340B and some of the competitive entrants and it's we will I'm not still confident that we will moderate that effect in the fourth quarter. In terms of the relevant percent of Avastin in lung cancer, it's around 15% for the entirety of lung cancer with obviously a smaller proportion in the unlabeled indication of second line lung cancer. There is no indication with Ocrevus that we would have a panel at this stage it's not in discussion or dialogue with the agency and I can't rule out the stocking for Tecentriq that is really demand driven that you see the 76 million Swiss Francs through the first three quarters.

The last question is from Stefan Schneider from Vontobel. Please go ahead.

Lot of thanks for taking the question, I come again with something that is being asked probably already two times at the read-out in first line lung to the EMPOWER 131, 130 and 1011. Having moved to 2018 from 2017, I don't know whether you already elaborated on why that is, is it because you show overall survival of no longer PFS or what is the reason for that and then follow to that is the Avastin combination trial there is that going to be OS data next year in the second half? Daniel O'Day: So Stefan no change to the lung program, I mean when we started the Phase III lung program we said read-out will be in 2017 and 2018 and it's very consistent with that, so timeline wise we haven't had any change and sorry the second question was, second question Stefan sorry.

The overall -- do we expect overall survival data already for the one in second half I should say. Daniel O’Day: Yes no I think we would expect PFS in the second half of next year with specific sample with Avastin and yes that's right, not OS.

There are no more questions at this time.

So thank you everybody for joining our briefing today. And thanks for your interesting on Roche, have a good day. Bye, bye.

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