U.S. Markets closed

Edited Transcript of ROG.VX earnings conference call or presentation 30-Jan-20 3:00pm GMT

Full Year 2019 Roche Holding AG Earnings Call

London Feb 6, 2020 (Thomson StreetEvents) -- Edited Transcript of Roche Holding AG earnings conference call or presentation Thursday, January 30, 2020 at 3:00:00pm GMT

TEXT version of Transcript

================================================================================

Corporate Participants

================================================================================

* Alan Hippe

Roche Holding AG - Chief Financial & IT Officer

* Karl Mahler

Roche Holding AG - Head of IR

* Severin Schwan

Roche Holding AG - CEO & Executive Director

* Thomas Schinecker

Roche Holding AG - CEO of Roche Diagnostics

* William N. Anderson

Roche Holding AG - CEO of Roche Pharmaceuticals

================================================================================

Conference Call Participants

================================================================================

* Andrew Simon Baum

Citigroup Inc, Research Division - Global Head of Healthcare Research and MD

* Keyur Parekh

Goldman Sachs Group Inc., Research Division - Equity Analyst

* Mark Douglas Purcell

Morgan Stanley, Research Division - Equity Analyst

* Michael Leuchten

UBS Investment Bank, Research Division - Co-Head of Pharmaceuticals Research of Equity Research

* Richard J. Parkes

Deutsche Bank AG, Research Division - Director

* Richard Vosser

JP Morgan Chase & Co, Research Division - Senior Analyst

* Sachin Jain

BofA Merrill Lynch, Research Division - MD

* Sam Fazeli

Bloomberg Intelligence - Research Director of Europe, Mideast and Africa

* Simon P. Baker

Redburn (Europe) Limited, Research Division - Head of Pharmaceutical Research

* Timothy Minton Anderson

Wolfe Research, LLC - MD of Equity Research

================================================================================

Presentation

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [1]

--------------------------------------------------------------------------------

Good afternoon. Great to be back in London for our full year briefing. Let me get right into the numbers, which we shared with you already. This morning, you have seen group sales up by 9%, in local currencies 8%, at constant rates. Core EPS growing at a stronger rate at 13%. And on that basis, we propose to increase the dividend to CHF 9.

On a divisional basis, a very strong result on the Pharma side, entirely driven by the newly launched medicines. I will come back to that in a moment. On the Diagnostics side, solid, more moderate growth at 3%. I'll get back to that in a moment as well.

Now if you look here on the quarterly sales development, you see a 6% sales growth after a very strong third quarter. And there are really 2 reasons why sales growth has come down in the fourth quarter. On the one hand, we do feel the impact of the biosimilars now in the U.S. So finally, they have entered. And on the other hand, we took a decision to take down inventories in China, actually for both divisions, for Diagnostics and for Pharma, even more so for Diagnostics. And that also explains why you have seen negative growth rates in China in the fourth quarter. And that has also affected, of course, the results of Diagnostics. If you look at the underlying development, Diagnostics continues to grow in the mid-single digit, and also the demand in China from the end customer remains very strong.

Now that's really on a high level the summary and the story about 2019. You see on the one hand, as expected, the significant impact from biosimilars, in total, CHF 1.5 billion, the majority in Europe, also Japan. And you see the first impact now in the United States with CHF 300 million. And on the other hand, you see the strong, as expected, growth and demand for the new medicines with over CHF 5 billion. And it's this balance which has been responsible for the growth last year. And Bill will comment on the various franchises in more detail.

If we look at it from a regional perspective, of course, very strong growth in the U.S. What I'd like to point out here is that we are back to growth in Europe. We have seen negative growth over the last 2 years. Now the new products start to overcompensate for the decline due to the entry of biosimilars. International, also very strong, primarily driven by China. That's also reflected on the operating profit level. You see slightly improving margins and actually a very strong free cash flow this year.

If I just turn to the portfolio. We have one of the leading portfolios in the industry from a quantity point of view. But at the end of the day, it's a lot about the quality of the portfolio. And one of the indicators here is the breakthrough therapy designation we get from the FDA where we have a leading position in the industry. And that, of course, speaks for the differentiation but importantly, also about the speed of the approval process, reimbursement and how quickly we can bring those new medicines to patients.

We've had quite some news flow. Last year, we brought 2 new medicines, Rozlytrek and Polivy, to the market. We have seen important readouts, also late-stage readouts, and we've brought a number of new tests and platforms to the market on the Diagnostics side.

We keep looking for external innovation. We invest a lot in our internal research and development, but a lot of good things are happening outside of Roche. You see here a number of transactions in both divisions. Just to highlight, Spark, which we finally could close towards the end of last year, it took actually longer than we would have thought, but now we are very happy that we could close the transaction, and we have already started with the integration. So that provides us with a new platform, an important platform initially for rare diseases, longer term potentially beyond rare diseases. So we're very excited about this.

Let me close with the outlook. First of all, really extraordinary strong clinical news flow in 2020. So this is really a year where we have a lot of readouts. Now we'll see how many are positive. Not all will be positive, but there is a lot of optionality here. We know for sure that -- for sure -- I mean, given the data, we are extremely confident that we will launch 2 new medicines in neurology with risdiplam and satralizumab. We have some very interesting opportunities in autoimmune diseases, in particular with etrolizumab, and also in ophthalmology, and there's a number of readouts in oncology. Just to highlight one with Tecentriq in liver cancer. The growth last year was very much driven by the new indications where we have been first to the market, which was small cell and triple-negative breast cancer. And with liver cancer, we have yet another opportunity to be the first on the market. And many of you have probably seen the results, which we recently presented. I mean this is really a big step forward for those patients concerned, a very prevalent type of cancer, in particular, in emerging countries such as China, and a type of cancer where there's still enormous unmet medical need.

Good. With this, let me close. We expect low to mid-single-digit sales growth for the current year. We expect again to grow the earnings roughly in line with sales. And on that basis, we should be able to again increase the dividend next year.

Thank you very much. And with this, I think, Alan -- no, Bill. And that makes me realize that this time, for the first time, we have Thomas Schinecker with us. I'm already so used to working together with him. He took over in August last year, but I believe it's the first time that you are in front of this audience. So welcome. And for the time being, over to you, Bill.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [2]

--------------------------------------------------------------------------------

Thanks, Severin. Welcome, everyone. It's very nice to be here and talk about the results we have. I think we have a duty to disclose our financial results, but the thing that we get really excited about is talking about our pipeline progress. And it continues to really amaze us to see the power of science brought to patients with many really serious diseases. And I'm very excited to say that just in the month of January, we've approved 8 new pivotal studies for 3 novel molecules. And I mean this is a pace of progress that every year we think it can't get any faster and there can't be any more breakthroughs at the pace that we have, and every year it seems to get faster. So I think you'll see the people of Roche are going to be very busy in 2020, but that's going to pay off in rich rewards for patients in the years ahead.

So back to the numbers. So we're very pleased to bring in a full year result above 10% or 11% growth. The U.S., obviously, a really stellar result, again, fueled by the new medicines. Europe, with a return to growth, which is encouraging considering that there was still a significant loss for MabThera. And then the bulk of the Herceptin losses in biosimilars happened in 2019. And so if you think about -- we can lose the majority of our Herceptin sales in 1 year and still grow in Europe. With all the price pressures and everything else, that's, I think, quite a remarkable result. Japan with 9% growth, and again, Japanese pharmaceutical market is shrinking, and so delivering 9% growth is no mean feat. And then overall, international, with 15% growth driven by China, so quite an excellent result as well.

This is what the P&L looked like. I think that the notable things here, royalties and other operating income, substantially down due to the loss of Cabilly primarily. And so, again, that's sort of a shock we weathered and managed to still deliver 11% sales growth and 12% core operating profit growth. So I think we felt quite good about that.

Cost of sales, up about 7%. Our COGS and period costs in manufacturing were up 3%, while our volume went up 16%. So I think this is further evidence of the work we've been doing with the transformation and new ways of working in manufacturing. This is really the third year in a row we've had similar types of productivity gains, so really a substantial feat.

But you might wonder, with a 16% volume increase and all these productivity enhancements, are we missing anything? And I'm pleased to say we had the highest level of on-time and in-full deliveries, the highest percentage that we've recorded in a decade in 2019. So I think the transformation is delivering not only productivity, but it's also delivering quality. And that's part of -- the mindset is really building that quality in rather than inspecting it at the end.

M&D seems up 10%, which was actually more than we originally intended to do. And essentially, what happened is as the -- as we got further into the year, we realized we had some good opportunities to invest in things that would continue to enhance the growth of some of our new products. And we thought that was an investment in patient access and in our future, so we took that. R&D, you can see up 6%; and G&A, 8%. But the core administrative cost was not such a big increase. This mostly reflects the incorporation of the Flatiron G&A and full FMI G&A on a full year basis. So that was the sort of overall P&L.

This, I think, is a very nice chart because it exemplifies the growth and the strong and broad base of growth we have, now with 3 molecules, with over $1 billion -- or over CHF 1 billion growth in a single year as well as Perjeta and Kadcyla with big contributions. And again, as you look down the list, a lot of new products on the list and a really strong base.

I'll say a few things about some of the oncology products. I guess I would just say, overall, again, some good features here, like for example, Alecensa with 38% growth nearing CHF1 billion. So we should get there this year. Tecentriq, 143% growth and nearing CHF 2 billion. And then I'll comment a little bit more on some of the other franchises as we move forward.

HER2 franchise. Take a look at Q4. I think this is particularly interesting because for those of us who've been working on new products in HER2 space for over a decade, I mean, it's really encouraging. So Q4 was the first quarter where our sales of Kadcyla and Perjeta exceeded our sales of Herceptin globally. And I think that's a pretty big milestone, and that's really driven by the uptake of Kadcyla and Perjeta in early breast cancer treatment around the world. So I think a great milestone.

If you look forward and what else is coming in HER2-positive space, we announced the positive study of the FeDeriCa study, which is looking at the fixed-dose combination of Herceptin plus Perjeta in a subcutaneous form. And we haven't had a tremendous uptake of subcu Herceptin, in part because the desired regimen -- the preferred regimen for Herceptin now is with Perjeta, whether it's in metastatic disease or in early breast cancer. And so if you have a subcu and an IV, that's not that helpful if you're already going to have to give someone an IV. But with this, you have the opportunity, once someone's come -- a patient's come through the chemo part of the therapy, for the adjuvant continuation, they can come in and instead of having infusions lasting from 2.5 to 7.5 hours, we're talking 20 to 40 minutes. So we think that's a pretty big advantage for patients. And it could be an important factor for converting patients to the fixed-dose combination, and so we're excited about it. We expect to launch in the U.S. in 2020.

Switching over to the hematology franchise. I think we -- as you know, we don't include VENCLEXTA sales. So while this sort of looks like it's going down, if you include our part of the VENCLEXTA sales, it's actually a positive trend. And I think broadening base, you still see a large portion of MabThera but growing parts for Gazyva, which is doing well in first-line CLL, for Polivy, which launched in the middle of the year in the U.S. and is launching in Europe. And then I think we're going to continue to see growth on both those products. So a pretty good outlook in hematology.

Going to switch over, talk about immunology for a moment. So we continue to serve a number of different types of diseases in patients ranging from rheumatoid arthritis, IPF to ophthalmology with XOLAIR -- I mean, sorry, LUCENTIS isn't in here right now, but with XOLAIR. And I think we've held up well in the face of a lot of competition. And so I think that's really the main message here is that we have some pretty robust brands. If you look at a product like XOLAIR, has had numerous competitors coming into XOLAIR's key allergic asthma area, and through areas like urticaria and other indications, XOLAIR's held in there really quite strong.

Neuroscience, it's really about OCREVUS and continued growth. And again, we really see no letup in this. The U.S., the share of new and switching patients has remained around 40%. And that's really held up over, I don't know, 8 or 9 quarters now. I think that's really remarkable if you consider I think there are 16 therapies in MS now, and OCREVUS gets 40% and the other 16 split the other 60%. So it's really proving through the twice-a-year infusion, through the strong dosing profile with the unsurpassed efficacy on disability progression, whether you're in primary progressive disease or relapsing. It's just -- it's a really good choice for patients. And we're seeing, again, increasing share in early patients but really use across the whole spectrum. Again, more to come on that, and we like what we see.

I think we're excited to have 2 new molecules coming now, medicines in SMA and satralizumab. We announced just in January the positive results of FIREFISH part 2, which is the second part of the Type 1 study. And so we're glad to see the first part results sort of reaffirmed, I think particularly because, again, the FIREFISH patients tend to be significantly older than the patients on the other therapy studies. And so we're really encouraged by the efficacy we see given the older setting. We've filed now based on FIREFISH and SUNFISH, and we're expecting approval in the first half of the year in the U.S. We'll be filing later in Europe because the EMA wanted us to file with both FIREFISH and SUNFISH, so Type 1 and Type 2 and 3. They wanted us to wait until we had all the data available before we filed. But I think we have a strong package here and a compelling -- I think both a compelling product profile, but a lot of patients who are not currently served, whether they're older and they have weight kind of limitations or whether they have limitations due to the need for intrathecal injections. So I think risdiplam is going to be a great addition to the arsenal against SMA. And then satralizumab in NMO, which we filed and we're looking forward to launching later this year.

All right. Let me come back to Tecentriq. So again, excellent growth curve here. You can see we're well over the CHF 2 billion run rate. And this has really been driven, again, through 2019, primarily by the small cell lung cancer indication, continued use in nonsmall cell, in first-line growth there as well as the triple-negative breast cancer. And we were very pleased to add the results from Tecentriq plus Avastin in liver cancer.

Some of you may have seen these curves. This is what we presented at the ESMO Asia conference in Q4. And I think for those of you who are used to looking at oncology overall survival and PFS curves, you'd agree, this is really a beautiful sight and what it represents for progress for patients. And I think, particularly, what you see on the progressive free survival curves, often, you see the curves coming together at the end. And here, you see quite a large number of patients that seem to be getting a prolonged benefit. And then you see that also playing out in the overall survival curve where the standard of care, sorafenib, was about 13 months overall survival median, and you see how we seem to be getting the tail.

Now what the final analysis will look like exactly, I couldn't say. But I think what's clear is you get an early and prolonged benefit. And this is in a chemo-free regimen. This is going to bring new hope to a lot, a lot of patients, including about 900 a day who are diagnosed with metastatic liver cancer in China alone. So it's a tremendous unmet need.

The regulatory authorities, when we provided this data, I mean, they literally are ringing our phones, saying, "Hey, how can we make this available to liver cancer patients in our areas?" So FDA put this in the real-time filing protocol, which allows us to continually submit data, and we look forward to a very speedy approval in the U.S. But we're also working very closely with EMA and China on rapid approvals. And we hope to launch in the first half in the U.S., but hopefully, in the second half, we have a chance to launch both in Europe and China. So we'll be following that really closely.

So this is a slide that just kind of demonstrates the breadth and the depth of our oncology pipeline, 45 studies ongoing. We've had a number of positive readouts. But there's a few things that are -- I think are pretty interesting to highlight. So these ones that are highlighted in this sort of yellow color are all adjuvant or neoadjuvant studies. And obviously, if we can make an impact in the curative setting, it's a tremendous benefit for patients, but it's also something that's very popular with payers and gives us a chance at a better reimbursement outlook. And so I think what's interesting, you can see long adjuvant, head and neck, renal. And so a number of important areas in addition to breast cancer where I think we're -- typically have been focused in the past.

Another way to look at this is through new molecular entities. So these are studies that with novel agents -- and these are all late-stage studies. So for example, in lung cancer, small cell lung cancer, there's a study with Tecentriq plus an anti-TIGIT molecule we call [tiraga] and -- plus chemo, and that study will be starting very soon a Phase III study. If you look over under breast cancer, there are novel studies with a number of agents, including a SERD, a PI3 kinase inhibitor and -- let's see what else is in there, and ipatasertib, which is a Phase III molecule with 3 studies reading out -- 3 pivotal studies reading out this year. So that's one that we hope to be launching next year and so on. So I think it's exciting to see the number of new targets that we're pursuing and what that can mean for patients as well.

Now I'm just going to finish out a few comments on the hemophilia franchise. So this is basically HEMLIBRA. We're excited now we have Spark, we have an opportunity to bring gene therapy. But today, patients are increasingly benefiting from HEMLIBRA. And essentially, you see more and more patients making that choice. And we had tremendous growth throughout the course of 2019. We really don't see any sign of this slowing down in 2020. If anything, we see sort of a momentum effect where once patients have been on a while and they start to tell their friends in the hemophilia community, there's a growing confidence in what this medicine can mean to patients and their families.

Okay. So in terms of the things to look forward to this year -- and this is a sampling because there were really too many to list. But I wanted to highlight a couple that are in ophthalmology because this is an area where we pioneered with anti-VEGF therapy with LUCENTIS back in 2006. And now we're really looking forward to 2 opportunities to go worldwide. The first is the port delivery system with LUCENTIS. And just as a reminder, we have several Phase III studies going on. We're in wet AMD, we hope to have readouts there around the middle of the year. We also are in diabetic macular edema, which is a large unmet need, and especially because you have younger patients. And so this idea of getting frequent injections in the eye for 10, 20 years, they really need a better solution. And so the opportunity to have a high sustained efficacy with twice-a-year refills of the device, we think, is really compelling. And we think this is a bit of a sleeper because it's not a new molecule, it's a device, and so it's something that maybe people have been sitting on the sidelines a little. But we're really looking forward to having the Phase III data and go into regulators and bringing this to physicians around the world.

And then finally, faricimab, which is Ang2 and VEGF, so 2 targets. And the goal here really is to get higher efficacy levels than we've seen with an anti-VEGF alone. We'll have -- we had promising Phase II data that we've shared. We'll have the first Phase IIIs reading out in Q4 in DME and in Q1 of next year in AMD. And so we're going to have really concrete answers on this. And I think you'll agree that if we're able to show substantial benefit over an anti-VEGF, this could be a really important medicine for the future.

So in summary, you can see a really excellent progress of the medicines that we've delivered in the past decade. I think we're really impressed to see 35% of sales in Q4 coming from the new medicines. That number will be over 40% by midyear of this year. And 3 new molecules or 3 new launches with satralizumab, risdiplam and the fixed-dose combination of Perjeta and Herceptin on track for this year.

This was the news flow in 2019. I think many of you have been following it as it came. But I think it's worth just looking back and seeing the progress we made, how many new approvals and, frankly, a lot of really important advances in pivotal studies. So a great chart. How is this year's look? It's a long list. And I guess that's one of the disadvantages of following Roche is you have to put up with long pipeline presentations. But we'll keep that problem hopefully for a long time.

And so we've got a number of really important approvals that we're looking forward to. I think probably the one that's maybe the most symbolically important is the Tecentriq plus Avastin, just in terms of its impact and the thousands and thousands of patients who are dying every year, and the opportunity to have really a prolonged sustained effect there is super exciting. But then in terms of Phase III readouts, you'll notice this list is even longer than last year's. And there's some really important things in here.

Three, ipatasertib pivotal readouts. Most of you may not be tracking ipatasertib, and so you might want to take a deeper look into what we're doing there. I think the etrolizumab product, which is the second one from the bottom, we've had in Phase III for a number of years. These are massive studies. We have Phase III head-to-head studies with Humira, with Remicade, we have in Crohn's disease and ulcerative colitis. And those studies are all going to be reading out over about a 12-month period, beginning in the middle of the year. This could be a major important advance in IBD, and we think also another molecule in the line of OCREVUS, HEMLIBRA, Tecentriq, that could be a really large medicine for Roche.

So anyway, a lot of exciting things. We're looking forward to it. But there's no treatment without a diagnosis, right, Thomas? And so I'm pleased to invite my friend and colleague, Thomas Schinecker, up to cover Diagnostics.

--------------------------------------------------------------------------------

Thomas Schinecker, Roche Holding AG - CEO of Roche Diagnostics [3]

--------------------------------------------------------------------------------

Thank you very much, Bill. Very excited to be here today. As Severin mentioned, I'm fairly new in the role but not new to the company. I've been with the company for the last 17 years in different roles in different parts of the regions, both in commercial roles and also product development organizations. So really excited because I see the great potential that Diagnostics has, as Bill mentioned. Diagnostics plays an important role, the great opportunities to continue to develop our pipeline.

Now if you look at the sales in 2019, we had 3% growth, as Severin mentioned. And this was impacted, of course, by the inventory reduction that we did also at the end of last year.

Now if I look at the different business areas, and I'll get to -- into more detail on a later slide, growth has been strongly driven by the Centralised and Point of Care business area and the Molecular Diagnostic business area.

Now if I look at the different regions, strongly, the growth is coming from the emerging markets, so Latin America and also Asia Pacific. Now you may have seen that growth last year in China was only about 5%. So the growth is lower in Asia Pacific because of the China situation. In all other markets, as I say, the growth was fantastic also in 2019.

Now if we look at in-market growth, in-market growth is still strong. So we see a much higher growth in-market when it comes to the sales growth in China.

EMEA, LATAM. So EMEA was 2%. Specifically on the Diagnostics side, we had very strong growth with 4% in very mature market where the market growth is actually very low. North America was 0%, and I'll get to it. The growth has been impacted by lower growth in Tissue Diagnostics and their coagulation monitoring business.

Now as promised, I'm going to go a level deeper, looking into different business areas. The Centralised and Point of Care business area, which is our biggest one, growth, again, driven by the immunodiagnostics business, but this is exactly the area where we reduced inventories. So without that, growth is still in a very good double-digit range. We have a decline in coagulation monitoring. This is predominantly driven by the new drugs that are available where there's no more testing needed. So this will continue to decline.

Molecular Diagnostics, we have good growth with 6%, strongly driven by the blood screening business and molecular point of care business. Here, we also report the sequencing business, which has had a double-digit growth last year.

Diabetes Care, in a more difficult market environment, has kept the sales stable with 1% growth. As you know, here, we have continuous glucose monitoring taking over the strip area. So here, we manage to actually compete very well in a highly competitive market. But of course, if you look at the overall diabetes care market, it's growing higher because of continuous glucose monitoring.

The Tissue Diagnostics business, you see with 0%. But looking at the reagents, we have a solid growth. And here, we expect much higher growth going into 2020 with more system placements and system sales going forward.

Now looking at the P&L, we were talking about the top line with 3% and the bottom line with 1%. Here, again, looking at the situation that we had with the inventory reduction, our growth on top line would be in the mid-single-digit range and the same in the operating profit line. And how did we do this? Well, if you look at M&D, R&D and G&A, we had very good cost control. So a lot of productivity measures were happening last year. We did have a higher increase in cost of sales. And this was driven predominantly by 2 things. First of all, the high instrument placements that we had last year. And this is great news because this will drive future growth, and I'll have a slide on that very soon as well, on the next slide, I believe. Now the second part was the IBD regulation, and you may have heard about that. So with the IBD regulation that has -- or is going to take effect in Europe, we have to re-register all of our products into that market. So that, of course, drives some of the regulatory costs, which we book into cost of sales.

Let me also point out the R&D line because we're spending CHF 1.5 billion in R&D. We are a company that strongly believes in innovation and the innovation potential in the diagnostics space. So looking at that and looking at all of our competitors, we are significantly outspending in that area, which is an investment into the future. So where are we investing in? We're investing in developing new systems. And here, you can see a typical large lab with pre-analytics on the bottom left side. You see on the right side here, you see the Molecular Diagnostics area. Here, immunochemistry, hematology, and here are the fridges. So example, after it's been worked on in the different systems, can then go back to the fridge, wait, if there's another request, it can go back out. So there's no human intervention. This increases quality, and this also makes this more scalable because as the volumes go up significantly, the hospitals cannot afford to hire more and more people to do manual labor. The people can actually focus on what's needed, and it's the patients. They can focus on the patients.

And you can see the instrument placements. We are significantly increasing the instrument placements. And this is driving the cost of sales, but this is an investment into the future. We have more than 100,000 of the large instruments placed in the world. We have more than 1.5 million point of care instruments placed in the world. So this is very important for us because this is where our innovative reagents run on. This is the infrastructure. And we have new systems in the pipeline, both on the pre-analytics side, which is the cobas prime, which can handle all kinds of sample types, also for cervical cancer screening but also in mass spectrometry, molecular diagnostics and the immunochemistry area with new products in the pipeline.

Now what is running on those different systems? So today, we already have the broadest menu in the industry. We have more than 500 assays available across the different platforms. And we have them available across many different disease areas, from infectious disease, oncology, women's health, cardiology, critical care and so on. And the key is to have the product menu available on single platforms because the labs don't have space for multiple machines. They need to consolidate everything on as few machines as possible. And this is exactly our strength. And this is the strength that we're building on also in the future because we're investing a lot of money into developing new assays that can really differentiate us going forward, even more than today.

And you can see some of the areas where we're doing research in. One example is in oncology. We have a marker in the pipeline, the early detection of hepatocellular carcinoma. So that's something that Bill was mentioning earlier. But also in other areas like in critical care, new markers to detect sepsis earlier. 50 million people die every year of sepsis. That means every 2.8 seconds, someone dies of sepsis. And sepsis is a disease -- it's not a disease per se, but it's a situation that a patient is in that can be detected much, much earlier and so that the doctor can react. So we have a number of different assays in the pipeline that can really change the standard of care and differentiate our menu. Same in the neurology area. We got breakthrough designation for all-time assays. We have assays in the pipeline for Parkinson's but also for skeletomuscular disease. So very key to drive that expansion.

And let me give you a couple of examples of recent launches. One is the Elecsys HIV Duo launch in China. So this is a combination of antigen and antibody. And with that, you can close the window significantly from the time that someone is infectious until the time that it's actually you can detect in blood. So that's the medical value behind it. And none of the large players, except us, actually has this assay. And specifically, we launched it in China. China is, in the future, still going to contribute 40% of our absolute growth. So this is important for the Chinese market, and there are no local players also that can offer that test.

Furthermore, we expand a lot in the blood screening area, another area of infectious diseases. So you can see that we have launched in 2019 Zika. We've launched Babesia. We have launched cobas pro. So all systems and assays in that market that help differentiate us. We've won a number of tenders across different countries around the world. At the moment, we have about 80 million blood donations worldwide that get tested by our systems. With these wins, we have additionally 11 million donations that will be tested in 2020. And why are we winning these tenders? We're the only player that can combine serology and nucleic acid testing from one company. And we can combine it with pre-analytics because no other company can connect a molecular system to a pre-analytics system. So really excited about this because with that, we keep blood safe in the world.

And how else do we keep blood safe? Zika, I mentioned that. And you may have remembered a couple of years ago, this was a real crisis with babies having really difficult situation because their mothers were infected with this virus. And this infection came by a mosquito, but also it could come by blood transfusion. And so we have launched it in the U.S. So in the U.S., in blood screen centers, they are screening for Zika, but also now we've launched in Europe. So we reacted very, very quickly. And we reacted quickly not only with Zika, we reacted quickly with MERS. We reacted quickly with Ebola and also SARS. And that's -- I think that's something that we're known for, to react quickly. And this is also what we have done in the situation with the coronavirus. And this is, of course, all over the news at the moment. But it's key that you can identify the patients quickly, that you can isolate the patients so that they can be kept away from the healthy patients, so they don't infect more people. And we have a combination of assays that help do this. Also in China, we're, of course, working very closely with everyone to support those efforts to make sure that everyone gets the support that's needed.

Finally, we have also launched a digital solution in the diabetes care space. We have these different glucose meters, and these glucose meters can be used in combination with strips. But with these apps, with these apps that you can do -- have nowadays on your iPhone, and with the camera, you can actually read the strips without using a meter. And this is specifically important for more emerging markets where the penetration of iPhones or smartphones is actually very high but maybe not so much the rest of the infrastructure. So very excited that we have launched this at the end of the last year as well.

So in 2019, we had a number of very key launches. So very excited about that. Even more excited, of course, also about the launches that we have in 2020. Let me just give you one example in infectious disease with EBV. So we have received breakthrough designation for EBV and BKV, and this is for transplantation patients. And why? Because many of us carry these viruses in us. However, these viruses are harmless as long as we're not immunocompromised. So it's very important that in such situations like with transplantation patients, you actually measure those viruses if they're in the body, so you can avoid complications. But we have many other systems and assays that we're going to launch next year. That's just an excerpt. So I'm really excited about the possibility that we have, both on our portfolio and pipeline in the systems, but also in our biomarkers because we really can change the standard of care with diagnostics.

With that, I hand over to Alan. Thank you.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [4]

--------------------------------------------------------------------------------

Thanks, Thomas. Thank you. Welcome. Great to see you. It's been a -- really been a solid year for Roche. Let me lead you through a couple of figures now.

Let me start with the overview. And when you look at it, sales growth, 9%. We had an impact of CHF 1.5 billion negatively from biosimilar competition, CHF 1.3 billion in Japan and Europe and then roughly CHF 300 million -- sorry, CHF 1.2 billion in Europe and Japan and CHF 300 million in the U.S. And you know we have given guidance basically on the impact in Europe and Japan, and we came out really, really on spot. So we're very happy with that.

The core operating profit up was 11%. You know really there was a major headwind on the royalty side from the loss of the Cabilly patent, CHF 708 million, CHF 705 million constant rates. So I think we mastered that and overcompensate. I will come back to that.

The core EPS growth at 13%. What really gave us is additional dynamic in the core EPS compared to the core operating profit taxes. We had 2 tax cases that we resolved, very old ones, 15 years old, and they gave us a little bit of a boost here.

Dividend in Swiss francs, I think Severin mentioned that. The cash flow, very happy with the cash flow at CHF 20.9 billion, and we'll show the development later on. Broad net debt to minus CHF 2.5 billion, so it's CHF 3.1 billion reduction in spite of doing the Spark transaction. And then -- which I think is pretty outstanding compared to the previous years is that we brought gross debt down, now from CHF 18.8 billion to CHF 14.4 billion. You might remember the last years, we were always hovering around CHF 20 billion and then CHF 18 billion, but there was never a major step. I think this year, we made a major step. And I will come back to a point called a bond redemption.

And the net financial result, we'll explain that. Two elements. On one hand, lower income from equity securities, and we'll tell you the story about it; and then the early bond redemption, CHF 202 million that we brought into the core results. The net income up 32%. Major impact is certainly the operations and how much we have improved there. The other piece is lower impairments compared to last year.

Here's the overview. I think 9%, I've mentioned, my colleagues did a great deal, Thomas and Bill, to explain that. Then the 11% core operating profit, I will go through the P&L. You see core net income has a higher dynamic. That's basically taxes. Then you see the core EPS losing a little bit dynamic, just growing with 13%, that's sugar. Sugar had a tremendous year, and that's even more dynamic in their profits compared to the dynamics that we have. And we have to take the 40% out. We own 60% on sugar, we take that out, and that brings the dynamic in the core EPS a little bit down.

Good. IFRS net income, I made a comment on that one. I will come to the question later on. Look at the CHF 20.9 billion, and I'm pretty proud about that number and the free cash flow at CHF 16.7 billion.

Here's the bridge. And I think the bridge is quite important because it tells a little bit the story of the year that will even give you the impacts first half and second half. So we went from CHF 18.3 billion -- CHF 18.3 to CHF 20.64 at the end of 2019. The first point to make is the gains on the product disposals. And you might remember in the first half, we had a little bit of an uplift coming from that. We had CHF 437 million in gains in. In the second half, we had CHF 62 million. So in total, roughly CHF 490 million, which is a little bit of a boost compared to last year. And you see really the impact here of plus 1.3 percentage points.

Royalty and other operating income, that's certainly excluding the gains of product disposals. This has been Cabilly. Cabilly is the major point here, and Cabilly is a bit of a reversal. We had an impact of minus CHF 271 million in the first half and then lost more, roughly CHF 430 million, in the second half. So I think really you see where we had an acceleration of that. And that has cost us minus 2.8 percentage points dynamic.

Then the gains of the equity securities, it's a bit of a funny story. We have a minority in a company called Allakos. And Allakos from end of 2018 to end of 2019 has basically doubled the share price. But it was a bit volatile, right? I think it came down in the first half and then came up again in the second half. The really -- the real difference maker here is -- the minus 0.6 percentage points is Avexis. In the older company, we held a minority, and then Novartis came and bought it, and we had a nice gain in our numbers, unfortunately, in 2018 and not in 2019. So we're missing CHF 100 million here, which explains the minus 0.6 percentage points.

The bond redemption happened in the second half, CHF 202 million, gave us a minus 1.1 percentage point impact. And then the resolution of the tax disputes. You know we had an impact in the first half. In the first half, that was the very old case, 15 years old impact, plus CHF 242 million. And in the second half, we had another impact from another tax case, plus a CHF 213 million. So these 2 together, minus -- plus CHF 454 million, they gave us a positive impact of plus 2.9 percentage points, and the rest is operations.

Good. Quickly through the P&L. 9% is explained. Royalties and other operating income was a minus CHF 375 million, that's Cabilly. That's Cabilly. Now you might ask yourself, okay, where's the positive impact coming from. This is, on one hand, the product disposal that I've explained already, and which was larger than last year, CHF 170 million more. And the other piece is really about VENCLEXTA. VENCLEXTA really gave us quite something, and some of you even had it in the papers. I say, we had a CHF 176 million increase, '18 to '19, from VENCLEXTA alone in that line positively.

Cost of sales, CHF 1 billion up, CHF 1.050 billion to be precise. It's tremendous how the volume has increased. Pharma had a volume increase of 16%. Diagnostics had an increase of 4%. I think really having a cost increase of 7%, I think, really shows how well we have dealt with the costs in that line.

M&D, I think Bill made the point. You have seen basically no increase in Thomas numbers. So that's driven by Pharma where we really use the opportunity to invest into the pipeline in launching the new products. R&D, up CHF 633 million, basically same here, driven by Pharma. And then G&A, an increase of CHF 151 million, a little bit with some service costs from last year. We had a positive impact in '18, which we didn't have in 2019, but the vast majority is really coming from administration. That's really coming from the acquisitions, Flatiron, FMI, et cetera, that we have done.

Good. Leads me to the royalties. And really, that explains Cabilly. I've mentioned that already. You see royalty income and others, out-licensing income pretty stable. Then other operating income, that is really the profit share with (inaudible) staff, which is driving that for the U.S. And then we have the gains and losses of the product disposal of CHF 170 million that I've explained already.

Good. Margins look pretty good. I think we had a slight increase in the group. We have a slight increase on the Pharma division. I think if we didn't have done the reduction of inventories in China, I think Thomas would have had the same margin than the year before. So I think we look pretty good here.

When you then look really at the core net financial result, a CHF 300 million deterioration, if you like. Well, the bond redemption, that's certainly something which will help us in the future, and I come to that on my next slide. Equity securities, nothing else than Avexis that I've explained before. Net interest income is a nice anecdote because as we've waited for the closure of the Spark transaction, certainly, we held all these billions of U.S. dollars available and reinvested it in short-term assets or assets with short-term returns, and that gave us a plus CHF 68 million. I admit, it's not our core business but we take it, so that's a positive here. A small impact from currency and some other points, so you see really the deterioration here a little bit by design.

Good. The bond redemption itself, don't want to dig too much into it. We bought back CHF 2.9 billion, which is a very significant number, helped us to bring the gross debt down. I think it's fair to assume steady state, certainly, that we will have a reduction for interest expenses in 2020. So that will be another help in the year -- in this year.

Good. Group tax rate went from 19.7% to 16.3%, so really a significant reduction. And certainly, the 2 old tax cases that we have resolved, one 15 years old -- and we can argue whether this will come back again or not. I think I don't expect that to happen in 2020. So I think really the 18.4%, this is really what we should have on our mind. And I think really a tax rate around 18%, this is what I would guide for, for the future.

Noncore and IFRS income. So let me lead you through that. The core operating profit, up CHF 2 billion, as you can see, to CHF 22.5 billion. I think quite an impressive EBIT. When you look really at the global restructuring plans, they came up with roughly EUR 300 million. And these are really projects that are going really across the company with a lot of transformation efforts. So we are constantly working on our financial flexibility, and that allows us to make all the investments that we're doing. So -- but you see really, I indicated to you, I think, whenever the charge is around CHF 1 billion, I feel pretty good with that. So I think we achieved this.

Amortization of intangible assets went up CHF 238 million. I think the major point here is really the revaluation of Esbriet. We brought Esbriet in the valuation a little bit up as we brought the goodwill down last year of InterMune. So I think that's a point here. Then impairment of intangible assets, a reduction of CHF 1.6 billion, so in fact, a positive impact in the comparison. That's really, well, InterMune, which accounted for CHF 1.8 billion last year.

M&A and alliance transactions, a positive CHF 43 million. You might be surprised, why is that a positive? Well, this is the release of contingent considerations of provisions, for contingent considerations for companies and goodwills that we've impaired. So that comes in up positively. And then we have legal and environmental with a minus CHF 480 million, an increase of CHF 300 million. I would call out one case here. That's the Meso case. I think it's page, I think, 89 in the finance report. Old case about an ECL technology. We went for that case for quite a long time. We lost the court case. We will appeal, but we made a provision of roughly CHF 200 million.

So all in, you can see an improvement at CHF 800 million compared to last year, and that brings the IFRS operating profit up by roughly CHF 2.8 billion. Then you see the total financial result and taxes was an improvement of CHF 464 million. Here, we have even 3 tax topics in. The 2 that I've mentioned before with the plus CHF 445 million, then we have an additional one, which we had at half year already, the change of the Swiss tax reform where we reevaluated our deferred tax assets related to this. And it doesn't mean that we pay less taxes in Switzerland. It just mean we had to look once again at our tax assets, the deferred tax assets, and that gave us a positive here of CHF 232 million. And then certainly, the bond redemption went against that. So you end up then with the CHF 464 million. And that means the IFRS net income goes up CHF 3.2 billion, which is 32% at constant rates.

Good. Cash, as I said, I think, a great year here. Look at the 29 -- CHF 20.9 billion. Look at the CHF 20.5 billion coming from the Pharma side. And then you really see the Diagnostics division. It looks like a little bit weaker with the CHF 963 million, but they did 2 deals. The first one, the in-licensing deal, will definitely help Thomas in the future. That's a good thing. We will see that positively in the P&L. And then the Santa Clara, we bought a site in the U.S. I think if we didn't have done these 2, I think, well, we would be on the level of last year, to well explain of.

Good. I think when you look at the cash flow overall, I think what is exciting to me is how well we did on the net working capital side because what you could have expected with all growth that we have had, that we have very high inventories, accounts receivables go, go, go through the roof, I think that was not the case. I think on group level, the inventories went even down. So I think really we did good. And we're pretty good structurally on inventory, so I think the supply chain did a great deal here. Accounts receivables went up, as you would expect, with the high sales growth. But at the same time, accounts payable went up and compensated for that. So very happy with that.

We had a little bit less investment in PP&E. Then you know that IFRS 15 that we now have to show the lease liabilities paid that was formerly in the core operating profit. And then we have more investments in intangible assets. Yes, you might have seen that we have roughly invested into intangible assets CHF 1.5 billion, CHF 1.6 billion, I think when you really look at the balance sheet. In the cash, it's CHF 1.5 billion. The difference is what we have not paid yet. So -- and that's the right spot to put our money, no doubt about this. But still, I think, went up with CHF 20.9 billion in operating free cash flow. That's a track record, which I really appreciate, and that's determining our strategic flexibility moving forward. So I would say the stage is set.

What happened to net debt? As I mentioned at the beginning, we went from minus CHF 5.7 billion to minus CHF 2.5 billion. So we are still -- we still have net debt. We're not net cash positive. Operating free cash flow explained. We paid taxes, CHF 3.5 billion. A little bit of outflow certainly for interest expenses on the treasury side. And then the dividend, certainly, the dividend for 2018 paid in '19, and then certainly Spark is in that number.

Good. Balance sheet, very quickly. On one hand, you see cash and marketable securities goes down from CHF 13.1 billion to CHF 11.9 billion, and certainly, that's the money that we have put into that bond transaction, buying back bonds. And you see on the other side, when you look at the noncurrent liabilities, they go down. And this is really why we have reduced our long-term debt.

The noncurrent assets, that's a bit special because we closed Spark at December '17, right, in the range that we gave in our guidance. But what we did is there was not enough time to really come up with a great purchase allocation. So basically, what we have done, we put that whole thing into the noncurrent assets. And now we have to look at it, what are the intangible assets, whatever. And we expect to finalize the purchase accounting until half year 2020. Formally, we have to finalize it at the end of 2020. I think we can do it until half year, so the numbers will shift a little bit here. Equity went significantly up, and the equity ratio is now at 43%. As I've said, the stage is set.

Outlook, currency, we came from half year, and it's quite interesting when you look at it. Here is the Swiss francs to U.S. dollar ratio. Here is the Swiss francs to euro ratio. And when you look at half year, so June, plus 3%, minus 3%, impact 0. And then what happened is the U.S. dollar remains strong but just a plus 2%. And then you see the minus 4% here, and then you see the negative impact, which came. So I think the U.S. dollar is still going strong, but the euro went even weaker. And that gave us then the negative impact on sales, core operating profit and core EPS. And you know our modeling if we remain all currencies stable at year-end 2020, how would 2020 look like? It would be a minus 3 percentage point FX impact on sales, core operating profit and core EPS. A very unlikely case, as we all know.

Good. Let me set the stage for 2020 and for your predictions by giving you the adjusted number for the core EPS 2019. So the core EPS as reported is CHF 20.16. And you know what we are taking out is the FX losses. You find the FX losses in the finance report on Page 59. Page 59, you find the CHF 205 million. We have an 864 million shares. So you divide -- now before you do that, you first take the taxes away. That's a good thing. 18.4%, I think we agreed on that. When you take the 18.4% away, it's a CHF 37.7 million charge, and you end up with CHF 167 million. This is the number you divide, so the 864 million shares. And then you end up really with the adjustment of plus CHF 0.19, and that leads you to CHF 20.35. And that's the basis for your projections.

Good. Here's the guidance. Let me make one point about the guidance right away. I think this guidance looks even more ambitious when you take into account that we expect a biosimilar impact in 2020 of roughly CHF 4 billion negatively. So that's really the number that we have on mind. And I think really when you take that into account, I think that's really -- that's a great guidance to have and quite an opportunity for our company. But as said, the stage is set.

Thanks. And now we are, I think, expecting the questions. Correct? Very good. Thanks for your attention.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [5]

--------------------------------------------------------------------------------

We'll just continue. (foreign language) Shall we directly continue? Or do we want to make a break? Is it fine to continue? Okay. Good. Excellent. So can I ask my colleagues to come up to the table? And can we take the first question here?

================================================================================

Questions and Answers

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [1]

--------------------------------------------------------------------------------

Do we have a mic?

--------------------------------------------------------------------------------

Richard Vosser, JP Morgan Chase & Co, Research Division - Senior Analyst [2]

--------------------------------------------------------------------------------

Richard Vosser from JPMorgan. Two questions, please. First question, thanks for the biosimilar impact. But thinking about some of the international operations on some of the products, obviously, weakness in China because of the destocking. So how should we think about the outlook of Perjeta, Herceptin, Avastin, Rituxan in the international operations? Specifically Perjeta on the NDRL, should we think of the bolus grade like Herceptin did? And then how should we think of the others, Herceptin, Avastin, Rituxan? Second question, just going back to the -- or going to the net debt to total assets. It's now considerably below, I think, the original target range, whether that was 10% or 15%. So how should we think about your putting to work the capital to restore the balance?

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [3]

--------------------------------------------------------------------------------

You want to start?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [4]

--------------------------------------------------------------------------------

Yes, sure. International operations and -- you said Perjeta, Avastin, Herceptin, MabThera? Okay. Now I'm going to add Alecensa because Alecensa and Perjeta were both added to the NRDL in Q4. And although the prevalence of ALK positivity is not necessarily higher in China, there's just -- there's a tremendous amount of lung cancer in China and so the number of patients that can benefit is really quite large. So we think we'll still have -- we think we'll have some underlying growth, although most of the growth in MabThera and Avastin and Herceptin in China has probably already happened. But Perjeta and Alecensa should be good growth drivers since they just received the listing, and that does dramatically expand the number of patients who are eligible and can receive it.

And then otherwise, in the rest of international, there's some gains and losses. There are some biosimilars or noncomparable biologics in some markets. And so we have some negative markets and then we -- other markets where it's really more of the store -- the effect of Perjeta and Kadcyla getting into the adjuvant therapy. Also, products like Alecensa and other products that we've launched in the last decade that are really just starting to penetrate some of those international markets. So we think it should be a reasonably good year in international. 15% is hard to replicate, but we feel good about it overall.

--------------------------------------------------------------------------------

Richard Vosser, JP Morgan Chase & Co, Research Division - Senior Analyst [5]

--------------------------------------------------------------------------------

Net debt to total assets?

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [6]

--------------------------------------------------------------------------------

We're now at 3%. So justified point. Well, I think we have set discussions since I joined the company 9 years ago. And every year, I think there was this concern we would become net cash positive. I think so far, we've put our money to work and really found solid acquisitions, good acquisitions, I would even say, even in some cases, which gave us good returns. And I think really in a period where we have -- where we have a good growth, I think we will find opportunities here to move forward here.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [7]

--------------------------------------------------------------------------------

That's a good problem to have. And we'll cross the bridge when we have to budget, right? Good. Please.

--------------------------------------------------------------------------------

Sachin Jain, BofA Merrill Lynch, Research Division - MD [8]

--------------------------------------------------------------------------------

Sachin Jain, Bank of America. A couple of questions, please. Firstly, on the CHF 4 billion, just another part of it. What portion of that is the U.S.? And what's your level of visibility on the erosion rate you've assumed there and level of confidence around that given there's not a lot of precedent as it sits? And second question, you've referenced a couple of launches through this year, risdiplam, satralizumab, Tecentriq in liver, across all regions. To what extent are those launches reflected in guidance versus offering upside? Background to the question, obviously, is your prior launches, OCREVUS, Tecentriq, HEMLIBRA, drove upgrades through the last couple of years. And then the final question is on gene therapy. No mention of Sarepta, so I wonder if you could talk about that licensing deal towards end of the year, your level of excitement and broader gene therapy intentions beyond that deal.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [9]

--------------------------------------------------------------------------------

If I could make the point, it's important to us that you say roughly CHF 4 billion.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [10]

--------------------------------------------------------------------------------

Yes. That's what we're guiding for.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [11]

--------------------------------------------------------------------------------

Yes. I mean, we really nailed it. I think the point is we really nailed it in 2019. We said the impact would be about $1.3 billion and it was $1.3 billion. We're not always that good. And so...

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [12]

--------------------------------------------------------------------------------

Roughly.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [13]

--------------------------------------------------------------------------------

Yes. We think it's roughly CHF 4 billion, we think -- of which, I think, as Alan said, we think the impact ex U.S. will probably be similar this year as last year, which was about $1.2 billion, so the rest is roughly -- the rest is the U.S. In terms of the visibility on that and how confident are we, I mean I think there's a fairly decent range in it because it's still early days. So keep in mind that there were 2 products launched: one biosimilar to Herceptin; and one to Avastin in late July. They didn't really get momentum with contracts and all that sort of thing really until the fourth quarter. And then there have been several other launches since that time, just in like November, December. There's additional launches happening kind of in Q1. So it's still pretty early. I think what we would say is it's consistent with what we've said all along. We expect the impact in the U.S. to be similar to Europe, maybe a little better, maybe a little slower. I think it's really very -- that's still our outlook. And yes, so far, that's been it.

You asked about whether the launches are in the guidance, is there a potential upside? There's -- we always like to think there's upside. I think if you consider the launches we're talking about, risdiplam is likely late first half. I think we've said May PDUFA date. So if we launch in May, its ability to drive a massive upside in the year, given that it's not given up front. OCREVUS, one of the effects we had there is that as soon as a patient's dosed, they just got 6 months of therapy. Risdiplam is dosed daily. So I think with risdiplam, with satralizumab coming later in the year, with the HCC, is a big indication. But the biggest demand for that will likely be outside of the U.S. because liver cancer is relatively less common in the U.S. and the first approval is likely to be in the U.S. So all that to say is while we -- while there is a range around those launches, I don't think they're going to -- I don't think they're going to be a big variance driver in 2020. I think it's really more the underlying strength of the ongoing launches like Tecentriq, OCREVUS, HEMLIBRA continuing. So that's the answer.

Sarepta, we're super excited about. I think the timing, that's going to depend a little bit on what the further development program looks like. There's been very promising data. I mean remarkable results in just a few patients, the kind of results that you just wouldn't expect to see in Duchenne's muscular dystrophy. That's what made us so excited about it. We're -- we remain incredibly excited about bringing that to patients around the world. But it's not a -- we don't think that's a 2020 phenomenon. And then finally, our intentions in gene therapy are -- we have big intentions. In fact, I've spent a significant amount of time, several of us have, with the leadership at Spark already this year, sort of helping plan how do we use the resources, the worldwide resources of Roche to help accelerate the programs they have ongoing, additional investment for Spark in Philadelphia to extend into new disease areas? And I think we expect big things about -- from Spark in the future. But again, these are more -- the majority of those things are new concepts that we're investing in, and so they won't have impacts in 2020.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [14]

--------------------------------------------------------------------------------

Let me just bring it over here.

--------------------------------------------------------------------------------

Michael Leuchten, UBS Investment Bank, Research Division - Co-Head of Pharmaceuticals Research of Equity Research [15]

--------------------------------------------------------------------------------

It's Michael Leuchten from UBS. One question on the inventory reductions in China, timing and rationale, both for diagnostics and pharma? And then, Bill, in your commentary around hematology, you didn't mention Polivy, the POLARIX trial for 2020. Is that still a data point for this year or not?

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [16]

--------------------------------------------------------------------------------

Do you want to talk about inventory?

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [17]

--------------------------------------------------------------------------------

Yes, sure. So we decided in Q4 also to reduce the inventories in China. We see a lot of pressures in China in terms of the 2-invoice policy. You may have heard about that. So basically, most of our business in China goes through distributors. And the Chinese government has implemented this 2-invoice policy, which means that between us and the end customer, there can only be 1 distributor and this used to be very different in the past. So what we did, we did the prudent step and we said, okay, let's reduce inventories rather than get into negotiations with our distributors. I think that's the best thing for the future.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [18]

--------------------------------------------------------------------------------

And in the pharma area, we went from 26 days to 16 days. And largely, what created that was our volumes increased substantially. I mean we had a 38% sales growth. Our volumes were even significantly more than that. And we reevaluated how much safety stock we needed in the channel because when you have higher volumes, you don't need as many days. So we basically -- we thought about it, and we said we might as well take that as we come to the end of the year and go ahead and line it up in a place where we'd like to be.

The Polivy study you mentioned, so this is the study of Polivy with R-CHP, which -- instead of R-CHOP, which is the standard of care in diffuse large B-cell lymphoma. It's in the curative setting. It's a big deal if we can replace the vincristine with Polivy and drive what we hope will be better tolerability and a higher cure rate. And so this is really the big study for Polivy. And the reason I didn't mention it is because we'll only get the results and I think it's like end of the year, maybe Q1 of 2021. So it's possible it will happen in 2020, but maybe more likely, we'll be talking about it around this time next year or shortly after. Thanks for the question.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [19]

--------------------------------------------------------------------------------

Good. I suggest we go over to this side. Yes. Very good, please. Thank you.

--------------------------------------------------------------------------------

Mark Douglas Purcell, Morgan Stanley, Research Division - Equity Analyst [20]

--------------------------------------------------------------------------------

Mark Purcell from Morgan Stanley. A couple of questions. Just in terms of, firstly, risdiplam, the opportunity and what we should think about from an uptake perspective. We understand that about 40% of patients diagnosed with SMA in the adult setting don't like to take the drug because of intrathecal delivery. I think, Bill, you mentioned that. So is there a sort of big sort of unmet demand here? And how do you see diagnostics playing a role in growing that market?

The second thing is on OCREVUS. Obviously, there's incoming competition but you have plans to reduce the infusion times. You also have your own [subcut] program as well. So could you help us understand that the plans and timing behind defending and building the OCREVUS franchise?

The third one is on the bispecific, the CD20/CD3. You said in the press release, there's a decision to be made soon. The 2:1 may be a better refractory option, fast route to market, the 1:1 maybe safer, better for earlier-stage patients. So it's obviously a competitive environment. You're currently ahead. You may not stay ahead. So just what's going on there? And how should we think about that opportunity?

And the last one is just a quick one. Liver cancer, Tecentriq in combination with Cabometyx. The data has come out in Q3, the COSMIC-312 trial. So how do think about having Avastin and Cabometyx as partners on top of Tecentriq in this setting? Will there be patients or countries or opportunities where one might be an advantage versus the other?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [21]

--------------------------------------------------------------------------------

Great. That's quite a list. So let's start with risdiplam. Yes. I mean, it's fascinating a field like spinal muscular atrophy, where essentially, there were no products, what, 4 years ago, there was no medicines. And now we're actually in the wonderful position that there's actually 3 choices or soon will be 3 choices. And they couldn't have more different modalities with an intrathecal administration several times a year versus a gene therapy, which is onetime but has certain limitations in terms of age, in terms of viral antibody positive, et cetera, and then an oral daily. It seems like each of the medicines is very efficacious. So they have that in common but that there's very different patient populations. So type 2 patients, often, they don't present with symptoms until they're toddlers or older. Type 3 patients sometimes don't present until later yet. And then there's also this complication around what benefit can you provide? In other words, if neurological damage has happened, can you repair that? Or are you really just preventing further damage? And so -- and we and the other companies have been trying to figure out as much of the stuff as fast as possible in types 1, 2 and 3, all at the same time. So I would say that I think there's -- it's going to take another decade or 2 before all of this is worked out and like which types of patients should get which medicines in which setting. But I think there's a large opportunity for risdiplam today because there are many patients who aren't going to be eligible for one of the other therapies. There is the fact that with an oral therapy, you have an opportunity. From a payer standpoint, you've got an immediate ability to have a therapy. You don't have a big price tag upfront. There's the fact that we will have data. We'll have more type 2 and type 3 data than the other therapies. We've got patients out and even including patients in their 20s in the SUNFISH studies. So I think overall, we're going to have a really good regulatory package and a great medicine to bring. And I mean, I haven't been very good at predicting the uptake of -- I didn't know -- I knew OCREVUS was going to be really big because -- but I'd say HEMLIBRA has surpassed all the market research we did. Normally, you do market research, and then you adjust it down because market research is sort of a bias. You're asking about this thing, and then they sort of -- but with HEMLIBRA, the actual uptake has far surpassed the market research.

On risdiplam, I think it's a difficult 1 to call. And I'm glad we have a lot of really excellent analysts here who will help us with that, but I think it's going to be very meaningful. OCREVUS, we have completed faster infusion study. We know it can be dosed about twice as fast with an IV infusion. Right now, it's usually 3.5 to 4 hours, can be more for a first infusion. And we think we'll be able to deliver that dose in 2 hours. So if you think about it, a 2-hour infusion twice a year, which is about as often as an MS patient sees their physician. It's a pretty good way to deliver a very efficacious and well tolerated therapy. We're just going to be going -- we're not going to be competing per se. We're going to be going out and talking to people about the benefits of OCREVUS and what it's doing for primary progressive patients, relapsing patients, the impact it has on the long-term with disability progression, with the well-characterized dosing. And we think that OCREVUS is going to continue to grow for a long time.

CD20/CD 3. I'm not going to say very much about it because it's still very much in the works. It's a very exciting field. We have 2 -- looks like 2 really excellent molecules. And they have different sorts of attributes so far, what we're seeing. But we haven't -- I don't think we've maxed out yet on the efficacy of mosun and we haven't hit -- we are continuing to work with the dosing profile of the 2:1 and continuing to get gains on tolerability. So it's hard to call and say, oh, mosun is going to be the front line and 2:1 is going to be refractory because those are -- that's a plausible hypothesis. But I think it would be prejudging. We'll follow the science, and we expect to maintain our lead, though. And then liver cancer. You asked about Tecentriq plus Cabo. It's all going to be about the data. Let's see it. We welcome it. Tecentriq is an excellent cancer immunotherapy. Cabo is a really compelling medicine with several applications, and we can't wait to see the combo.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [22]

--------------------------------------------------------------------------------

If you just continue in this whole perhaps -- just hand over. Good.

--------------------------------------------------------------------------------

Richard J. Parkes, Deutsche Bank AG, Research Division - Director [23]

--------------------------------------------------------------------------------

Yes, Richard Parkes from Deutsche Bank. Firstly, obviously, thanks again for quantifying the biosimilar impact this year. I wonder if you could commit to whether you think this is definitively going to be the year of maximum pressure from biosimilars. I know you've obviously got European Avastin biosimilars launched this year. I just wonder how you think about that erosion. And maybe you can give us a number of -- what that number would be in 2021? Or maybe that's too cheeky.

The second question is just on Tecentriq in adjuvant lung cancer. I think at the Pharma Day in September, you had those 2 trials potentially reading out this year, but they haven't been mentioned in the presentation. So could you just talk about where we are in terms of interim analysis there? And where you think you are versus the competition in adjuvant lung? And then final question on HEMLIBRA. You've highlighted the strong initial uptake in the initial European markets and the noninhibitor setting, but the sales at the moment look quite modest. So could you talk about the reception and then compare and contrast the uptake and the reception from physicians to the U.S. experience?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [24]

--------------------------------------------------------------------------------

Yes. So the first question on biosimilar impact. I mean, I think it's been evident in our modeling for some time that the year of maximum impact would likely be '20 or '21. Now I haven't answered anything because you sort of had that in your -- honestly, I mean, I'm going to give you an answer that might seem sort of obvious, but it sort of depends on what the ultimate impact is in '20, whether '20 is the maximum. If it's bigger in '20, then it's going to be less in '21. And so -- and the roughly, I would heartily agree with. So let's wait and see. Maybe by midyear, we'll have a better idea of whether it's -- but in either case, I think what's most exciting is with the trajectories from the products we've already launched and the additional ones we're adding, I think we get increased optimism that we can continue to grow through it.

Tecentriq and adjuvant lung. Yes, the trials are all on track. There's a potential, depending on the level of efficacy, that we could hit an early readout in 2020. And we've been anchoring on the base case with the final analysis in 2021, but it's possible. And we think we're at least equal or ahead of competition, so we look forward to the results there. These things are somewhat unpredictable, but we hope we'll have a good benefit there.

HEMLIBRA EU uptake. Yes. In noninhibitors, because the unmet need is not as high, obviously, as the inhibitor patients, I think the Europe -- the system, the health care system in Europe has been a little slower to respond to it. We see all the same dynamic in Europe as the U.S. in terms of the attractiveness for patients. We don't see any fundamental reason why it won't do as well in Europe as the U.S. ultimately. But there's been just maybe a little slower start in some major markets. Germany is one. We think there's some ways about how the factor is reimbursed. That gives pretty big economic incentives to the hemophilia centers. But we think that the compelling profile of HEMLIBRA will ultimately prevail in that.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [25]

--------------------------------------------------------------------------------

Let's just move on.

--------------------------------------------------------------------------------

Sam Fazeli, Bloomberg Intelligence - Research Director of Europe, Mideast and Africa [26]

--------------------------------------------------------------------------------

Sam Fazeli from Bloomberg Intelligence. Just 3 questions. One, following on from Richard's on Tecentriq in adjuvant. Obviously, we had the failure of the monotherapy in bladder cancer. Is there anything you want to comment on in terms of what that might mean for -- obviously, we all hope that the lung and renal monotherapies will read out positive? But obviously, it raised the question about what -- whether the right term -- test is being conducted there.

And on the biosimilars, just for some of us who cover some of the other companies that do -- are launching the competitors, could you give us a feel for what your expectations are between volume and price impact? And no specifics, but if there is a split, 50-50, what are you thinking in terms of the reduction? And then lastly, I saw a headline pass with the risdiplam pricing. I'm not sure, I'm not going to attribute the word aggressive to anyone because I didn't actually hear it. What does that mean? Right. But what does that mean? How does that -- we know what the price of the current drugs are roughly per annum on their lifetime, et cetera. What does -- give us some color, if you may, please.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [27]

--------------------------------------------------------------------------------

Yes. And just to clarify, so when I was asked about that, I talked about what we've done in other medicine launches and somehow it ended up that we're going to be aggressive there. So we're really -- we don't typically comment on the pricing of a medicine before we launch it. I think the only thing I would say is that we will stay true to our values, which is we want to make sure that it's perceived as a good value, not just by us. It's always easy for -- yes, that's easier for us to perhaps perceive it. It's really important for us to talk with the patient families, the insurance companies and see what they believe is the value, and we'll make a good decision on that. Biosimilar. The impact will be primarily on a volume impact, not price because we -- yes, we don't have major changes on our prices. And so the main impact is that we lose the business.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [28]

--------------------------------------------------------------------------------

And Tecentriq (inaudible)

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [29]

--------------------------------------------------------------------------------

Yes, sorry. You started with that, didn't you? So...

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [30]

--------------------------------------------------------------------------------

You're talking U.S., I assume.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [31]

--------------------------------------------------------------------------------

Yes. So there are some countries where statutorily, they force a price cut when a biosimilar launches or when the patent expires. And so some of that has already happened. But then once the biosimilars enter, then it tends to be more of -- they come in with big discounts. And yes, we lose the volume.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [32]

--------------------------------------------------------------------------------

And the number, I mean we can provide the number roughly for the half year in Europe after 2.5 years in the market, we have to make an impact of like 1/3, maybe it's priced at 2/3 (inaudible)

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [33]

--------------------------------------------------------------------------------

Okay. And then you asked about the -- so the metastatic -- sorry, the muscle-invasive adjuvant study for bladder cancer with Tecentriq. We recently announced that we didn't hit the primary endpoint. And so of course, we have that same question. What does this seen about -- is this a good prognostic, an accurate prognostic for the opportunity to do good in other adjuvant forms? And we also asked that question beforehand because it's always important to get the answer before the result's in. And basically, what we were told is, hey, I wouldn't put too much weight on the bladder cancer study because bladder cancer, even in the metastatic setting, there's a lot of patients who don't benefit. And so I think that lung and melanoma, those may be more of, I guess, the bellwethers for the impact of cancer immunotherapy in adjuvant, and that's why we're very anxious to get the lung result.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [34]

--------------------------------------------------------------------------------

There's also 1 question on the phone.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [35]

--------------------------------------------------------------------------------

Okay. So perhaps we can take a question from the phone in between. Can you connect us?

--------------------------------------------------------------------------------

Operator [36]

--------------------------------------------------------------------------------

The first question from the phone comes from Tim Anderson from Wolfe Research.

--------------------------------------------------------------------------------

Timothy Minton Anderson, Wolfe Research, LLC - MD of Equity Research [37]

--------------------------------------------------------------------------------

A couple of questions. The -- just on Alzheimer's, you have gantenerumab and the DIAN 2 trial. And I'm wondering if you can give us your perspective on that as something that should be reading out very near term. Do you view that as a trial that's extraordinarily high-risk or a medium level of risk or what exactly?

And then another question on etrolizumab. Bill, I think I heard you draw parallels between that and some big recent launches like HEMLIBRA and OCREVUS. Does that signify that you're highly confident in the outcome of those results? Or is that simply just a reflection of what the theoretical market opportunity is for a product like that without necessarily meaning that you're highly confident in the outcome of the Phase III trials?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [38]

--------------------------------------------------------------------------------

Alzheimer's -- so the DIAN study is a -- it's an independent third-party group that's running a study in familial Alzheimer's. It's a rare inherited form, and it includes a medicine from Eli Lilly as well as gantenerumab. And I think there's -- I want to say 87 gantenerumab patients in the study. So I mean by definition, it's an exploratory study. In fact, most of these patients, they started on the low dose, which is about, I think, 1/4 of the current dose. And so I think based on a number of those factors, we think it's quite a high-risk that it would read out with a positive result. It might yield something interesting. But to hit statistical significance with a small in would be, I think, unusual. And as such, we haven't really factored it into our main plans. But we hope it will offer some clues and some useful information.

In terms of etrolizumab, I think -- we think it's very likely that the mechanism works because there's another program. I think it's Entyvio that has an anti-integrin in IBD. That molecule, I think, is closing in on $4 billion in annual sales. And so I think Entyvio addresses alpha 4 beta 7 and etrolizumab hits alpha 4 beta7 and another integrin called alpha 4 beta E. So we don't know yet what the significance of that dual mechanism is. In fact, the Phase III will kind of answer that question. We had good Phase II data, and that makes it something that we think is very promising. We're pretty confident it's going to work. The question in ulcerative colitis and Crohn's disease is, how well does it work? And these diseases have been really tough over the years. There have been a lot of good molecules that have gone in, had Phase II data and then missed on Phase III or disappointed on Phase III. So I think we -- we think it's going to be -- I mean, we hope that the clinical trials will be positive and strongly positive. We've done everything possible to make this a really important medicine in terms of the formulation. We'll have a once-a-month subcu with an auto-injector. It's -- we're setting it up for success, and now we've got to wait for the Phase III results. Thanks, Tim.

--------------------------------------------------------------------------------

Karl Mahler, Roche Holding AG - Head of IR [39]

--------------------------------------------------------------------------------

There was maybe -- there was the 1 question...

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [40]

--------------------------------------------------------------------------------

Can we have a mic?

--------------------------------------------------------------------------------

Karl Mahler, Roche Holding AG - Head of IR [41]

--------------------------------------------------------------------------------

There was 1 question via the web on OCREVUS from [Manasi Agarwal] and he was asking about the future of OCREVUS. He said you reached CHF 3.7 billion sales in 2019. Do you think this has reached now the peak sales? And how do you see the ongoing competition for Mavenclad, Mayzent and any other kind of competitors going forward?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [42]

--------------------------------------------------------------------------------

Sure. No, I mean, we don't think it's reached peak sales. We think -- yes, I mean, in fact, outside the U.S., OCREVUS is really just getting started in a number of places. I think there's still something close to a dozen countries in Europe that don't have full reimbursement for relapsing and primary progressive. So we think ex U.S., there's a lot of growth continuing. Inside the U.S., we have 40% of new and switching patients, and we only have 20% total market share. So as we continue to get a large number of switchers and new patients, the market share will continue to grow. We really don't see any impact on OCREVUS from any of the recent launches. The 40% number has -- I think we vary between sort of 37% and 41% for 7 or 8 quarters. Obviously, there's now -- will be competition probably sometime in 2020 from another B-cell directed therapy. We think it's good for patients to have more choices, and we think B-cell therapy that we've pioneered is clearly a leading approach in MS. And so -- but we don't think that's going to stop the growth of OCREVUS.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [43]

--------------------------------------------------------------------------------

Okay. If we move on with the mic in same row over there, right.

--------------------------------------------------------------------------------

Keyur Parekh, Goldman Sachs Group Inc., Research Division - Equity Analyst [44]

--------------------------------------------------------------------------------

It's Keyur Parekh from Goldman. Three questions, please. Two for you, Severin; one for you, Bill. Severin, for you, you've spoken for multiple years about the direction of travel, which is growing through biosimilars. Can you talk a bit about the momentum of travel? So on an underlying basis, what you're guiding to today, assuming around CHF 4 billion in biosimilar hit is underlying growth of 8% to 9% in 2020? You've just done 9% in 2019. You're sitting on eculizumab, risdiplam, Tecentriq in adjuvant. So as we think beyond 2020, can you talk a bit about the momentum of that growth exiting 2020? That's first.

And secondly, in this medical bridge that you will do at the end of this year on kind of 0 to 15% net debt to assets, of course. But just philosophically, can you help us think about your preferences between returning money to your shareholders versus paying it to somebody else's shareholders? I'm not asking for modalities in which way you do it, but just help us think about one versus the other.

And then, Bill, lastly for you, Herceptin, Perjeta fixed-dose combination, help us think about the ambition there. Is it to protect what you've got there? Is it to recapture some of the volume share you've lost to the biosimilars? And in that context, how should we think about kind of broad picture pricing for that combination?

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [45]

--------------------------------------------------------------------------------

So let's start with your question on the longer-term growth perspective. And as Bill has alluded to, we have another 2 years, '20 and '21, where we have significant impact now from the United States, right? We still have Avastin in Europe, but we have Herceptin and MabThera basically behind us. And our experience has shown with MabThera and Herceptin, it takes about 2 years to wash it through, right? So that's our best estimate at this point, considering a significant impact from the U.S. in the short term. So on a very high level, I see '20 and '21 is the peak for the biosimilars, and then you'll have a tail, of course.

The other side of the equation, of course, is the new product. Bill, you have alluded to the continued growth of the medicines we have already launched. And then the big question, of course, is how many of the other medicines will actually make it to the market? And I would say that's the most uncertain component. We have -- really, I think that's really special now also in 2020. We have a very, very rich news flow. And you can play through all the scenarios. You mentioned itolizumab. Alzheimer was discussed, PDS, faricimab. I mean, you go on and on and on. And now all of those actually have a huge potential. So these are not small franchises if it works. But the question is, does it work? And you can play through these scenarios? If I look at the success rate of our Phase III trials over the more recent past, it's about 66%, 67%. Now that would suggest that some of those will work out and some will fail. But you can have a scenario where everything fails.

I've gone through that as well. I remember very well back in 2010, I will not forget this year. We had lots of stuff coming through, those of you who were around, in metabolism in diabetes. And I mean one after another kind of fail, and all our growth expectations turned into actually a decline of sales. So you can have a scenario like that. And then you can have times when kind of everything works. I've seen these phases as well. So that's the biggest piece of uncertainty. And now of course, what you always hope is that one of those big one makes it. Etrolizumab would be a blockbuster. I mean, without any doubt, imagine you have something in UC and Crohn's. All right. If you have something really differentiated in this field, your question gets obsolete, right? If indeed, by time it works, that's a shift, which kind of puts the growth on a completely pattern. But if -- I would put it like this. There's a lot of headwind still in '20 and '21, which is compensated for now with what we have already on the market. Now beyond that, I would hope for an acceleration of growth because the headwind goes away.

And then there is this kind of nagging thought in the back of my mind where I say, that's one or the other hits, and that could put us on a even different growth pattern. But I just don't know. And you know better than me, how binary it is. Look at the growth which we had in 2019. I mean at the end of the day, it's basically 3 medicines, which have made the difference, right? So just imagine, all 3 of them would have failed. We would've been in a very different space, and we can't get rid of this binary kind of situation in our industry. It remains high risk and high opportunity, of course, I hope for the latter. I guess...

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [46]

--------------------------------------------------------------------------------

One perspective, though. I mean, if this whole approach -- no, here's the thing. The worst year, the year you refer to I mean, again, in hindsight, we had -- we sort of had a lot riding on 3 things. One was Avastin adjuvant. The second was some cardiovascular metabolism mechanisms that I don't think we or anybody really understood very well. Now we understand that, but -- and the third was, we had a whole bunch of antibody drug conjugates. And so we had a relatively narrow base of things. So it turns out, Avastin adjuvant, that didn't work. A lot of those antibody-drug conjugates had toxicities and the cardiovascular mechanism we were targeting all had -- right? If I look at things like etrolizumab, LUCENTIS, PDS, polatuzumab. I think these are areas we understand pretty well. And the question may be how powerful are those results? So I don't think we have the potential for an all red Xs slide. I think some of these are going to hit and some of them are going to be big, and it's just a question of how many.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [47]

--------------------------------------------------------------------------------

Thanks for the encouragement. I feel much better now. So on the capital allocation question, right? Yes. Just go back a bit in time. Remember, when we gave this guidance of 0 to 15% net debt relative to assets that was actually at the time after the Genentech acquisition, and we gave this guidance for very different reasons. At that time, the bondholders were concerned that we were over leveraged, right? And they were interested that we reduce our debt, and we gave a guidance where we want to land. And that was the 0 to 15% relative to assets. Now of course, the situation has turned the other way around, and now the people are worried that we have not enough leverage over time, just to put that into perspective. I should also say that there is no -- I mean, there's no science around that. I mean what we did at that time is we looked at the industry, and we said what is reasonable and it provides us with the necessary strategic flexibility. So it's also not like if you get over this line, then suddenly, the world looks very different. This is kind of a high-level range, which we gave at the time, but it's still valid. And we have to see how things develop.

Now I would add the various components. I mean, as far as dividends are concerned, we have committed ourselves to an attractive dividend policy and certainly keep that going after now, what is it, 33 years of consecutive increases so we want to -- I don't want to end up as the CEO who first kind of even reduced the dividend or -- okay. And what that tells is we look at it very long term, right? So we want to increase the dividend, but we want to do it in a very sustainable way. So that will not change in essence.

And then the other question is around M&A, right? That's what you, Alan, alluded to. And that's, of course, very opportunity-driven, and I don't see a fundamental shift here at this very stage. We do see that late-stage assets remain very priced as scarcity around those. And in most of those transactions, we actually step out for economic reasons. The focus continues to be on earlier stage deals. That's typically where we can bring in other capabilities and assets. And I'm an interesting partner for those companies because it's not only about price, it's a lot about bringing it to the next inflection point and generating value for their partner this way rather than just the planned exit.

Then again, if there are opportunities, right, we would seize them, and we would be very willing to take external innovation into the company. So that's opportunistic, and we'll see how it develops. And I don't see, however, a major change in the dynamics here. So really, what I'm going to tell you here is really, we take it step by step. We don't have kind of scenarios, what we do, depending on those opportunities. Really, it's what I said before, and we'll cross the bridge when we have to cross it. And then we will cross it.

--------------------------------------------------------------------------------

Keyur Parekh, Goldman Sachs Group Inc., Research Division - Equity Analyst [48]

--------------------------------------------------------------------------------

And the fixed dose?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [49]

--------------------------------------------------------------------------------

Yes. So the fixed-dose combination, and you asked about the pricing. And I think for starters, we've been actually pursuing combination-based pricing in a number of countries around the world for some time. In other words, we don't necessarily think if we have 2 molecules that the price should be 1 plus 1 equals 2. We like to bring better value than that. And so I think it's almost certain that we'll do that in the case of the fixed-dose combination, but it will differ by country. And our goal really is to offer both convenience. And I think the difference between hours and 20 to 40 minutes is going to be significant, but also to bring value to the health care systems. So we'll -- yes, we'll be taking those matters very seriously.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [50]

--------------------------------------------------------------------------------

So let's move on a little bit to the back if we can go to the next row, and then we'll come forward again. We still have 15 minutes -- 10 minutes. Okay.

--------------------------------------------------------------------------------

Simon P. Baker, Redburn (Europe) Limited, Research Division - Head of Pharmaceutical Research [51]

--------------------------------------------------------------------------------

Simon Baker from Redburn. Three questions, if I can. Firstly, Bill, you mentioned that you've signed off 8 new pivotals in January. I'm assuming that's not the run rate for the year, but perhaps you could give us some of the pushes and pulls on the R&D line for 2020.

And then moving on to gene therapy. Could you -- I know it's early stages post the Spark closure, but could you give us any thoughts on the requirements for manufacturing capacity investment within gene therapy. I'm guessing that at the moment, Spark's relatively limited capacity is fine for an ophthalmic indication. But if you move to more systemic administration, you're going to need more capacity there. And then finally, for Thomas, given you're relatively new in the job of diabetes, the 2 classical questions. Firstly, is diabetes care core to Roche Diagnostics? And also, could you give us your rationale for why Roche Diagnostics is part of the Roche Group?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [52]

--------------------------------------------------------------------------------

Yes. So the 8 pivotal studies, I hope it is indicative of the pace for the year because that would mean we have a lot of really important studies to run. And -- but I don't know if I would take 3 weeks and multiply it by -- what would that be, 17 x 8. Probably not. But that would be a great problem to have. But I do expect the R&D line to grow disproportionately to sales and marketing, G&A, cost of goods. We're -- I mean, that's a major call to action for everyone in the company right now is we just believe that more and more in the future, society is going to be demanding and needing breakthrough medicines but also that we can deliver those at an affordable cost.

And so our vision for the future is very much that we're really maximizing our investment in R&D, improving R&D productivity also, but both the investment and the productivity, but then really emphasizing new ways of working to drive productivity across the organization. And I think it's really encouraging to see the gains we've made already and the fact that, that's not been a result of cost cutting. We haven't put out any guidance to people who say, hey, you've got to go cut your costs. I think I've shared with most of you we eliminated budgets in our commercial organization about 2 years ago in the U.S. and last year, in international. We've seen tremendous gains in productivity in marketing and sales.

Similarly, in manufacturing, I think over the 3-year period, we're up about 45% on volume, and our headcount is down about 18% over the same period. And that's not based on cost cutting. It's really been based on, hey, how do we make everyone's job count for more and make sure that people have great jobs and that they're able to make a maximum impact. And so yes, and we're going to take the gains from that and largely put them into R&D. So I think that's really where we need to be as a company. And frankly, I think that's where the life science industry needs to be for the future.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [53]

--------------------------------------------------------------------------------

Can I add to that, Bill?

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [54]

--------------------------------------------------------------------------------

Please.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [55]

--------------------------------------------------------------------------------

That certainly, that means we are not changing the guidance. I mean the guidance is the guidance. What's really happening is there's a reallocation that we really increase R&D on one hand and then take it from other cost lines. I think that's the idea here. Just to put that very straight. This is not a contradiction to the guidance.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [56]

--------------------------------------------------------------------------------

It's partly also driven by the acquisitions. So we continue to invest into the digital space, for example, with the FMI and the Flatiron acquisition. And then we also have Spark now. So Spark is really an investment case at that time. So there is a need to find the money somewhere, and it's a matter of reallocating resources accordingly. But I think those are good investments for the future.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [57]

--------------------------------------------------------------------------------

And of course, when we're growing, we're growing the top line as well. That's a big part of it. Gene therapy requirements, capital requirements for manufacturing, it depends a lot on the dose. And in gene therapy, the doses, you start talking about in logs. So the dose ranging can go from like 1 to 1,000. And then the manufacturing requirement has a lot to do with whether you end up at the high end or the low end. But I think on average, the manufacturing costs are going to be relatively low in gene therapy because you're still -- you're providing generally 1 dose. And so even if you end up at a higher viral load, it's going to be -- tend to be rather low. But exactly -- so I think it's basically on a P&L the size of ours, the gene therapy manufacturing capital is not going to be a point of discussion.

--------------------------------------------------------------------------------

Alan Hippe, Roche Holding AG - Chief Financial & IT Officer [58]

--------------------------------------------------------------------------------

Yes. To add to that as well, I think we have CapEx in 2019 of CHF 3.5 billion. I think it came down CHF 500 million, as I've presented before, from roughly [CHF 400 million]. And honestly, I don't expect that number majorly to change. And as Bill said, I think, really, this gene therapy piece is not really moving the needle so much.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [59]

--------------------------------------------------------------------------------

Thomas, you have the big question.

--------------------------------------------------------------------------------

Thomas Schinecker, Roche Holding AG - CEO of Roche Diagnostics [60]

--------------------------------------------------------------------------------

Yes. So let me start with diabetes care. So we are very committed to that business. I mean, that area is extremely important because, A, this is one of the areas with -- or diseases with the highest disease burden worldwide. And also, if you look at a diabetic, usually, they don't have just 1 disease. So they have usually also cardiovascular diseases, neurological diseases, et cetera. So we can combine our strength between diagnostics and diabetes care to really help the patients. So from that angle, I'm very much convinced.

Also, given that we've just launched a patch pump, the Solo micropump. We have new products there in the pipeline. We just saw another app. So especially, also there, we have a lot of digital capabilities also in the diabetes care area that we can leverage, not only for diagnostics but actually across the whole group. I mean we have so many software installations out there that are regulated, yes. And we think about some tools that we're looking at on the pharma side, specifically for neurological diseases on how you can monitor progress. I mean we can do that in a regulated fashion that pharma doesn't have the history in. So that helps, definitely.

But let me give you other examples that are key from my perspective. One is, I mean, we have a ton of joint development programs ongoing. In fact, we have a meeting set up in very near future where we're going to go through all of these. I mean I had this meeting also with our head of pharma research and early development. We were both surprised by how many programs we actually have ongoing between the 2 groups. And let's not forget, when pharma launches a medicine, they would want that companion diagnostic to be available as broadly as possible across the world. And that's why it's powerful to work with #1 in the diagnostic industry because we have the largest installed base of platforms around. So you want to make sure that the hospitals don't have to start installing different systems. They don't have space. They need to work with that organization. So from that perspective as well.

Another key reason for me, and when I was leading the research and development on the centralized area, it really became evident for me. I mean we're not only one of the largest, if not the largest antibody manufacturer in the world on the pharma side, we're also the same on the DIA side. So on the research and production side, we leverage a lot of technologies across the 2 divisions to be more effective, make better antibodies that are more specific, so we can really leverage that. So I mean, I'm very excited. And I think I want to even collaborate even more.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [61]

--------------------------------------------------------------------------------

I'm excited too. And I think what -- you didn't even share and we talk about this a lot, more and more health care systems when we come -- when we show up, Roche Pharma or Roche Diagnostics, they want to talk about the whole picture. I mean how are you guys going to come together to help us manage oncology care in our country? And we actually have solutions we can bring. And especially with the insights businesses, like FMI and Flatiron and DIA and pharma, I mean, we can come in and talk about how do we actually get patients having truly personalized care? And we have these shining tower examples we're running now in a number of countries. We've already announced several of them, including Croatia and South Korea, where every patient diagnosed with metastatic cancer will get a full workup, including things like foundation medicine testing, the tissue testing from Roche Diagnostics, and every patient will get a targeted therapy. And it might be one of our targeted therapies or someone else's. And that's fine because that's what leaders do in a field.

We have a commanding lead in diagnostics in oncology and in therapeutics in oncology, and we can bring that together. So -- and I think, really, the cost of us being together, I'm not sure what those are. In other words, when you ask the question like, well, why diagnostics and pharma together? It's not obvious to me, why not? Because we do, I think, gain a lot of knowledge from each other about the other part of the patient journey, and it works great, I think.

--------------------------------------------------------------------------------

Thomas Schinecker, Roche Holding AG - CEO of Roche Diagnostics [62]

--------------------------------------------------------------------------------

I mean, in neuroscience, for example, we have about 15 markets that we have established with pharma so they can do research in that area. So it's very key also in early research to be able to work with someone who can then develop these assays so this research can be done on the pharma side.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [63]

--------------------------------------------------------------------------------

We have to move on. Yes. Right.

--------------------------------------------------------------------------------

Karl Mahler, Roche Holding AG - Head of IR [64]

--------------------------------------------------------------------------------

One last one.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [65]

--------------------------------------------------------------------------------

Andrew at the back.

--------------------------------------------------------------------------------

Andrew Simon Baum, Citigroup Inc, Research Division - Global Head of Healthcare Research and MD [66]

--------------------------------------------------------------------------------

Not sure it's on. Maybe it is. It's Andrew Baum with Citi. A couple of questions. Number one, in reference to the administration's IPI proposal, I know Roche Genentech has historically great links with Washington, but this administration seems to have a way of take -- forging its own path. So I'm interested in your anticipation of timing impact and the enactment and how, of any IPI proposal as we indeed hear, is happening?

And then second, in reference to your adjuvant bladder trial, which failed to meet its primary endpoint, I'm less interested in bladder but I am interested in the messages that we can take from that to other indications despite the obvious differences in biology. So could you share with us, particularly when you're looking at the subgroups, any information on baseline PD-L1 status, baseline antidrug antibodies to Tecentriq? Were there any positive subgroups there which may have significance to how we think about future adjuvant trials were the current trials not to be positive?

--------------------------------------------------------------------------------

Karl Mahler, Roche Holding AG - Head of IR [67]

--------------------------------------------------------------------------------

We will disclose the data at a later point in time.

--------------------------------------------------------------------------------

William N. Anderson, Roche Holding AG - CEO of Roche Pharmaceuticals [68]

--------------------------------------------------------------------------------

Yes. Karl is reminding me that, I mean, honestly, some of the analysis you asked about, we haven't even done yet. And we won't be disclosing it until we bring it to a major medical meeting because I think it is important for the world to understand what we learned from that. But honestly, I don't even know the answer to the question you're posing. I think they're very good questions.

Regarding IPI and timing and all of that, look, we are -- I think what we have in common with the policymakers and really everyone else involved in the U.S. health care system is we think that patients deserve better. We think that patient out-of-pocket costs are too high. We think that there are reforms needed in the way that medicines are priced and sold. We've been driving for, on the pharmacy benefit side, rebate reform, lower out-of-pockets. Out-of-pocket cost should be capped for patients. It doesn't make sense that patients with serious diseases are paying a price for medicines that the insurance companies aren't paying, that these discounts happen or rebates happen without the patient benefiting. And so driving for that. On the hospital-acquired medicines and Part B, we've been in discussions with key policymakers on both sides of the aisle about market-based pricing reforms. We think that's really the way forward.

And I think something like IPI, I think, would be difficult to implement because it requires new channel partners that don't exist at this point. It would be very disruptive for the health care system. And we think that reference pricing is something that, frankly, it's a bad idea because the market-based pricing approach has led to this revolution in biotechnology that is bringing all these new medicines. And you've probably seen that the White House Council of Economic Advisers, so their internal group has said that foreign reference pricing in the form of H.R. 3 would result in 100 new innovative medicines and not being launched in the next decade. Something like IPI would have likely a similar effect. So yes, I mean, we just -- we're working on better solutions. And we remain optimistic that in the end, better solutions will be found.

--------------------------------------------------------------------------------

Severin Schwan, Roche Holding AG - CEO & Executive Director [69]

--------------------------------------------------------------------------------

Okay. We have to come to an end. Thank you very much for your interest. And for those of you who are still around, we have the usual upper [row]. However. I think you have to go to the U.S., you have to also go back to Switzerland. I will be around. And most importantly, Karl will be around. So see you in a moment. Thank you very much.