Severin Schwan - Chief Executive Officer and Director Daniel O'Day - Chief Executive Officer of Roche Molecular Diagnostics and President of Roche Molecular Diagnostics Roland Diggelmann - Chief Operating Officer of Roche Diagnostics Alan Hippe - Chief Financial and IT Officer

Sachin Jain - BofA Merrill Lynch, Research Division Andrew S. Baum - Citigroup Inc, Research Division Richard Vosser - JP Morgan Chase & Co, Research Division Alexandra Hauber - UBS Investment Bank, Research Division Luisa Hector - Crédit Suisse AG, Research Division Keyur Parekh - Goldman Sachs Group Inc., Research Division

Good afternoon, ladies and gentlemen. Welcome to our year-end briefing. 2013 has been a very good year for Roche, both in terms of financial figures but also very much in terms of our pipeline progress. So let's get right into the 2013 performance. We achieved all our financial targets: sales up by 6% in local currencies; core earnings up by 10%; and the Board will propose a dividend increase of 6% to the general assembly, raising the dividend to CHF 7.80. We've really had a fantastic year in terms of pipeline progress. To start off, in the HER2 franchise, we launched Kadcyla, Perjeta, neoadjuvant over Herceptin subcu [subcutaneous]. We got breakthrough designation for Gazyva in the hematology franchise, which we launched at the end of last year already in the U.S. And very importantly, today, we have 15 new molecular entities in our late-stage pipeline. We also had a number of important product introductions in diagnostics. For example, the cobas 8100, which is a new platform which brings the automation in the lab to a completely new level. Growth was very much driven by oncology, both the established products and the new medicines we have brought to the market. But also outside of oncology, with Actemra in rheumatoid arthritis, that's grown by 30%, has now surpassed the CHF 1 billion mark. And we have also seen very good growth with Lucentis in ophthalmology. Pharmaceutical sales up by 7%; diagnostics up by 4%. So on a group level, 6% in constant currencies. What you can see is that the growth in Swiss francs is only half of that, with 3% growth. And about half of that is due to the strengthening of the Swiss franc versus the Japanese yen, and the rest, due to the strengthening across other major currencies such as the U.S. dollar. We continue with the good dynamic on the sales growth. You can see 6% to 7% in the second half of 2014 -- sorry, '13. And this good performance on the top line also translates to the bottom line, a record operating profit, with CHF 17.9 billion and a margin of 38%. And that is also reflected in the cash flow, which now stands at over CHF 16 billion or 35% of sales. Based on those positive results, the Board will propose to increase the dividend again. This is the 27th time in a row and represents a payout ratio of 55%. We really had a tremendous development in terms of our late-stage pipeline, and don't forget, a number of products have now moved out of those pipeline, because we just launched them, like Kadcyla, Perjeta, Gazyva, and still, we have 15 new molecular entities in late-stage. Many of them in oncology, good development there. But you can also see an emerging portfolio here in immunology and ophthalmology, with important compounds such as lebrikizumab in asthma or lampalizumab, that we presented stunning Phase II data actually for geographic atrophy in ophthalmology last year. Now let me just give you a bit of flavor in terms of the outlook for 2014. So on the one hand, we have headwinds, in particular, mainly with Valcyte and Xeloda, which are going off-patent in 2014, as you know. We also divest our Neupogen portfolio to Amgen, and we had some special one-off effects with 340B rebates and the past service income. This is overcompensated as we go into 2014, above all of course by the strong growth of our established brands and very much so by the new medicines we are launching outside of the U.S. now. And we should expect some additional leverage from lower interest expenses and also product disposals, in particular, the sale of the Neupogen franchise to Amgen. I expect that the income from product disposals will roughly be in the magnitude of the one-off effects we had in 2013 for the 340B rebates and the past service income. So based on that, we expect group sales to grow in the low to mid-single digits. We expect core earnings again to outgrow sales growth. And based on these results, we expect to increase dividend again. And with this, I'd like to hand over to Dan. Daniel O'Day: So good afternoon from my side, everybody. It really has been a terrific year for the portfolio this year. I think the numbers, and I'll get into some of the detail, but just before I get into it, there's been significant, I would call, medical practice-changing type products that we entered into the market here this year. Just to remind you, I mean, first of all, with Perjeta and neoadjuvant, really paving the way for potentially a new regulatory path for products with a significant PCR effect in patients, up to 40% of patients treated before surgery really had no evidence of tumors by the time they hit surgery. So when you think about the period of time it took for Herceptin to go from metastatic to the first use in adjuvant setting, that was close to 8 or 9 years. And this was 1.5 years, so pretty phenomenal. Secondly, the Gazyva head-to-head trial -- I have a slide on that, but going head-to-head with gold standard of care and showing a really significant difference in CLL. And thirdly, I mean, this is really the year when the antibody drug conjugates were fully launched into the marketplace in HER2-positive breast cancer with Kadcyla. So there's a lot of rich news here, but I just want to say that those 3, I think, really will fundamentally shift the way that cancer is treated in those 3 different areas, and plenty more to come. So just to review the 2013 results in a bit more detail, from the Pharma division side, we had a 7% sales growth overall in constant currency exchange. I would call it a really outstanding growth in the United States, with 10%, driven by, of course, the oncology portfolio -- the established oncology portfolio, the newly launched oncology portfolio with Perjeta and Kadcyla, and also driven by the ophthalmology, immunology portfolio, Actemra as well as Lucentis. Europe, with a 2% growth, significantly ahead of the market. In austerity-strapped Europe, to see a 2% growth is really a very good performance and I think shows the value of innovative products. In Germany alone, we had an 8.5% growth last year, obviously offset by some other European markets. But 2% in Europe, really strong performance. Japan, at 2%, that also needs to be put into perspective because Evista, which was a product that Chugai had, was given back this year. And also there were some price -- management price declines in Japan. If you take those 2 things out, actually, Japan would be growing at 8%; really strong growth in Avastin and Actemra in Japan. And then finally, the international region growing at 8%. If you look at the E7 markets, they are growing at 12%. Very strong growth in China overall, 21% growth; in Brazil, 9%. Also demonstrates, I think, again, sustainability of the strategy of taking highly innovative products into the emerging markets with innovative access programs to penetrate both the private and public sector. On the P&L, the core operating profit grew in line with sales. I want to explain a couple of these line items. You see most of the aspects of the P&L are growing less than sales. Cost of sales, obviously, we had a robust volume growth on the pipeline. But in addition, we're investing now in our -- particularly our biologic manufacturing network to be able to prepare for the future of all the new biologics that are coming out and meeting the increased volume demands there. Let me talk a little bit about the royalty and other operating income, as well as the R&D. The operating income was really -- had the biggest effect from the disposables of product that occurred in the second half of 2012 that did not repeat in 2013. So you see here Ostac, Rocaltrol, Vesanoid, Rohypnol were divested in 2012, and that really offsets about CHF 130 million of the pharma income from disposals and other royalty income. I would just point out that it's public knowledge now that we have a deal with Amgen on Neupogen and Neulasta to return that product to them, and that obviously will mean, in 2014, we'll see some additional income in this line moving forward. On research and development, it was a 5% growth. I'm going to talk about the pipeline. We had significant evolution from Phase II into Phase III last year. And it really splits out in 3 major categories, somewhat equally, actually. The first one is the high number of projects that came into the late-stage portfolio that required investment to either start Phase III or prepare for Phase III. Secondly, the increased technical developments. The more we become creative with our biologics, particularly when we look at armed antibodies or bispecific antibodies, the technical development to produce those products is actually quite sophisticated and it takes significant resources. So, I think on the one hand, I mean, it's good news because we have new products coming into the pipeline and, of course, the technique and the know-how that goes into that also allows those products to be quite robust for the long term. And then the third major bucket is really the investments in Phase IV trials, with some of the accelerated approvals we've had and the faster review times. From a regulatory perspective, that puts some demands on our Phase IV spend. And in addition to that, compliance standards are increasing everywhere in the world in terms of pharmacovigilance, safety reporting, and we need to increase resources to meet those demands as well. So those are the 3 major impacts on R&D in 2013. Really strong sales, you can see here, across the region for all the products across our line. Obviously, the 3 major oncology products continue to produce robust growth: Avastin, MabThera and Herceptin. Perjeta and Kadcyla together had sales of about CHF 0.5 billion last year. And of course, Kadcyla, just recently approved in Europe. Actemra met the CHF 1 billion mark last year, growing well, subcutaneous launches moving on, good data in monotherapy. And Lucentis, arguably, you can call me on this one, it did much better than we expected. I'll explain a little bit about that, but at 15% growth, that was really a strong performance. And I'll give you a little bit of an insight into 2014 as well. Now, obviously, I should say that MabThera also hit CHF 7 billion in sales last year. You look that it offset, obviously, the products that are declining in our portfolio. Evista, this is what I talked about in Japan, Neorecormon and Epogin, and Pegasys. I'll just warn a little bit, it's the only slide I'll talk about Pegasys. We'll be on a little bit of a wild ride again this year, with the second-generation orals coming. There is, as you know, the cycle between next therapies and hepatitis and warehousing. And right now, we're seeing the launch of the second-generation orals in the United States, which will re-instigate, if you like, Pegasys at least for the -- expect the first half in the United States next year. And then in the second half, we would expect warehousing of patients to begin again in anticipation of the interferon-free regimens that are to come after that. So stay tuned on Pegasys, it's going to be interesting. For 2013, continued duration of use in MabThera and Rituxan and also penetration into many new markets, and emerging markets is growing that line. The HER2 franchise, growing at 14%. Xeloda, a very important point as we look at next year, Xeloda is -- has lost patent in Europe as of the fourth quarter. It will lose patent in the United States. There are several generic entries ready to go. This will occur in February and March of this year. And we also expect Valcyte, which is already off-patent, to have generic entrant during the course of the year. All in all, we would expect that we would have about CHF 1 billion loss in patent expiries next year, that will be more than offset by the growing oncology and x oncology franchise as well. And Zelboraf, fully penetrated in the United States, will -- 2/3 of Zelboraf sales today, by the way, are in Europe. And so we think that will obviously have some evolution as we go into next year, also with the launch of the MEK inhibitor. And we look forward to our data on the MEK inhibitor in the middle of the year. As we look at Avastin, you can see the growth here continuing. It continues both in colorectal cancer, with treatment beyond multiple lines, and ovarian cancer. As you'll see in a later slide, we're quite encouraged by the fact that we expect Avastin to be filed for its 7th now indication globally, and that will be cervical cancer. And I'll give you a bit of an update on that as we continue. So on the HER2 franchise, you can see here that we had a lot of advancements in 2013. We had, of course, several approvals: Perjeta in Europe; Kadcyla in Europe and the United States; Perjeta for neoadjuvant in the United States; and also Herceptin subcu. We saw a pretty significant uptick in Perjeta in the fourth quarter of last year in the United States after the launch of the neoadjuvant indication. And that was driven by -- predominantly by increase in the metastatic setting, in the first-line metastatic setting. So what we see now at the end of the year in terms of new patient starts in the United States is around a 50% share for Perjeta in the first-line setting. Kadcyla continues to do well in second line and beyond as well. And we see a share there of around 40%, actually. Herceptin subcu, where it's been launched, Germany, the U.K. and others, is actually doing very well. What we find is when a center adopts Herceptin subcu, they tend to convert all the patients in the clinic because they change their workflow. So they tend to convert not just adjuvant but also metastatic patients in their Herceptin subcu. And we think, particularly -- of course, this is only approved outside the United States, this is going to be a very convenient strategy for patients, for healthcare systems and continue to support our franchise in relation to biosimilar competition that will come. So for Lucentis, as I said before, a stronger year than we had anticipated at the beginning of this year. That's predominantly due to a number of factors. Our share in AMD has stabilized. The growth in the AMD market continues. So the market is increasing both in terms of number of patients, but also in terms of conversion from Avastin. We're seeing conversion from Avastin that is really moving to Eylea and to Lucentis. And clearly, we're getting very good share in that area. Also the new dosing regimen that we had in AMD is helping in supporting this year as well. In DME, we continue to gain share and do well there. I would just remind you that in the second half of the year, we expect to have competition from Eylea in this DME setting, just to raise that to your attention for the course of the year. So overall, we see, clearly, growth for Lucentis in 2014 with the following dynamics. In terms of the emerging markets, it has been a bit of a lumpy ride. And you can see now in the fourth quarter that, that lumpiness has kind of smoothed out a little bit, particularly with Russia growing now at 34%, big tenders coming in, in the fourth quarter. And you can see across the board, even in China in the fourth quarter, we had a 17% growth. So overall, across the emerging markets, we had a 12% growth and 14% growth in the second half of the year here. So it speaks, I think, to the continued value of our products in these markets and how they're doing outside the United States and Europe. Good. So that's on the results. I want to turn the attention to the innovation as well. Since we last spoke with each other, we basically had 4 additional pieces of news and information in the regulatory front. We received the Actemra subcu approval in the United States. We received the Gazyva approval in the United States. We received the Kadcyla approval in Europe. And we also, finally, had the final readout, not unexpected, of the Tarceva trial in adjuvant, which was not positive. And of course, at this stage in the game, we also expected that as well. But it leads us to the full reading of milestones and approvals for this year. I had the opportunity, actually, to be at ASH this year to see the data presented in the plenary session, and it was obviously very well received by the CLL community and the leukemia community in general. And what you see here is, of course, the Kaplan-Meier curve on progression-free survival. And it's pretty astonishing that after we saw the hazard ratio of 0.14 with the comparison of Gazyva GA101 to chlorambucil, you see on a head-to-head basis now with MabThera, a 1-year -- close to a 1-year difference in progression-free survival. Big impact on patients and certainly encouraging for us as we continue to develop Gazyva in the other indications, in Non-Hodgkin's lymphoma. On the Bcl-2 side, also been some good progression there. You know we have the Phase III study going on right now with relapsed or refractory in MabThera. In addition, the Phase II CLL faster market strategy in the 17p deletion, we expect data end of this year, early next year, with that trial. We will be beginning a trial with Gazyva, Phase III in combination with Gazyva, and we expect to start that trial in CLL by the end of the year. And I'm pleased to report that on the tumor lysis syndrome, we feel fairly confident. We've been able to handle this situation. We are now in dialogue with regulatory authorities, and we expect to have the tumor lysis syndrome agreed and resolved with the regulatory authorities by the middle of the year. Gazyva, clearly with this type of efficacy, has been looked at now by -- in the leukemia lymphoma community, as potentially the new CD20 backbone, certainly in CLL and potentially in other disease areas. And with that, of course, we are pursuing, as you know, combination medicines within our own research and development pipeline. But also, we've started to look externally, and there is a number of small molecules out there that are using Gazyva as a backbone out there through investigator-initiated trials or collaborations with other companies as well. So I believe that this concept of replacing the backbone, adding on to that backbone to increase survival and efficacy in lymphoma is really getting some good ground. So immunotherapy, I think this is something that you've probably heard about. We're very excited about it. I mean, it's an incredibly exciting area. It's an incredibly complex area. I'm happy to say that, before I get into any of the science, I think that one of the things I am very impressed at within Roche and Genentech is we have some of the very brightest minds on this, some of the very best experts in the field of oncology and immunology that are working with us to design, if you like, the next generation of cancer therapy. And what you see here is, of course, what we've disclosed before in non-small cell lung cancer and renal cancer. I'm happy to say that the OAK trial, the Phase III in non-small cell lung cancer, we expect to start enrolling relatively soon this year. And you see the data expectations on the other trials that we have going in this area. In addition, we have disclosed a variety of the ongoing combination studies. And we have others that we haven't yet disclosed but will be disclosing over the course of this year. We're really looking very comprehensively, both within our own portfolio and within the broader cancer portfolio, on what types of combinations we think can best be engendered by working with PDL1. And of course, we will keep you informed. It's a highly competitive field, as you can imagine. And therefore, the release of data, we will moderate with that level of competition. But I would say, in terms of 2004 (sic) [2014], amongst other potential information that we'll be providing on immunotherapy, we will be providing data on a new cancer beyond the 3 that have been spoken about before in the first half of this year. And as soon as we have more information on that, we will let you know. But we are seeing some encouraging data in another tumor type with PDL1 which is something that we felt the PDL1 target has the potential to produce, which is more pan-tumor type activity. The additional thing that I would say is we do have other additional combinations that we are working on right now, that we will update you on, including immune doublet combinations, both within the Roche Group and also partnering outside the Roche Group. In addition to the pretty extensive cadre of compounds that we have within Roche, there's a couple that have been disclosed that we've licensed-in from the outside or acquired-in from the outside. And I just wanted to quickly take you on a little bit of a walk through these 3. When you look at the cancer immune cycle, you have agents that both stimulate the immune system and inhibit the immune system. As you know, PDL1 is an inhibitor. And what anti-PDL1 is, is it removes that inhibition so the T-cell can work on the cancer cell. These are basically different types of stimulus that we brought in from the outside. And as you know, oversimplifying, you have 3 major aspects to the tumor cell cycle process. And that is that you have the tumor cell itself, you have the T-cell and you have these antigen presenting cells as well. And just starting out with anti-CD40, what anti-CD40 is, is something that instigates antigen presenting cells to instigate tumor cells to kill cancer cells. We're quite excited about that program. If you look at the cancer vaccines, they also act directly on the antigen presenting cells to stimulate activity towards tumor cells. And we have 2 there, a DNA vaccine and a peptide vaccine, that we've disclosed so far. And then the last one is this Immunocore collaboration on ImmTACs. And this is a pretty interesting technology because this has specific antigens that target different tumor cells on one end of this molecule, and on the other end, they attract and connect T-cells directly to the cancer cells. So if I confused you a little bit, it's -- welcome to the field of cancer immunology. I guess, the real message here is that this is an exciting area, but it's a very complex area. And I think the companies that have the best minds working on this are the ones who are going to find the best combinations to be able to find the way forward in cancer immunotherapy. Very exciting area, and it's one that we are fully dedicated to and are investing quite a bit in, in the years to come. But of course, cancer isn't everything at Roche. And we have had a very attractive year in our immunology and ophthalmology portfolio, more than CHF 6 billion in sales last year, growing at 12%. So already, we have a very attractive and durable pipeline on the market today. And we added last year 3 new compounds to lebrikizumab. So lampalizumab, as you know, for geographic atrophy. And we're in the process of discussing with the regulatory authorities now the trial design and preparing that for the future. Etrolizumab, as you know, both lampalizumab and etrolizumab have PHC or companion diagnostics hypothesis to be able to target and increase efficacy in sub-patient populations. And then finally, oral octreotide, which is a collaboration that we have with Chiasma for acromegaly, and we'll be getting results from that this year, with the potential to file next year, depending on those results in acromegaly. Just so you know, this is an oral medication and the current standard of care is infusion by a rather large bore needle. So it could be a significant advance for patients. Lebrikizumab, moving well with the clinical trials there. There will be some additional Phase II data presented at the AAAAO [ph] meeting in San Diego in March as well, so you will have some more data on lebrikizumab. And then we have a variety of compounds, but the M1 prime that we call quilizumab for allergic asthma will be looked at in Phase II for this year for potential movement into the late-stage portfolio. So an exciting portfolio in immunology, ophthalmology. Really, a big change if you look back a year ago in terms of the pipeline and driving innovation and development here. So on the outlook side, pretty exciting. We have more than 10 compounds now in CNS. Obviously, you heard about the first 2 studies in negative symptoms, bitopertin, which were not successful. This was a huge unmet medical need, but as you know, quite a high-risk program. And yet, we still see that there is a significant unmet medical need and good scientific hypothesis that we need to pursue here. We still have the positive symptoms or suboptimal symptom trials for bitopertin that we will need to read out as well. Ocrelizumab, moving well, and gantenerumab, as you know, we started the second Phase III in mild patients at the end of last year in addition to the ongoing trial with prodromal. Immunology, ophthalmology, I've covered. PDL1 and Bcl-2, I've covered. Just the ALK-inhibitor is now in a Phase II, with crizotinib -- you know what I'm talking about, that Pfizer drug. I have a blank on this. Why is it that we make these drugs -- products so difficult to pronounce. Even ours, we can't pronounce. In any case, it's in those failures right now, and we'll be working -- the interesting thing about our ALK-inhibitor is it seems to have an effect on brain metastasis, which obviously that other compound doesn't. So we're interested in that, and we may be looking at that as Phase III as well. I got to get that one down. The other exciting news on the oncology portfolio is 2 PI3 kinase coming into the portfolio in December. Now you may ask why 2? One is actually focused on a mutant PI3 kinase, and the other one is more focused on wild type. And they have potential different applications in different types of cancers. The mutant, most likely specific to the hormone-positive breast cancer, whereas the wild type or pan-PI3 kinase could be functional in multiple tumors as well. So early data on that, but exciting data on that, such that we wanted to move it right into late-stage. It's obviously one of the most mutated forms of mutations in cancer, and it's something that we feel, not only on its own but in combination with other products, could make a big impact in a variety of tumor types. And then anti-CD79b for leukemia, this is one that we will look to take also into the clinic in a later stage. Looking at -- potentially looking at chemotherapy sparing or reducing in the regimens that are associated with leukemia. So we're pretty excited about this portfolio, 15 compounds. We have a lot to do in 2014. But a big and talented organization that's excited to attack these compounds on their own and in combination. So in terms of what we expect for the late-stage news flow this year, on the regulatory side, we still expect to hear on Actemra subcutaneous in the EU as well as MabThera subcutaneous in the EU. And then as you see here, we're expecting to file Avastin for cervical cancer in the EU and the U.S. this year. And on the readout front, just to remind you, we've got MARIANNE reading out later this year with Perjeta and Kadcyla in the frontline setting, which we're all looking forward to. We've got the MetMAb inhibitor in non-small cell lung cancer, which we expect to read out this year. And we have our MEK inhibitor in combination with Zelboraf in melanoma that we expect to read out this year. So I would say those are the several readouts that we expect to be -- I wanted to highlight here. And then we've listed just some of the Phase II programs we have, the ones in oncology, I've spoken about with the exception of anti-HER3 EGFR, which we expect some information on this year. And then we also have some in Alzheimer’s, neuroscience and also the asthma program. So with that, plenty to do. Thanks for your attention. And I'll turn it over to Roland for the Diagnostics.

All right. Thank you, Dan. Thank you, and good afternoon, everyone. It's a pleasure to be here. And it's a pleasure to report on the Diagnostics results. Could I maybe add the slides, Dan, here as well, if possible, or I'll go through here as well. Anyways, Severin already mentioned a good year in Diagnostics as well, 4% growth overall, which puts us above the market growth. And you can see here, led by a very strong franchise in Professional Diagnostics, growing 8%. Molecular Diagnostics, the underlying growth, if we exclude for the Sequencing performance, which is largely playing in a niche only, is a very healthy 6% growth. And then 7% for Tissue Diagnostics. So together, the Lab Diagnostics franchise is growing 7%, which is a very good result. And of course, you can see here, an ongoing challenging environment in diabetes care, minus 3%. Here, too, we see some encouraging signs in that Latin America, Asia-Pacific and EMEA actually contributed to positive growth. The negative performance coming from the U.S., where we had, as you all know, massive reimbursement cuts on strips, 72% mandated by the CMS on the public sector. So this is what affected the result, minus 15% in the U.S. Other than that, growth in diabetes also in the rest of the regions. So with that, turning over to the regions and the breakdown. And you can see here good growth, very strong growth in emerging markets in the E7. 14% in Asia Pacific here, led by China with a 27% growth. Very good growth as well in Latin America. And I'd like to point out the positive performance in a flat market in EMEA, 2% from a market-leading position, capturing market share in the European setting. And then also in the U.S., 1% here including diabetes care, excluding 4% growth again here, the Professional Diagnostics franchise allowing us to drive ahead of the market. On the P&L, a couple of comments. The 4% growth translating in the 4% operating profit growth. You can see at the different line items here. The COGS was over proportional gross, which is driven by a couple of factors, will be on one hand, a price decrease that we see that averages out at about 3%. But on the other hand, also positive in that we were able to continue to place instruments in the market, where we'll actually fuel future growth, and we had a higher proportion of cash sales of instruments as opposed to amortizing over 5 years, which has an impact on the COGS. You can see as well, the R&D line, investing in new platforms, which is across the different segments, and also in addition into our new franchise in hematology. G&A, one-time affect or the initialization of the medical device tax, first time it hits our P&L, and then also internal programs in advancing IT and other investments. That all compensated by cautious spending on the M&D line, also an effect of the restructuring that we undertook in diabetes, but also in integrating our Applied Science business in Professional and into Molecular. So this is just a rundown on the P&L. We had a good year as well on the pipeline. On the launch, 11 out of 13 products launched in time. The 2 missing will be launched in the first quarter of this year and the second quarter this year. So overall, good launch. Accuracy, and I think this also speaks very much to us continuing to pursue our strategy. You can see on the top, the instruments the platforms, which is really looking at driving testing efficiency in the laboratories. And here the most important one, Severin mentioned it earlier, is the cobas 8100, the fully automated workflow system. So testing efficiency on one hand, and then high medical value tests on the bottom end, which actually fuel the largest menu through this instrument base. And with that, a couple of highlights for the fourth quarter. I'll touch separately on Professional Diagnostics. On Molecular, what you can see is a very good performance on HPV, 90% growth in the HPV franchise. And also what you can see is that we start to complement this with other tests. Our cervical cancer testing is the most comprehensive. We have HPV for primary screening, which is a molecular test. We have the CINtec PLUS, which you can see at the lowest line, which is a histology testing largely for triage. And then we have a confirmation test, which is again a molecular test, which is the CINtec test. So a comprehensive suite of tests that we offer in cervical cancer and certainly, the most comprehensive. On diabetes, we see good uptake of differentiated products, which are aimed at free testers, which are aimed at people who like to test more in a more structured way. So the Accu Chek Mobile here is one of these examples, growing nicely at 40%. And finally, on tissue on the advanced staining, low growth in the U.S., which was expected; a reimbursement reduction in the U.S. and then also the CAP guidelines that actually make some of the negative testing obsolete. So that it has a negative impact on the number of tissue tests performed. So this was a base effect and we expect this year to be faster growth in tissue. And we also had a very, very strong growth, 14% in EMEA, and 34% in Asia Pacific, expanding our customer base in histopathology. I'd like to touch on a couple of highlights. And as we move beyond the instruments and as we focus more on workflow and rendering laboratories more efficient, of course, laboratory IT becomes very, very critical. It's a means of integrating and connecting the instruments. This new system is called Infinity. It can connect individual instruments in a laboratory setting, from pre-analytics to analytics to post-analytics. It can also connect third-party products and it's modular in its nature, so it can also connect to a laboratory or a hospital information system. So with that, we're trying to consolidate the offering in the lab, driving loyalty as well and, of course, helping the labs to become more and more efficient. Fantastic engine for us for growth in Professional Diagnostics, which as you seen grew 8% last year, is immunology, 14% growth across all the regions, way ahead of the market. A leading franchise in cardiac menu, but also in woman's health, vitamin D, 30% growth. And we have now approximately 60% market share in Europe on vitamin D. So immunology continues to be very, very important. The engine of growth for us and we see good momentum along with the increased placements of instruments. I'd like to move to Molecular with this example, which are 2 tests, which are actually not new in being completely new tests. But what they are, they are high medical value tests, which we bring on to an existing platform. So again, increasing the efficiency opportunities for a lab. And on the other hand, providing high medical value assays, which really can make a difference. MRSA, being the first one, which is a test that can literally save lives, in that it has a screening or diagnosis potential for hospital-acquired infection. And as you know, a hospital-acquired infection, growing increasingly with the resistance to antibiotics, 1 out of 20 patients in the U.S. at risk of acquiring a hospital infection. So this is a test that can make a huge difference. And then at the bottom, the herpes simplex, which consolidates our offering, on the first place. Also allows to put that offering on the same system and provides, again, a huge medical value, in that compared to the current standard where you do a cell culture, you would have a result in 4 days. Here, you'll receive a -- you'll obtain a result in 4 hours, which obviously allows you to move much faster into the treatment options. So with that, a brief outlook. We believe that we're well positioned. We will continue to provide our installed base into the market, complement with the largest menu in the industry. We have an exciting pipeline, which I'll talk a little bit about of new products and platforms. We have a leading franchise in emerging markets, which we'll continue to expand on, and of course, we need to continue to address, adjust and restructure the operating model around a Diabetes Care business, which continues to have very, very good fundamentals if you think of nearly 400 million patients -- or 400 million people affected by diabetes in -- and looking to be tested. And of course, last but not least, being an integrated part of the Roche Group, our contribution to the strategy in personalized healthcare and continuing to develop companion diagnostics, we have more than 250 projects internally between our Pharma and other diagnostics entity. So really advancing here and with all the associated bio-market programs. We've had 2 launches already, which I touched upon, MRSA and HSV. Good coming out of the gate. This is the launch table for 2014. I will just point out the top 3, which are new platform coming this year. And on the top, you can see the cobas 68 and 8800, which is our new molecular platforms. Two platforms, same system approach, modular in nature, the highest throughput that you will find in the industry and the highest degree of automation. We believe this will have a huge impact in how molecular tests are performed. Then, the entry into hematology, the acquired company from last year, CMI, that we are now bringing into the market towards the end of this year, in the CMI [ph] countries and then in the end of next year, for the United States. So here, coming from the central lab, moving into the specialty testing, an exciting opportunity to actually contribute to the existing customer base by providing a new and innovative platform. And then finally, urinalysis, same approach, serum work or work area approach was a modular approach here also to drive efficiencies. And then, of course, driving all this with the menu and you can see here a host of new assays that are in development, and that will hit the market to complement the menu that we have. So with that, I thank you for your attention and I will hand over to Alan.

Roland, thanks a lot. Yes, a warm welcome from my side as well. Great opportunity to give you a little bit of an overview about the financial results. And I would like to start, yes, really with a couple of highlights, yes, that we have seen this year. First of all, the strong core EPS growth of 10%, and there were different elements to it and I will explain them. And the other point is really a point I'm really excited about is the strong operating free cash flow of 5%. And you might say this 5% is -- what's the deal about it? And you might remember last year, we got a cash-in from the Montoro plan from Spain of CHF 700 million. So really, overcompensating this inflow in the year 2013 was quite an achievement, and I will show you later on why that was and why that happened. And that also leads to the point that the free cash flow went up 6%. So also, we overcompensated the increasing dividend from last year and showed some growth in the free cash flow. The improved financial result, positive development of the core net financial result, 11% lower expense, quite significant, you will see it later on, was quite a boost for our core EPS growth. And what we have done is, we had an early bond recall of CHF 1 billion with CHF 168 million in costs in December. In total, we have called back CHF 1.4 billion in the course of 2013 with the related costs of CHF 248 million. When you look at the net debt, net debt down by CHF 3.9 billion and the ratio of net debt to total assets went to 11%. I think it's not really telling the story about 2013 because what we have really done is we have reduced the gross debt by CHF 5.9 billion. And that was also the major trigger why the interest expense went down so significantly, but I will get to that. Good. Let's talk about the performance first. Here is the whole set of numbers, just to give you an overview. And I would like to start certainly with the sales Severin has talked about, its 6% up. He has explained already the difference to the Swiss franc increase of 3%. And what you're seeing is, there is kind of a box here with a dotted line around it. And these are the effects for PSI and 340B, and I will explain that on my next slide. But I think, you should see, really, how our performance has been when we put that out. Because you know our guidance has been that EPS growth is higher since sales growth, excluding these effects, and this target we achieved. When you go through it line by line, I think you'll see the core operating profit CHF 17.9 billion, 8% up even when you exclude PSI and 340B, 5% up. And here you see, the margin went a little bit down. But I would talk, really, about stable margins here. You see then, the core net income, plus 10%. Question is, where is the additional momentum coming from when you compare it to the up 8% in the core operating profit? It's not coming from taxes, but we're able to keep the tax rate relatively stable. The major point is here, the financial results. And the financial results went down and gave us the boost on the core net income, which translates very nice into the core EPS growth of 10%. And when you exclude the effects, up 7%. Operating free cash flow, I've praised already to a certain extent, CHF 16.4 billion and the free cash flow at CHF 5.4 billion. Good. That leads me to the 2 effects I would like to explain again. I think we've made some transparent to you during the course of the year already. I think the first one is really the 340B resulting from the year 2010 ObamaCare Affordable Care Act and the law was passed, passed in 2010. And in fact, it said that we have to give discounts here to certain hospitals for the so-called 340B hospitals, normally public hospitals, and in fact, for patients who are either under or uninsured. At that time, it was not clear whether this is for all indications. And also from a timing point of view, it was not so clear. We took a conservative stand, admittedly, at that time and started to reserve, and this was sales provisions that we have taken at that time, for these discounts. There was a ruling then in 2013 in July from HRSA. And the ruling said that the discounts are applicable for non-orphan indications October 1 onwards, so not retrospectively. And that really led to the point that we could release these provisions. And as you can see on the sales side, CHF 182 million just affecting Pharma. And you see on the cost of sales lines, these are the collaboration expenses which go with that. Then we have on the G&A line, and solely on the G&A line, the PSI, and that's the past service income. We have changed a couple of our pension plans. Evidently, it led to the point that the pension liabilities went down. And it gave us a positive impact on the P&L of CHF 302 million for the group and then you see it’s splitted up between the divisions and corporate. It has a tax effect to it here and then you see the net income was boosted by CHF 330 million based on these effects. Good. That leads me to the P&L. And in fact, a lot is said already from -- a lot said already, from Dan and from Roland. Sales 6% up. The royalties and other operating income down, minus 4%. And really the point this year, we have had less product disposals. And especially, the effect is obviously when you look at the second half of the year. The cost of sales line, plus 8%, higher volumes evidently. We had very good volume growth in both divisions that triggered once a year. And the other point is, really we started to invest in the network that we want to build, a manufacturing network in biologics, where we have announced to the market that we would like to spend CHF 800 million in CapEx over the next 5 years. M&D, pretty reasonable. R&D, as said, 15 NMEs in late-stage now. So more activity in the late-stage pipeline, more projects than ever. We have the Phase IVs, which are really related to the launch of the new products. And they're accounted here on the R&D line. So that's another point here. And also, development, health and work a little bit on the early stage and helped to boost the pipeline there as well. Good G&A in total. Let me say when you really put -- really look at G&A, all in, and you'll first it's a minus 3%, certainly triggered by the PSI effect. When you exclude it, I would say, 2 effects basically: IT systems that we have built and the other one is business taxes like the medical device tax, Roland has point to. Good. The core operating profit, up 8%, including the effects. Good. When you look at the margin, and this now includes everything, as I've said, when we exclude the special effects, I would say, still a robust picture that we are showing here. And as you see the contribution from Pharma CHF 16.1 billion on the core operating profit side and from Diagnostics, CHF 2.2 billion. Good. With that, let's talk about the financial results. And you see really here, the upticks that we have taken. First of all, the reduction from 2012 to '13 is CHF 267 million. Major positive impact comes from the interest expenses. And as said, I think it looks quite significant and it comes from the point that we have reduced the gross debt by CHF 5.9 billion. And that happened pretty much the vast majority right at the beginning of the year. That's why we could benefit so much from this effect. And the other point I would like to make is on the gains and losses on debt redemption. Because I've outlined already that we have done a couple of recalls -- bond recalls already in 2013. But I think it's quite comparable, the impact on the P&L compared to last year. So that's not a lot of shift here. And the other point is, I have to make it and there might be later on a couple of questions about currencies, this is the impact here from Venezuela. Basically, Venezuela negatively the devaluation effect, as you can imagine, was a minus CHF 63 million. And I guess, this year we will see a little bit from Argentina. Good. I think you'll see here the balance sheet in rough terms. Cash and marketable securities goes a little bit down by CHF 2.1 billion, not to worry at all. Basically, that's the difference between the gross rate reduction -- gross debt reduction and really, the net debt reduction. When you look at it and we have bridged that gap by giving you additional cash to pay, really, our debt back. You might remember the net debt reduction was CHF 3.9 billion. So I think you covered that bridge with the addition of cash. Other current assets and also the noncurrent assets are pretty stable. When you look at the current liabilities, and the non-current liabilities, you see quite a bit of a shift. The current liabilities, that's pretty much the reduction in short-term debt. And when you look at the noncurrent liabilities there are 2 effects: one is the reduction in the long-term debt and the other one is the pension liabilities, the reduction in the pension liabilities. And it's not all the changes, really, that we have done in our pension plans. It's the discount rates. Discount rates went up. And so far, it really the pension liabilities went down, which gave us a nice uptick in equity and you see, really, an increase of CHF 4.4 billion in equity and now with an equity ratio of 34%. Good. That leads me to the cash. And I think, as I've said at the beginning, excited about the development. CHF 16.4 billion, up 5% here. And I think when you really go through the divisions, I think you see here Pharma goes up 5%. And let me make a point here, Diagnostics goes up 9%. And I can say that diabetes care has contributed quite significantly to that increase here that we are seeing here. So CHF 2 billion from the Diagnostics division alone. When you look at what happened, I can give you, first, a message about the net working capital. And when you look at networking capital, we just had a slight increase. The major point here is when you look at the accounts receivables. And the accounts receivables, absolute number went down. Even in constant currencies, it went down by 1.4% despite the fact that we had a sales increase of 6%. So I think that's quite an achievement. And one element of this is really what happened in southern Europe and you might remember what has happened. In the year 2012, where we brought our exposure from CHF 2.5 billion to CHF 1.8 billion. And we benefited massively from the inflow from the Montoro plan, as I've outlined already. Now we move further on and you can see we had another reduction -- overcompensated the effect we have seen in the year 2012. And you see, really, pretty much every country reduced the accounts receivable exposure further. The only exception here is really Spain. But Spain, as said, we had a major inflow last year. And getting to this CHF 600 million number was quite a struggle, yes, in the year 2013. And I think, really, a fantastic achievement. It also, I think, outlines to a certain extent, the payment behavior in these countries has changed. We see, really, a positive uptick here. I don't want to paint a too rosy picture here for the future, but I think we see some improvement in that region, and perhaps, even in Europe as a whole. Good. Putting that together, net debt at the beginning of the year, CHF 10.6 billion. The operating free cash flow that I've outlined already, the nonoperational free cash flow, so dividends, the taxes and also the treasury impact. The currency translation -- transactions in non-equity instruments led to a net debt number of minus CHF 6.7 billion. And what does that mean? It means we have really entered the range of 0% to 15% for net debt on total assets. And you see here the development, half year to half year and here is pretty small admittedly to CHF 6.7 billion over here. Here is the year end of 2012. So the roughly CHF 10.6 billion over here. And you see what has happened. I think basically, the assets are rather stable and we have reduced the debt quite significantly. At the end of the year 2012, we have been at 16%, so not within the range. And as said, in 2013 end of the year, we have been in there. Good. What does that mean? Is that a major trigger point for us to be in the range. I mean, I wouldn't say so. I think we had all reasons to argue that 0% to 15% is the right area to be in. And still I would say, we are above the average of the industry. You might then argue, "Okay, also your cash generation is above the average of the industry." That's fine. But I think when I just look at this, and the area here, I think we feel comfortable with it. And we don't think that we are a complete outlier here. Good. Let me make a point about currencies. And I can really tell you this morning with the media, a lot of questions about that. And that we can encourage these, not just here Japan but especially in the emerging markets. And what we are showing you is quite some heavy impacts for the year 2014. You know how we do it? In fact this projection here for 2014 for the quarters to half year September, third quarter and also for the full year, are done on the basis that we keep all currencies stable at the end of the year 2013. And then just projects the impacts during the year. So it's really early days. And the question is how conservative is that stand? I think the major point I would like to point out is, really, where the U.S. dollar has been. As Severin pointed out, the major impact in the year 2013 was the weak yen against the Swiss franc. But now it looks like that the U.S. dollar, here, will have a major impact in the year 2014. Question is, is the U.S. dollar really staying at this level here? That's the major question. If the answer is yes, I think most probably you will see these impacts on our numbers over here. I think basically what I can say when you look at other countries and we've seen the weakening of currencies in other countries, I think we feel well prepared. When you look at Argentina, for example, and I think -- and I look at Dan and I look at Roland, I think really the preparations here for such a situation in Argentina, in fact, started 2 years ago. I think what we really ask ourselves, "What can happen? How can we mitigate such an impact?" I personally have been to Argentina last year. So I think we do our best really to get control about these things. I'm not saying that there might be major fluctuations in the year 2014, but it could imagine that there might be some -- in some of the emerging markets. Good. And with that, let me reemphasize the outlook that we're having low- to mid-single-digit growth when it comes to sales, core EPS growth ahead of sales growth. And the dividend outlook further increased for the dividend, which would be the 28th dividend increase in a row for Roche. Thanks a lot for your attention.

Alan, thank you very much.

And with this, I suggest we go right into your questions. Yes, please here. Can we have the mic here in the second row? Sachin Jain - BofA Merrill Lynch, Research Division: I'm Sachin Jain from Bank of America. Three questions: 2 product and 1 financial. Firstly, on the new Phase III starts, the CD79 79 and the PI3 kinase, you gave some headline comments, but I wonder if could provide some more color on the differentiation you think you've seen in Phase II for each of those assets for the CD79? Any color on the efficacy tolerability and where you plan to fit that in within your hematology portfolio, given you have a lot of assets there to potentially combine it with? On the PI3 kinase, any differentiation comments versus a multitude of PI3 kinases that are out there? And I think you mentioned plans for breast cancer, where would you plan to position it in the paradigm there? So that's going to be topic 1. Topic 2, on the PDL1 space. You've mentioned immune doublets. I wonder if you could comment on the breadth of that pipeline, just to get a feel of the breadth of your immune doubles versus Bristol and Astra? And then also on time lines [indiscernible] 4 combos, how do you view your immune doublet time lines versus competition? And the final question, I guess, you're expecting on the capital allocation, and Alan you've kind of preempted it, but whilst you commented that you're within the range, some could argue that your commentary on path from here is, perhaps, the least transparent of your peer group. So given the dividend of plus 6% doesn't seem to suggest the willingness for excess cash return to shareholders, [indiscernible] is within the range. I think, Severin, you previously commented, you see U.S. [indiscernible] valuations are stretched. What is the plan for cash from here or which of the above variables are you willing to be most flexible on?

Okay. Dan, you want to start with the product-related questions? Daniel O'Day: Right. Crizotinib, first of all, that's the name of the product I was forgetting. Thanks very much, Sachin. I'll try to do my best on some of these. I'm not trying to be evasive, but I just have to say that on the PI3 kinase and 79b, the data is pretty fresh. And in-house, we're evaluating this right now. So I'm not going into a whole lot of detail, but I'll give you a little bit of color. And on the PDL1 side, the reason that I may come across without all the details, is because it's a highly competitive field. So -- but on the PI3 kinase, as I've tried to articulate a bit before, we've got this beta-sparing PI3 kinase. I mean, that should give you a little bit of an indication for how we're approaching that. And that's for PI3 kinase mutants. There's probably in hormone receptor-positive breast cancer, we think there is about a 45% representation of PI3K mutants. So it's a significant size of the breast cancer population. The data is early yet. We're still going through it. But certainly, we're going to look at an ability to look at this possibly in a future metastatic trial. And then we'll take a look as well as what role this could play in adjuvant as well. I believe what we're really -- because it's -- really, that's an early one. I mean, we're really working with Phase I data there, just so you know. And we're kind of digesting that and deciding how to go. On a quarterly basis, I'll do my best to try to keep you updated on that. The other one is a little bit further ahead, the pan-PI3 kinase. We have Phase II results for that. Again, we think this could be differentiated in the field in the wild-type setting that could be complementary in nature. And we may look at this pan molecule also in combination with other products within our portfolio. We also have already on this pan molecule some Phase I combinations that we've already done. So we know -- I won’t get into the details yet, but we know there is tolerability with some other agents within our portfolio, which encourage us in terms of moving a bit more rapidly into Phase II or directly into Phase III depending on the nature of that. So it's probably all I'm going to say. I'm going to let the teams digest it, work on it and try to inform you as time goes on. On the anti-79b, ABC, as you know, we had again pretty early data here. And we do have the data in-house on the Romula [ph] study on Phase II. This also, we also had anti-CD22, which we're still evaluating, but the 79b looked encouraging, something we wanted to try to move ahead in the field. We're still trying to figure out exactly how and what indications. But as you would expect in this field of hematology, we'll try to look at potentially some faster market indications. And then look at doing a bit more derisking in Phase II with some other chemotherapeutic combinations to try to determine how we best proceed along those lines. So again on -- that's early days. I'll inform you a bit as we move ahead. But it's a really -- I think one other, if you like, tool in the toolbox relative to our strategy for working on the backbone of the CD20, in addition, to the biological modifiers like the Bcl-2 inhibitor. And then finally, on the PDL1, your question was around immune doublets and time line versus the competition. I'm afraid I can't give you a lot more at this stage on that. Obviously, when we can, when we begin some of these trials, they'll also be public on different websites and we'll inform you at that time or before that. But as I -- what I can tell you is we're looking at immune doublets both within our pipeline, so you can articulate some of the things that I talked about here today that we just recently brought in. We have other check inhibitors in our pipeline that we're looking at in terms of can they combine and how do they combine with PDL1. And as I said before, we may be also looking outside at other immune agents from other companies and looking at those types of combinations. So what I could assure you is that as the year goes on, there will be more and more data released, but particularly where we think we have a significant competitive advantage we want to be a little bit careful about data release in this area.

Dan, thank you very much. Your third question was on cash allocation and you made, in particular, reference to M&A. Our M&A policy as such has not changed. As you know, we are not into mega matches [ph] we have never been. We are looking for hold-on acquisitions both for products and technologies in both divisions, Pharma and Diagnostics. It is true that we have not done a big deal of later-stage acquisitions in the recent past. We did one or the other, for example, in Diagnostics when we acquired CMI in the field of hematology. And indeed, it is very much related to the evaluations, we just can't make the numbers for many of the assets, in particular, late-stage assets in the biotech sectors. But what we have done is we have focused very much on early-stage deals. Because when you work together with an external partner on the early-stage opportunity, typically it is not -- it's also but not so much about the money. It is much more about the capabilities we can bring to the table as a company and then move together with this partner, move the asset together with this partner to the next inflection point. So no change on the M&A side. I just wanted to make sure that this is clear. We have a question here in the fourth row, please. Andrew S. Baum - Citigroup Inc, Research Division: It's Andrew Baum from Citi. Three questions please. I'm delighted that Roche is now recognizing the promise of immunotherapy a little bit more openly than, perhaps, they've done historically. So 2 questions. Number one, you referred to an undisclosed hematite that you're going to be presenting later in the year. Is your observation for efficacy in the PD1 or PDL1, you believe, a reflection of the unique activity of the PDL1 or do you think it's simply a reflection of the patients you have selected within that indication, who may have been treated in order to bring that benefit? Secondly, with regard to the CD40 which, Dan, you highlighted, should we assume this is ready to go straight into Phase II or are you going to engineer the antibody? And then separately, do you actually have enough supply of this antibody, given you acquired it from a third-party? And then finally, just on capital allocation to Severin or Alan, Novartis has been pretty clear in its thought process regarding its stake in Roche. What would it take for you to become interested in that stake? And if you were, what are the restrictions on you in dealing with the nonvoting shareholders if you brought it back? If you could clarify that from a legal perspective that would be helpful.

Okay. Dan, if you take the questions on the cancer side, perhaps, I can directly comment on Novartis stake in Roche. Really, I have to refer you here to Novartis, because Novartis are the owners of the shares and they have to decide what they do with the shares, rather us telling them what they have to do with it. But certainly, we're always open to look at that. But I wouldn't speculate about something which has not yet been brought to our attention. Again, I think the addressee for this question is really Novartis, as the other owner of the shares. Dan, you want to comment on... Daniel O'Day: Right. I'm afraid, Andrew, you might me disappointed with additional information that I'll give you on the PDL1 immunotherapy. Thank you for acknowledging that we're providing some additional information. I think that is directly in line with the data maturing and I'm actually being more prepared to provide it. But relative to the new cancer type, I really can't comment on your question right now. I think it will be released in the first half of this year and I think then we can have a good discussion around -- rather, that's mechanistically-based or some other advantage when you have that. But I think it would be better to see the data before I infer on that. On the CD40, it's Phase I right now. It's been brought into our pRED unit in Basel. So it's being evaluated now. It'd be a little bit premature to discuss exactly how that will progress, how quickly that will progress. But certainly as we evaluate it and as it comes more deeply into the integration of the portfolio, I'll keep you informed.

We can have the next question here in the same row. Let us turn on the mic here. Thank you. Richard Vosser - JP Morgan Chase & Co, Research Division: Richard Vosser from JPMorgan. A couple of questions please. Firstly, on MetMAb, I noticed you're starting Phase III trials in first-line lung, just wondering what you've seen in Phase II that you can tell us that increases your confidence to start now, or is there anything we can read into the MetLung trial and its potential for success? And when can we see the data, more specifically? Second question, just going back to the dividend. Obviously, the dividend was in line, as such, inside [ph] with core EPS growth. Is that how we should think of the dividend going forward? Obviously, the corridor for -- you're well within the corridor now, so should we think of further dividend increases ahead of core EPS going forward? And finally, the third question just on R&D ratios. Clearly, there's a lot to be done. Presumably, the R&D is increasing ahead of sales this year in '14. Is that how we should think of it going forward? Or can we see some sort of offset from the fact that the CNS pipeline has had some disappointments, would we see maybe less investment there?

Okay, thank you for your questions. First, on the dividend. We wouldn't give any guidance on how much we increased, but what we really do is we put in a floor. And we give you a guarantee that we increase the dividend in Swiss francs, as we did last year. So irrespective of currency developments, we put in a floor and you can count on a dividend in a strong currency if you like. And then at the end of the year, we -- when we get to the decision, we will inform you as usual. Now on the R&D side, you should expect to grow -- to see R&D grow roughly in line with sales. Now I'm always careful in giving a precise guidance because it also very much depends on how the portfolio develops for the opportunities that come along during the year. But in rough terms, that's how you should think about it. Dan, on MetMAb. Daniel O'Day: Right. On MetMAb, you're right, we have started. The first patient in was in quarter 4 last year on the Phase III first-line. The second-line travel readout in the first half of this year. The intention with this program was always to look at both different tumor types, but also different stages of small cell lung cancer. We don't have any readouts yet on the second-line. But we should get that in the first half, and that'll be the readout that I also indicated that will be a regulatory readout if it's positive as well.

Okay. Can we move on the mic in the same row? And we'll work our way backwards. Thank you. Alexandra Hauber - UBS Investment Bank, Research Division: Alexandra Hauber from UBS. Three questions please. Coming back to the beta-sparing PI3 kinase inhibitor, I think the interesting thing when we saw in San Antonio that we got responses with the molecule rather than just stable diseases the other PI3 kinase inhibitor. And things seem -- tend to happen in oncology pretty fast these days, so would it be fair to assume you're also going to pursue a faster market access large Phase II with this agent? And if you do, can you just confirm us -- confirm or explain to us how much work you need to do on the companion diagnostic, which only existed as a footnote in research using the olden's [ph] presentation for something that is -- that has -- where you would pursue a faster access strategy or potentially sort of looking for breakthrough designation? So the question is really how much work do you need for a brand-new companion diagnostic? Second question, coming back to the R&D. I wasn't surprised to see an acceleration of the R&D spend, but I was a bit surprised about the explanation you provided or the components of that. I think one component was this technical, I can't remember exactly the term, technical development. In order to make -- for technical development to make in early, to make a real dent, are we really, for us, visible in the Roche R&D budget and you need spend to the tune of $100 million? And that's a big number for technical development, in my view. Is that -- am I'm thinking about this as the right thing because if you need to spend that kind of money in this area, we're always getting to the point we can talk about realistic barriers to entry for smaller players? And the final small question I have, Alan, on the 6% bond where you have, 2 times exercised an option to call some of the principal early. Can you remind us whether you can -- the remain principal, can you call it at any time you like, or is there a special timeline to that?

When the bond is due or what? What's the maturity... Alexandra Hauber - UBS Investment Bank, Research Division: I know when the bond is due, but since you had twice exercising option to call some of the principal early, can you just -- can you call the entire principal early at any time you like and so keep that...

Yes. We can call it anytime we like. And just to answer this question right way, we can call it anytime we like. We have such a clause in our terms and conditions if you like. And the bond is due 2019. So it's really a stretched out impact that we are going to see.

Perhaps I can just make a comment on the technical side, I mean, what we are talking here at CMC. So it's ramping up the material which we need as we go into the clinics. Alexandra Hauber - UBS Investment Bank, Research Division: So it's not in COGS that we are seeing right now?

No, no. This is still in R&D. It goes into COGS, into cost of sales, when we have a product on the market. So when we do the ramp-up of material for Phase I, Phase II trials then it goes into R&D. And what really happened was that we had these fall-offs [ph] of Phase II trials. And that, of course, you see now being translated moving into Phase III. That's why we have now 5 -- 15 new molecular entities in Phase III. But first, we had to get there. And that required that we ramp up CMC. So we had to expand resources in that area, and that is expensive. This is not cheap. And if you like -- this is a hurdle of entry for many players. And actually what we also see is very often if we have collaborations with third parties, smaller companies, typically this is an area where we have to catch up and do a lot of work to get this ready for pivotal trials. So this is quite something. That was -- perhaps, Dan, you want to add to that? And then we had an additional question. Daniel O'Day: Just to add to that, I mean, I think it is a competitive advantage for sure, the knowledge in CS [ph]. Beyond the cost, just a simple knowhow and expertise associated with biologics is significant. It's an integral part of our product development piece. It happens earlier and earlier these days, particularly in oncology, where you're moving programs from Phase I to potential label-enabling Phase IIs or Phase IIIs. And it's not uncommon for that to be on critical path, too. So the investment is fundamental. I think the complexity of these molecules, we talked about Kadcyla in the past. I mean, what it takes to put an antibody, a linker and the chemotherapy in just the right balance so that the chemotherapy isn't released before it hits the tumor cell. I mean, this is really, really a significant competitive advantage. So it's -- sure, it's -- there's a cost hurdle, but I think there are many more hurdles beyond the cost hurdle that we have to consider. So -- and the exciting news is when you have products like Gazyva, that have breakthrough status and the agency is moving so quickly. It's all supposed to put real demands, not just on the clinical supply, but on the commercial demand side. And with Gazyva, we were able to work with the agency, the FDA, to all see you a similar clinical supply for early launch, which allowed us to get that out into the field. So it's really fundamental. We could have a whole separate discussion on that. In terms of the beta-sparing PI3 kinase, again, you have seen some of the Phase I data on this. We're pretty excited by what we see in terms of the biomarker. We have been working on that for a while now. This came out of Genentech research and our colleagues in Genentech and molecular diagnostics in Pleasanton that have been working on this PI3 kinase assay for quite some time to fine-tune it, to get the cutoffs right. Clearly, we have more work to do on this. But as you said, it's a very competitive area. We think we have unique differentiated complementary PI3 kinase in the mutant in the pan. I'm going to let the teams dig into it, but I think we're going to move as quickly as we can with this. Alexandra Hauber - UBS Investment Bank, Research Division: I said this is not in the traditional language of the time line from catching one another program... [indiscernible] Daniel O'Day: You mean, the companion diagnostic itself? I mean, I think we're in pretty good shape on this project. It can be. I mean, it can be that you need a validated assay to go into, similar to what we just talked about in CMC, to go into label-enabling trials. But I feel on PI3 kinase, we're in pretty good shape on this one. So I wouldn't expect that to be a stumbling block for us, at this stage.

Okay. If we can just move on in the same row. Yes, please. Thank you. Luisa Hector - Crédit Suisse AG, Research Division: It's Luisa Hector from Crédit Suisse. I've also got 3 questions. The first one, if we look at the full year '13, we can see that margins were flat if you strip out the 340B effect. And we also saw that we had a boost from the net financials that then helped you see the earnings growth ahead of sales. And so looking now to the guidance for 2014, with earnings to grow ahead of sales, and we've got the similar potential effect on the net financials, a booster there. Can you give us any confidence that we will see margin expansion in 2014 especially given the generic effect that you're likely to see within Pharma? And the second question would be on the CD20 franchise, and just whether you got any early comments on the Gazyva launch and the competitive landscape with the first oral launch, just any early commentary on that. And if we look at Gazyva and the approval that you have so far, is it fair to assume a fairly low sales potential, because it's quite limited in the CLL, especially with the chlorambucil combination? So we shouldn't get too carried away with sales forecast until you have more data in the 2016 time frame. And third question, hopefully, quite quick. You had a comment on the EMA investigation, so I think there was a reinspection in November and that brought up some issues. So I just wondered what they were, and what you need to do to fix them?

Okay.

[indiscernible]

Yes, please. Okay.

Yes. First of all, I think the guidance is the guidance. I think that's the first point I have to make here. I mean, we don’t have some margin in the guidance. I think that's, how should I say, it's a formal point I have to make. But having said so, I think very clear. I've been at JPMorgan and I said, I think that we are expecting rather stable margins moving forward. And I think that's a little bit -- I think the area we're in. I am very clear. I think we don’t expect a margin deterioration. Are we clear? But I also wanted to, how should I say it, flag to the market. That the steps we have had in the past, and especially in the past years, which were very significant. That this expectation might be a little bit overdone. Daniel O'Day: Right, then let me try to get -- I think I'll answer the numbers 2 and 3 together, but the CD20 and the chlorambucil-type question. I mean, maybe just to frame it, just take a step back from MabThera first of all. I mean, there are a lot of companies in hematology that are going for fast to market strategies right now in this area. And it's essentially today of course in CLL, which is about 18% -- 17%, 18% of total MabThera revenue. Just to put that into context. And in terms of real competition in non-Hodgkin's lymphoma, either [indiscernible] or aggressive [ph], which is 70% of the MabThera sales. We think people are either pursuing this strategy where they're adding on top of the CD20 backbone, whether its MabThera or Gazyva in the future. Or perhaps going head-to-head, but those would be at least 5 years out. I mean, just to put that into context. So let me concentrate a little bit on CLL, where there -- clearly, it's a competitive environment. We feel good about our position in CLL both with Gazyva and the data I presented here, the additional studies we have going with Gazyva with the biologic modifiers including the Bcl-2 inhibitor and the ability to combine those two. And you'll see, even within CLL, we see companies looking at using the CD20 backbone. But specifically, what could we look at for Gazyva in the United States? You're right. I mean, the label in the United States now is for a chemotherapy back component of chlorambucil plus unfit patients. And we obviously are continuing to work to move that up and to expand the chemotherapy backbone. But I think we can be successful in the early days so far that we have in the market are that it is being accepted and endorsed in that patient population, and potentially starting to be experimented and used beyond that patient population. So I hope that gives you a bit of an idea. I think for both for Gazyva and for other -- some of the competitive products, I think we should let a little bit of time go throughout this year and keep you updated in the first and second quarter and see how things are moving. I would just make one other comment on Gazyva is that we've priced that product in the United States with an intention to replace eventually the MabThera franchise. And also, with Gazyva, it's a set course of therapy versus some of the other competitors there that treat through to progression. So I think that's something, too, that you would take into your potential future estimates as well.

Dan, you also want to make a comment on the EMA orders? Daniel O'Day: Yes, sorry. So EMA, we did have a reinspection, as you may now recently. We -- first of all, I would say that we have been working very hard on our CAPA [ph]. And in fact, now for 19 medicines that have been reviewed both by us and by EMA, there has been no difference to the safety, benefit-risk profile of the product based upon this additional safety data that's being processed. And I think, that's very encouraging and in line with what we expected as well. So we do have additional observations with the audit that we're continuing to work on. We've got the infringement procedure, which is still ongoing in Europe. And we would expect, over the first half of this year, for EMA to be working with the commission to progress that infringement procedure -- we don't want to anticipate what that could or could not be. But we'll get answers, I think, in the first half of this year. The important thing is, I think, we're very clearly committed to working with the authorities to make sure that we've got the data that's -- that they're requesting for all the products.

Thank you, Dan. But let me reemphasize, I mean, the safety profile of the all these medicines, which were concerned by these audits has been reconfirmed, and publicly reconfirmed by the EMA. Can we have the next question please? If we work our way up, perhaps over there. Yes, please.

Mary Anne [indiscernible], Prime Avenue. Two questions, please. One is, what sort of growth emphasis should we expect in 2014 and maybe medium term from patient access programs in China and maybe some other Asian markets? And I think it has been a very significant growth driver for some products in the past, and it's just very difficult for us to gauge the dynamic there. I mean, are the new patients sort of phasing in at a regular rate, or is there a certain lumpiness? Just any thoughts you could share with us there on how to model that. And my second question is on HER2-positive adjuvant breast cancer. Just wondering if all the Phase III trials that you now have ongoing with the newer agents were to come out positive, then what do you think you would need to do in terms of ancillary trials, in terms of patient numbers in order to convince physicians that really the vast majority of patients who are currently on Herceptin would in the future be getting one or both of the newer agents, and what sort of the time lines around that? I mean, the reason for the question is that I think on the one hand obviously looking at the Herceptin adjuvant program where you had about 12,000 patients, if you multiply that through for Perjeta and Kadcyla and the combination, you might be hitting some feasibility issues. But I'm also wondering, I mean, with all of the trials that you now had also in the metastatic setting where -- that were positive irrespective of the chemotherapy, do you really have to do a full trial with every chemotherapy and every patient population look at hormone receptor co-positive breast cancer again? Or is there somehow, I don't know, sort of a cut-off where you can say, "Okay, with the program of this size, we can probably switch 80% of the -- or 90% of the current Herceptin franchise and adjuvant to one or both of the newer agents?"

Dan, you want to give it a try? Daniel O'Day: Sure, I will. I'll do my best. So first on the patient access programs, I don't think they're really lumpy. I think they're relatively constant and consistent. Probably the most successful one so far that we've had is in China, in particular. And what we're seeing in China is a pretty consistent uptake when you look at the growth rates of products like Herceptin. I don't have it right in front of me, but I think it's in the magnitude of 60% last year. That's really driven -- really completely by the patient access programs. So as you know, the way they work is that there's a certain component that's private pay. There's a certain program that's funded through a charitable contribution. And we make sure that patients get the entire Roche-sponsored treatment, of course. Now what I would say is what's interesting is for the patient portion of the contribution. We've now had 2 provinces that are starting to offset some of the costs that the patients would have. Now this is a province-by-province decision, but we're encouraged by that. If you like it, it's semi-reimbursement at the government level, some oncology products in China. So I think that we continue to see good strong robust growth with those programs in China. In other countries where that's appropriate, it doesn't work in all countries, and it's not appropriate in all countries. But we also had other strategies, of course, in other countries where we work with, second brands, or we work -- second brands that are priced differently in the public markets in some country where that's appropriate. We have a big private/public split. And all of those programs are leading to the growth numbers that I showed before. I mean, the 21% growth in China this year is really driven predominantly by the oncology portfolio in these programs, just to give you a feeling for that. Relative to the HER2, well there's a lot of forecasting in that, I guess. The fundamental thing I would say is like everything in oncology, it's data-driven, right, and data-dependent. I think what we do feel pretty comfortable with is that we have a very comprehensive program in the adjuvant setting with Kadcyla and Perjeta. I mean, you know it, you've seen it in different combinations. I can't tell you today which combination in the adjuvant setting is going to be the very best. I can't even tell you what combination in the first-line metastatic center is going to be the very best on Mary Anne [ph]. But we certainly have, I think, thought through both the biologic treatment regimens, as well as chemotherapeutic treatment regimens. And our full expectation is that if we have a significant advantage like we had in the metastatic setting and adjuvant, that we will see patient conversion because, particularly in the adjuvant setting, when you're talking about potentially moving in HER2-positive breast cancer from Herceptin today, which is a 50% curative rate almost, patients that do not go into progressed disease. If you were able to increase that by 60%, 70%, whatever it's going to be, I think in a curative state you would certainly want to give every patient the very best chance of being cured. So it's even more compelling, I would say, in the adjuvant setting than it is in the metastatic setting. But I wouldn't want to forecast exactly what that data will be. Other than to suggest, I think the program is comprehensive enough to answer the questions that you pose.

Perhaps, can I add one point on the bumpiness [ph] in terms of access programs and increasing access to the public segment. China is a case where you have a pretty decentralized decision process and, as such, there is the first city, the first province stepping in and you have a pretty gradual development. There are other emerging markets where you have a much, much more central decision-making on a national level where you can have 1 or 2 tenders for the annual demand of a certain medicine. And even though in the longer term, of course, you can argue this is an evolutionary development, what we do see is a very bumpy development on a quarterly business. And have seen this over the last year. You remember in one quarter, you asked, "Why is international market so low?" And we said, "Yes, because the tender didn't come in." And then in the next quarter, you said, "Can we predict that this goes on like this?" And we said, "No, no, be careful." There was this one-off tender in Russia, for example. Or there was this one-off tender in a country like Algeria, other countries in North Africa. So you can have a bit of a bumpiness when it comes to public tenders for the public segment. Daniel O'Day: Yes, that's a good point. I would, too, restrict the one-patient access program. Those are public tenders that really are not connected with the patient access, but they have a lot of lumpiness.

Good. Can we have a question there? And then, we'll have one more question before we break out into the divisional sessions. And next question? Daniel O'Day: Yes, please. Keyur Parekh - Goldman Sachs Group Inc., Research Division: It's Keyur Parekh from Goldman Sachs. And I've got 4 questions, please. Two for you Severin, 1 for Alan, and 1 for Dan. Starting the first one for you, we've heard Roche recently talk a lot more about annual pricing or price caps across therapies. Can you just help us think about what -- how this might actually be implemented especially from a U.S. perspective as we look at the next kind of 3 to 5 years? Secondly, just trying to get a sense of triangulating different things, you've given us a net debt to assets corridor on the balance sheet. Cash generation is growing at 5% to 10% per year. And you're telling us you won't do large M&A and bolt-ons. If you triangulate the 3, you get to bolt-ons that are going to be somewhere in the region of $30 billion to $40 billion over the next 3 years. Are those still bolt-ons? Alan, on the margin side, if I look at any large-scale Pharma company that's going revenues at 6% per annum, it's very difficult to see why margins won't expand. Can you -- especially given the geographic mix you have, the product mix you have, and if R&D is going to stay roughly flat as a proportion of sales, what is going to go up? And then lastly for Dan, on the 79b program, the Phase II that you did was in combination with rituximab, is there a chance that you can do the next stage of the program directly in combination with Gazyva, or do you need to do early work on that end?

Okay. I'm not sure whether I got the first question right. But if I understood you correctly, you were referring to the pricing and how pricing develops across different regions, in particular the U.S. Was that the question? Keyur Parekh - Goldman Sachs Group Inc., Research Division: No. What I was talking about was we recently heard Roche talk about moving to a per patient per year or the per cost of therapy trial, I think, using multiple...

Okay. Thank you, I misunderstood, yes. Yes, very much so. Our current pricing in the industry as a whole is very much based on a price per vial on milligrams, if you like. And we feel that we should develop more sophisticated pricing models as we go forward. Also in the developed markets, in particular, in cancer. If you look at Avastin, for example. It is used for different cancer types and we know that effectiveness of Avastin is different for the different indications, and the question becomes even more critical as we have more and more combination therapies where we stack biologics one after another. So our aim clearly is that we come to more sophisticated pricing models where we can link the value of the medicine closer to the price rather than having an average price across indications and across combinations. Now from what I see in our discussions with the various national health care systems is that payers, as such, are pretty open to that concept. They want to pay for the value rather than just for a vial or for milligrams. That the trick or, let me say, the hurdle is the implementation. Because if you do, for example, indication pricing, what you need is an IT system which tracks which vials has been used for which indication. And that is in many countries today, not the standard. They have only systems, which tracks how many packs of Avastin have been used, not necessarily for which indications. And it gets even more complicated if you then look at the outcome because then you do not only need the data for which indication it has been used, then you'll also need the data of what has been the clinical outcome for a specific patient group. But having said that, we are closely working together with governments and with payers to develop more sophisticated models as we go forward. But I think, Dan, it's fair to say, it's still early days, right? Daniel O'Day: Yes, yes. I mean there are some countries that are further ahead than others. I think we're paving the way here. We're reliant upon a lot of other stakeholders. But Italy, U.K., couple of markets are quite advanced. And we know this is the future, so we have to keep pushing it. I would say the U.S. is not at the cutting-edge of this, currently, to your question before.

Right. Then we had a question on the margin side. Alan, you want to take another shot at it?

Yes, sure. Yes. And I think Keyur makes a good point here, and sort of I expected it. And you've seen it. I think when you really strip it out, in fact, it's really -- it's a thin line between stable margin, margin expansions, et cetera. And admittedly, I think this year we have invested in a couple of things, and I think rightly so. And when I look at the Phase IV, so I think that's something and potentially that stays a little bit with us. More late stage is evident. So I think R&D is an element here. Before I said, okay, there is just flatting, but give us a little bit of flexibility, we need it. I always made this point. And I think very clear this year was a year, where we had to take that credit and use it. And I think we did well with that. The other point really I'm looking at is a little bit cost of sales. I think higher volumes are -- it's evident. And on the diagnostic side, we have price pressure, admittedly. I think that's a little bit of a driver that we're finding over here. And really, we are going to build the manufacturing network. So when you put that into perspective in 2014, where we have said the guidance for the sales is low to mid single, that's very -- that's the point here. I think really that puts things a little bit into perspective. I think going forward, we will see what that means and whether there is more momentum. It looks like that there might be more momentum. But okay, the guidance is the guidance for 2013 -- '14. And let's see how that goes. I've never said the margin goes down or significantly or whatever. I think we're talking really about here, how should I say, relatively small amounts.

Dan, there was a question on 79B...

We can pretend they all start to articulate. We've moved the 2 PI3 kinase and the 79B in the late stage development. But we did that in a fairly aggressive early way, and for good reasons because we want to get quickly into labeling enabling trials. But at the same time and specifically with 79B, the Phase II trial still continues. We have a final readout of it in 2014. And the team is looking at a variety of combinations, certainly looking at Rituxan with potentially other chemotherapy combo backgrounds. Probably, a little bit early to think about Gazyva with this age and at this stage. But we'll know more as the year goes on. And we'll also have the final conclusion on the CD22 as well, which is still -- we're still getting data on the Phase II on. So I would say, 2014 for this compound will be a year where we're planning, where we're trying to decide exactly how to move this product forward, getting some additional data on it. And then being prepared to move with speed in the future years. So I'll have to keep you updated on that.

I must say they have to... Keyur Parekh - Goldman Sachs Group Inc., Research Division: One more question on the evaluation of the balance sheet, guidance, the M&A guidance.

Yes. Time is advancing, so I'll keep it very short before we break out into the divisional sessions. And I think the Finance session is here. Carlos [ph], just correct that. Yes, what I can say? We have given you a range. We have now entered that range. I think there is still some flexibility. As I said beforehand, there is no change in our M&A strategy. We have clearly stated that we want to increase the dividend. And we'll see how the cash flow develops. I agree with you. There should be a strong cash flow as we go forward. I think it's a good problem to have. And we'll cross the bridge when we get there. Now I'd like to ask those of you who have additional questions on the balance sheet to stay here. And then, we have 2 breakout rooms. One is...

The old one up [ph].

The old one up [ph] for Diagnostics and for Pharma. And may I also remind you that after the breakout sessions, we are happy to meet you again for some cocktails before -- and some, perhaps, a bit more informal discussion before we go home. Thank you very much for your attention.