Roche Holding AG - Analyst/Investor Day

| About: Roche Holding (RHHBY)
This article is now exclusive for PRO subscribers.

Roche Holding Ltd (OTCQX:RHHBY) September 5, 2012 4:00 AM ET


Severin Schwan - Chief Executive Officer

Roland Diggelmann - Chief Operating Officer of Roche Diagnostics

David Loew

Hal V. Barron - Chief Medical Officer

Sandra J. Horning

Richard H. Scheller - Head of Genentech Research & Early Development

Andrew Chan

Daniel O'Day - Chief Executive Officer of Roche Molecular Diagnostics and President of Roche Molecular Diagnostics

Alan Hippe - Chief Financial and It Officer

Mike Burgess - Head of Roche Pharma Research and Early Development

Luke Miels


Steve Scala - Cowen and Company, LLC, Research Division

Sachin Jain - BofA Merrill Lynch, Research Division

Andrew S. Baum - Citigroup Inc, Research Division

Alexandra Hauber - JP Morgan Chase & Co, Research Division

Vincent Meunier - Exane BNP Paribas, Research Division

Michael Leuchten - Barclays Capital, Research Division

Seamus Fernandez - Leerink Swann LLC, Research Division

Jeffrey Holford - Jefferies & Company, Inc., Research Division

Gbola Amusa - UBS Investment Bank, Research Division


Ladies and gentlemen, good morning and good afternoon. Welcome to the Roche Investor Day 2012. I'm Selena, the Chorus Call operator. [Operator Instructions] And the conference is being recorded. [Operator Instructions] The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to hand over to Mr. Severin Schwan, CEO of the Roche Group. You will now be joined into the conference room.

Severin Schwan

Good morning, and welcome to our Investor Day in London. Also, a warm welcome to all of you who are joining us via webcast. Before we get started, I'd like to shortly comment on the management changes in the corporate executive committee, which we announced last week. As I have mentioned to some of you over dinner last night, I believe that above all, this is a powerful signal about the strength of the management team, both regarding Pascal Soriot, who is joining Astra Zeneca, as well as Daniel O'Day and Roland Diggelmann who take on their new responsibilities. So let me also take this opportunity to congratulate Pascal for leading one of the major companies in our industry, and I'd like to thank him for the many contributions to Roche.

I am absolutely delighted that Dan O'Day will take over the pharma business and succeed Pascal. As you know, he has expanded our leading market position at the very top of the in vitro diagnostics industry, and actually, he has spent most of his career in pharma. He has also been instrumental in driving the collaboration between the 2 divisions across pharma and diagnostics. It is also a great pleasure to introduce a new face to you today, Roland Diggelmann, who will succeed Dan O'Day as Head of the Diagnostics business. He will also shortly introduce himself to you right after my opening presentation.

So from my perspective, I think this is a very smooth transition both from a strategic and from an operational point of view, and I'm really proud that we can count on such capable leaders to take on the new responsibilities. Since Roland only took over literally 2 days ago, Dan will still cover diagnostics today, but Roland is around today. He joins us today and is certainly available for questions during the breaks or over lunch.

So let me get right into the agenda. We have a very rich program today. I will cover the strategy. We will report about the progress in our businesses in pharma and diagnostics, with a special emphasis on our late-stage pipeline as well as how we continue to maintain our strong cash generation. In the afternoon, we have 3 parallel breakout sessions, 2 on the early-stage pipeline for both gRED and pRED and another one where we talk about our various initiatives in the emerging markets.

Our strategy. As you know, we are facing an increasingly challenging environment. More stringent regulators and payers are under enormous cost pressure. And as a result of that, your investors are concerned about declining returns and the lack of growth. Our response, from a strategic point of view, is very clear. We focus on innovation. We focus on true medical differentiation. We focus on patient benefit, on our 2 core businesses, pharma and diagnostics. And we increasingly leverage the synergies between the 2 businesses to drive more tailored solution, personalized health care.

And I think increasingly, in particular with this challenging environment, we see that this strategy is paying off. Yes, regulators are very stringent. But if I look back at our most recent launches, particularly in the U.S., if you have true medical innovation, then regulators are very willing to bring those medicines and novel diagnostics to the market. Zelboraf was launched in a record time. Erivedge was launched on the base of Phase II data. Perjeta was launched ahead of our plans, and I'm equally confident about T-DM1 on the fantastic data we have seen with the EMILIA trial.

And likewise, if you look at the payers, there is cost pressure. Of course, we are also under price pressure. But if you look into Europe, where the cost pleasure is most elevated across the world, in the first 6 months of this year, we had a price pressure of about 2%, and that compares to a price pressure across the industry of about 4% to 6%. And that is fundamentally a reflection of the innovative nature of our portfolio and as such, of course, our ability to negotiate better prices with payers.

One of the key strategic competitive advantages we build on is this leverage between pharma and diagnostics. And you will see this as a common theme throughout the day, in health presentation, on the late-stage opportunities, very much, of course, in the diagnostics presentation but also in the early-stage presentations this afternoon. So I won't dwell into this in much detail. I'd just like to emphasize that I believe we are uniquely positioned in the industry by having the 2 businesses under one roof. And that allows us, in particular, to exchange know-how, expertise, intellectual property at the very early stage of research and development, at an early stage where independent companies would have to overcome many hurdles to work so closely together as we can.

Now of course, the first step is to create the innovation, the novel medicines, the diagnostics. It's then equally important to bring those medicines to the market, both in established markets and increasingly so in emerging markets. And again, we will cover this during the day in various presentation in the morning and our breakout sessions in the afternoon regarding emerging markets, and we'll share with you some of the, I believe, innovative approaches we have taken in terms of value-based pricing in established markets as well as new innovative pilots for tiered pricing in the emerging markets, which is both increasing access for patients and also driving our above-market sales growth in the respective markets.

Now let me get to a topic which I believe is really, really fundamental for the long term success of the industry, R&D productivity. And it is also a topic which is, in particular, important for us as a company, since we have this innovation-based strategy and as we invest more than our competitors over-proportionally into research and development. And I know it's a topic which is very close to your hearts. There is a number of studies out there which clearly show that R&D returns have been declining over the last decades. And we are now getting for the industry, on average overall, into a critical zone where those studies would suggest that returns on research and development are between 8% and 10%, so we are hardly earning back the cost of capital.

Now I believe what is really relevant from an investor point of view is not what is the industry, as a whole, doing. You invest into a specific company. So what I think is very important to look at is what is the R&D productivity across the various players. And to this end, we did a very, very simple analysis; you can easily reproduce that yourself. So on the horizontal axis, what we looked at is how much have our major peers invested into R&D over a 10-years period. And then we looked -- and this is represented on the vertical axis. Then we looked at how much has come out of this R&D investment in terms of newly generated sales by new molecular entities. So it's a rough measure. It's not looking specifically at the cash flow. We took sales, as a, if you like, surrogate for that, but I think it gives you a fair feel.

When I saw this data first, I was actually surprised. Intuitively, you would think there is a difference across the various players. I would not have expected how big the difference is. So if you look over a 10-years period, what you see here is roughly a 4x difference in productivity. So in other words, the best-performing companies get out 4x as much for each single U.S. dollar which is invested into R&D compared to the least-performing companies. So I believe this is very important to keep in mind if you look into research and development.

Now what does this mean for the industry? My prediction is that we will have more of a segmentation of the market. Increasingly so, we see this trend already happening. Those companies who don't deliver their cost of capital eventually will disappear, and I believe that companies will only go for marginally differentiated product. Differentiated products, there simply won't be markets for them. Payers simply will not reimburse that. And as a consequence, what we will see is, on the one hand, of course, a segment for the generic market. There's no doubt there will be a demand for that. And on the other hand, we will see a number of companies who go for really true medical innovation, and this whole area in the middle will disappear.

Now it is one thing to look at the R&D productivity retrospectively. It's easier. Of course, it's a bit more difficult to predict the R&D productivity as we go forward. So let me give you a bit of color, a bit of insight how we look at R&D productivity in Roche and how we manage R&D in our company. And there are really 3 basic elements I'd like to touch. The most fundamental one, in my opinion, is the understanding of the disease. If you don't understand the disease in the first place, not cutting-edge science in the first place, I think the likelihood to go for the right targets, to move on the right opportunities through the pipeline is just lower than if you really understand what is going on in the human body, if you understand what is going on in the disease. I think this is kind of the fundamentals. And at Roche, we put a lot of emphasis on that. We speak a lot about science. We speak a lot about excellence in science, cutting-edge science. You will hear this kind of everywhere in Roche. That is the basics.

Secondly, there are always more projects and opportunities than we have funds. So there is always a need for trade-off between different opportunities. Furthermore, of course, as you all know, there's an enormous attrition of projects along the development of a new medicine. So that means you have to constantly, constantly, in our business, on a daily basis, you have to make many, many trade-off decisions, and this is crucial. You have to decide early enough to invest enough resources into the most promising projects and likewise, stop and kill the projects which are less promising. So the decision-making, the governance in terms of how you allocate resources across the different opportunities, I believe, is absolutely crucial, and I will touch that -- I will touch on that in a moment.

And the third element, which I believe is very important, is this right balance between internal and external innovation. I keep saying there are lots of excellent -- some of the best scientists in our company. But nevertheless, the vast majority of scientific and medical progress is, of course, happening outside of the Roche walls, in academic institutions, other biotech companies, pharmaceutical companies. So it is crucial for us that we link into this pool of external innovation and leverage it for our company.

So let me start with the first one, the cutting-edge science, which is certainly the most difficult one to grasp, because it is not something that you have kind of very clear-cut criteria. It's, I believe, very much about the quality of science. It's very much about the quality of people. Now on this slide here, you see one possible indicator is actually the patent applications of Roche versus the other health care companies, and you can see that we are clearly leading in this respect. You can argue whether this is -- I think it's an indicator for the productivity, the vibrancy of science. It's certainly not a matter of quantity. Also here, it is a matter of quality.

What I believe is really, really important for us as a management is that we provide an environment where we attract the best scientists into our organization and where we provide them with an environment where they can really put their discoveries into new medicines, translate them into clinical benefits. And that has a lot to do, I believe, with the management approach. We believe in a decentralized management approach. We believe that research and science is not a matter of scale but only a matter of quality. That is also why we organize ourselves in more decentralized units. That is also the very reason why we kept the gRED organization and the pRED organization separate after the Genentech indication, to keep the vibrant culture of each of the places, to not increase the complexity by putting this all into one global organization. I do believe this is really, really essential. If you tell top scientists all day long what they have to do, if you bury them in standard operating procedures, you shouldn't be surprised if you don't get innovation. Scientists also need a very clear budget frame. They always would spend more money than you have. But within this frame, I think it's a lot about autonomy and letting good people make their decisions.

I also believe as a second element that this interaction with academic institutions is very important. At Roche, we have a long tradition of also hiring leading scientists from academic institution that brings in a lot of scientific knowledge, but it also keeps all these networks with academic institutions. We have a huge post-doc program in San Francisco and other places, like Basel with over 200 post-docs, which we have on an annual basis in the organization. Again, that keeps you fresh, that exposes you to new ideas, which are coming up in the academic environment. And we also encourage our scientists to do high-quality publications. Just last year, we had 12 publications in the top journals, Science, Nature and Cell. This is more than most universities around the world would have in terms of publications in these high-impact journals. And then, of course, it is very much a matter of attitude not only being top in science but doing this all with the intention to translate it into clinical benefits and bring it to the patients.

Now let me switch to R&D resource allocation, and let me start off with a general statement here. People talk a lot about the internal rate of return on research, et cetera. I do think that there is a huge difference in terms of the physician position criteria, whether you speak about research, early discovery on the one hand, on the one extreme or late-stage development on the other extreme. Of course, it all starts with a medical need. But then in research, it is very much about the plausibility of the scientific hypothesis, it is very much about the expertise you have in-house. And then it needs a lot of scientific judgment, whether you're on the right target, whether you can make something out of it.

If you are -- on contrast, in later stage of the development process, there's still a lot of interpretation. But very often, at least you have data, you have first clinical data, so there's a lot of room to interpret this data, but at least there are data you can talk about. There's an analytical element to that. You have a better perspective on the market potential. You certainly know more what is going on in terms of the competitive landscape. And at this stage, typically, you get also much closer into technical questions, whether it's feasible to produce a certain medicine, whether you can scale it up, whether you can provide it in the right formulation, et cetera. So the more you go to the later stage development, the more it gets, if you like, technical analysis. And I think this is important to keep this in the back of the mind, because it influences how, from a management point of view, you set the frame, and you actually do your R&D allocation. At the earlier stage, you rather work with overall budgets, and you leave a lot of freedom to the scientists to make the trade-offs within this given budget. Whereas at the later stage, you look much more at the specific opportunities, and you have a bigger discussion on the concrete programs.

Now let me just explain you how we do this in practice at Roche, and let me start with the Corporate Executive Committee, the various levels we get involved in R&D allocation. First of all, on a very high level, we take a stance on how much we want to spend for R&D overall. And based on our innovation strategy, what we say is we stand behind spending over-proportionally to the industry. At the same time, given the challenging environment, given the inherent risks of our business, we also don't want to let it go through the roof, and we committed ourselves to keep R&D stable in absolute terms for the medium term. So that is, if you like, the first very high-level approach we take in terms of R&D allocation.

In the next step, typically, on an annual basis, we look at the budget for the prospective units: how much we want to spend into diagnostics; how much we want to spend into research and early development; that is into gRED and pRED; and how much we want to spend for later-stage opportunities. And more on an arm's length basis, we have a discussion with Chugai how much they would spend into research and development.

On the next level, the CEC also gets involved in the review of the various disease areas. We have what we call an R&D steering committee. This is actually a subcommittee of the Corporate Executive Committee. Also Hal Barron from Development and David Loew from Global Marketing are joining in this committee, and we have reviews of the various disease areas. We had one of those earlier this year, in April, a more comprehensive one. And there actually was also a very important input to the decisions we announced at the middle of the year in reorganizing our pRED organization. I'll get back to this in a moment.

And lastly, we get involved in what we call LIP transitions. This is happening kind of 3 to 5 times a year when we take a decision to move a specific medicine into pivotal late-stage trials, when we commit to late-stage trials. LIP stands for life cycle investment point, so that is moving it into late-stage development. And that's what we do at the CEC.

And all the rest of the decision making, which of course is the vast majority of the decisions which are being made on a daily and constant basis, that is delegated further down into the organization, and we have respective processes at the appropriate levels. gRED and pRED, we cover this to a certain degree in the breakout sessions this afternoon. So we have Research Review Committee. We have the Early Stage Portfolio Committee. We have, of course, the Late Stage Portfolio Committee. Those are all very interlinked, so we have standard respective processes down in the organization.

Coming back to reorganizing our pRED activities, that is a concrete example where this whole governance process and this decision-making process came to action. We started on the basis that we wanted to keep R&D stable for the organization. And that was a challenge in itself, because we had many projects coming through the clinical development pipeline. And we had a more detailed review of all our various disease areas at the various sites. We saw some more promising, some less promising areas. Amongst them was inflammation in [indiscernible], where we thought there is less promise, so we de-prioritized that. As a consequence of that, we looked into the overall structures. We also looked at the management complexity in pRED and decided to consolidate our activities, to co-locate the management team in Basel and eventually close the Nutley site to reinvest all the savings from infrastructure cost, leveraging support functions into our clinical programs.

Now let me get to the third part of R&D productivity, and I'd just like to re-emphasize how important that is. We have teams who are constantly screening the market for opportunities. We have currently about 150 ongoing active partnerships with third parties, and you can see, it is really material for our business. About 1/3 of our pipeline, about 1/3 of our marketed products directly stem from external partnerships, and we keep it very much in focus.

All of that -- all of these efforts eventually should be reflected in our pipeline, and I'm -- yes, I'm proud to present here one of the leading pipelines to you. We have today over 70 new molecular entities in clinical development. Since 2011, we have had 25 late-stage clinical trials delivering positive results. And that, of course, is feeding our pipeline. That very much is feeding our late-stage pipeline, and Hal Barron will cover the various compounds in much more detail later this morning.

Let me just highlight here that we have another potential 5 LIP candidates for the remainder of this year. Our target is to bring another 3 compounds into late stage. And actually, one of those, rontalizumab in lupus, has already been moved into late-stage development, and we will also cover rontalizumab in one of the breakout sessions this afternoon. If you look a bit ahead into 2013, again, that looks promising, and again, this is a result of the many programs we are getting through the pipeline for 2013. Today, we already see 10 new opportunities, late-stage opportunities and candidates to move forward into late-stage development.

With this, let me conclude. First, our strategy remains firmly focused on innovation, on medical differentiation, on driving personalized health care by leveraging our capabilities across the 2 divisions. Secondly, we see above-market growth, [indiscernible] growth in the emerging markets. That, of course, is very much driven by our products as such but also by new innovative access models. And thirdly, I believe we have one of the leading pipelines in the industry, which will drive the value for our various stakeholders in the future.

Thank you very much. And with this, Roland -- here you are. Can you just shortly come up on the stage and introduce yourself? Thank you.

Roland Diggelmann

Thank you very much, Severin. A very good morning, and welcome, ladies and gentlemen. It's a true privilege to be able to be here today. Just wanted to very briefly introduce myself, and I look forward to the interaction with all of you in the coming months. As Severin mentioned, Dan O'Day will present today still the diagnostics business section, but let me convey to you my excitement about diagnostics, which is a growing market, which is a market that we have a goal to continue to outgrow going forward. It's also a market that is exciting in that it is, in itself, in its own right, a market that we will continue to expand, that we will continue to deliver the appropriate solutions to the market, to our customers, where I believe we are well positioned but where we will also, of course, make sure that we adapt to a changing environment.

Beyond that, it is also an exciting market, because we have the unique opportunity to collaborate with our pharma colleagues in personalized health care. And I think we are better positioned than anyone to do that and to achieve that going forward. And our goal remains to be the partner of choice, of course, for companion diagnostics and the importance they will have going forward. And I think these 2 pillars together make for a great business. They make for great opportunities. And with that, we will also be able to continue to expand our market presence in our key markets but also beyond, in our emerging markets. And with that, I think we can contribute to also expanding health care and, of course, to expanding access to diagnostics across the world.

So I really look forward to interacting with you over the next months and years. Thank you very much for the opportunity to address you, and I would like to hand over to David Loew for -- as Chief Marketing Officer and Head of Global Product and Strategy. Thank you, David.

David Loew

Thank you, Roland, and welcome to this new position. We are all looking forward to working with you in your new role.

Good morning, ladies and gentlemen. In the next 25 minutes, I will provide you insight on the key strategic drivers we have defined for pharma on how are we going to succeed in a challenging environment, as well as show you how we assess our performance in this environment. The 3 pillars we have defined are a focus on innovation, a willingness to expand into emerging markets and to protect the patients by ensuring that governments set adequate guidelines for the development of biosimilars.

We have seen in the last years that the market growth rates have come down quite significantly. This was triggered by 2 factors. One, there was a bolus of patent expirations; and two, governments under the financial pressure have set tougher standards on pricing in terms of asking for price reductions for in-market compounds or being more demanding when it came to launching new drugs. I think we are very well positioned with highly differentiated products in this environment, and you can see that in the first half year of 2012, we have, in fact, exceeded the performance of most of our competitors and have had a strong first half year.

Also, our innovative products have helped us penetrate into emerging markets where, based on tiered pricing approaches, patient access schemes, we have managed to actually increase significantly our performance in the emerging markets, deriving a very strong growth as well as making sure that patients get access to our drugs. So the international region has actually now exceeded Western Europe as a region. What we can see is that with this innovation focus, we have managed to get into a #5 position with a very good performance in comparison to others, because this came from organic growth. So we didn't go into mega mergers like other companies did.

The question is, now, are we well placed for the future where we see that Western markets are going to have a limited growth, and most of the growth is going to come from emerging markets? And I think the answer is we're going to be successful in the future. And I say this because when you look to the past, we have seen a strong market growth, which was only partially compensated by a decrease because of patent expirations leading to a relatively significant net market growth. And in the second part of the period in 2010 and '15, we see that the governments that are under financial pressure are not going to see a very strong net market growth because there are a lot of patent expirations. However, the market can still grow with in-market compounds, which are highly differentiated and with the newly to be launched drugs. So the question is for Roche, can we tap into this market growth? And looking at our pipeline, I think I can say we are very proud to have such a strong pipeline with highly differentiated products, and we are well placed to tap into that future market growth.

Let me zoom in now into the oncology products as this is still our main part of our pipeline. You can see that oncology is projected to grow further, because the incidence of cancer death is on a rise. And we have successfully launched in the recent part a lot of new highly innovative products and are about to launch still also new drugs into oncology in the very large cancer incidence settings. So for example, breast cancer is an important part of cancer death, and here, we come with Perjeta and T-DM1. So let me just show more details around those 2 products.

You know that Herceptin has set the standard of care in first-line metastatic HER2+ breast cancer, and it was a tough hurdle to beat. Now I was very pleased to see that when we reported on the CLEOPATRA trial in first-line metastatic breast cancer, we have not just shown that we have significantly increased the progression-free survival but we have also increased the overall survival. That is very good news for us and for patients, because we are really pushing the boundaries of the efficacy here.

We have also published just last week that T-DM1 also hit an overall survival improvement, which was significant, and the increase in efficacy was also accompanied by an increase in the tolerability. And that's great news for patients, because now, they don't need the chemotherapy anymore in second-line breast cancer, and all of you who have some family members or friends who were affected by cancer know what it means to receive chemotherapy. So I think that is great progress. The question now is, can these 2 products be combined? And we are running this trial right now. That is going to deliver results next year in first-line metastatic breast cancer, and we will bring both treatments, Perjeta and T-DM1, into the adjuvant early breast cancer segment as well. So if we are successfully delivering those results, I think the HER2+ breast cancer space has been completely changed and redefined. Many of you will remember, we talked about this 2 years ago at the New York event, and we had this vision. We have it really delivered now. So I'm very proud to see that we have made huge progress here for patients.

Focusing on hematology, we have a similar situation where MabThera has set a very high standard of care. Now results can still be further improved on the efficacy but also, in the way we administer the products. And here, we are going to bring new treatments to patients in the form of the subcutaneous form of MabThera, making the administration much, much more easy, and I will just show you an example on what this means, as well as developing GA101, which is hopefully going to bring more efficacy to patients, as well as new entities, like CD22 or Bcl-2.

Let me show you what it means, subcu, for the hospital system and for patients. You can see if an infusion needs to be administered, the patient needs to come to the hospital or the clinic. The pharmacy needs to reconstitute the product, then an IV line needs to be put, and the patient needs to be administered the Rituximab over a relatively long period. That period can take -- the whole thing can take about one day. Now with subcu, we are going to significantly change this, because the pharmacy doesn't need to prepare the product anymore. The nurse can administer it in only 5 to 7 minutes, and the patient can then be discharged. That has big benefit for patients but also for clinics, because they need much less infusion chairs. They need less physician time, less nurse time, less pharmacy time, so that has big pharmaco-economic advantages. This data is going to be presented at ASH in December this year.

Now innovation doesn't only come in terms of bringing new products to market. It can also be that we come up with new pricing models. And you have seen that in -- with Avastin, we have already launched capping products that I'm going to walk you through in just a minute. And we have also started to develop a tiered pricing strategy for the emerging markets, including our highly innovative products.

Now what is going to happen in terms of pricing for the future? Especially in oncology, we're going to see a change from a per-pack-based pricing, where the hospital needs to pay for a given vial, towards more a patient-based pricing, where depending on the indication or the setting or the combination, there are going to be different prices applied. That, of course, requires that governments or payers have a patient-based information, and we are developing those databases together with the governments around the world in order to make this happen.

Now this has already become reality. Avastin, as an example, in Germany, is being priced differently according to the indication. You know that with colorectal cancer, you need the low dose. With breast cancer or ovarian cancer, you need the high dose. And so we have developed a commercial package where patients need to pay up to a certain amount of grams, up to 10 grams here in Germany. Then they get the drug for free up to 12 months, and after this period, the scheme starts again. Same thing in Italy, where we have a pay-for-performance agreement at the beginning. In the 6 -- first 6 weeks, there is a cost sharing, and then patients need to pay up to 11 grams. The rest they get for free up to 12 months. These new pricing models have been very successful. We have seen that Avastin now is growing, including in Western Europe, very strongly again in breast cancer but also in ovarian cancer.

Now when we start building up new highly innovative treatments, like Herceptin with Perjeta, or we're going to see something similar with Perjeta and T-DM1 or Zelboraf and MEK, we will have to think about different ways of pricing those new combinations. And this is a schematic illustration of what's going to happen outside of the United States, where we have figured out a way how we can price them together and have a per-patient treatment cost that we are going to implement, making it affordable for society and making sure that patients can really access those treatments.

Let me spend some time now on our strategy, how we expand into the emerging markets. You did not hear us spend that much time talking about emerging markets in the past, but I think we have been very successful with our strategies, building up how are we going to penetrate the emerging markets, helping the diagnostic testing rate to increase, helping KOLs to understand the data and coming up with tiered pricing approaches so that patients can actually access. Now the good news is that this has also led to a significant above-market growth. You can see that in almost all these 7 markets, we have consistently outperformed the market. And you pick the first 3, Brazil, China, Russia being the biggest ones, we have very, very strong growth rate in those markets.

What are we projecting is going to happen in the future is that governments under the financial pressure will be forced to actually give more room to protected originals, more room to unbranded copies, and the branded copies and the non-patented products are going to shrink. So I think Roche is very well positioned with highly differentiated products to actually benefit from that growth of patent-protected originals. Also, we are going to see that there is a growth of patients having access to some form of health care coverage. The out-of-pocket part of population is shrinking at the moment, and the public payer part is increasing, as shown here. Now we have to make sure that we get included also in this public payer market, and we have developed ways, as this example shows with MabThera, to get included in public payer market and where we have derived now a significant growth of patients being treated and also deriving sales, of course, from it with those new access schemes.

Now the pool of patients that still can get treatment is, of course, very large, especially in China and India, and there are large efforts underway to help to tap into those pools. Luke Miels, in fact, is going to give you much more details where he is going to tell you about what can we do with testing KOL interaction in the workshop that we are going to hear soon.

And then our third pillar is we want to protect patients in -- by ensuring that governments establish adequate guidelines for the development of biosimilars. We have had a lot of interactions with different governments around the world, WHO, as we bring a lot of expertise to the table how monoclonal antibodies need to be developed and what criteria need to be looked at for the clinical development. And so when you look at the 2010 situation where we had very few countries actually having had biosimilar guidelines, and there were basically none for monoclonal antibodies, this has changed completely now. When you look at the world map today, and you can see that the majority of countries have developed those biosimilar pathways or have laws in place that are specific in terms of how do you need to develop those biosimilars. Now clearly, when you look at, for example, at Reditux, that has consequences for biosimilars, because there are these higher demands now, and some of the clinical submissions were considered as not sufficient enough and have been rejected by several governments around the world, where they demanded more clinical data.

So in summary, I think our 3 strategic growth drivers work. We are focused on an innovation, redefining the standard of care. We have been successfully expanding into emerging markets and will continue to have a strong focus on that, and we have a focus on protecting patients by ensuring these biosimilar guidelines are well entrenched.

With this, I hand over to Hal Barron.

Hal V. Barron

Thank you, David. Okay. Good morning. It's great to be here again to see all of you. Since last year, we've had a lot of great progress in the pipeline. And it's my pleasure over the next 30 minutes or so, if the timer gets corrected, to share with you some of the really exciting things that are happening. Severin -- and this is a little bit of the agenda. I'm just going to spend most of the time on the update of the portfolio and then end with a little bit of strategic direction we're taking in product development, an initiative we're calling ReThink D and finalize with a couple of slides on the news flow that's coming.

Severin showed this slide in the beginning, and it really is a very important slide. It's a slide that highlights, in fact, why we believe our innovative model is going to result in a highly competitive company and a company that's going to do great things for patients. Now in order to stay up in the upper left-hand corner, it's really critical that we deliver on a number of things. First of all, we need to really focus on great science. You'll be hearing a lot about that in some of the breakout sessions from gRED and pRED. We need to make sure that science translates into great medicines by designing trials that take smart risk into account, that really focus on ensuring that the molecules are being developed in the right diseases; to make sure we have the right dose; to make sure, whenever possible, we have the diagnostic -- companion diagnostic strategies you'll be hearing a lot about over the time period today; and really ensuring that when a molecule -- the data that's generated in Phase II isn't promising, that we remove it from the portfolio so that we don't end up spending money on molecules that won't ultimately be successful. This whole strategy needs to result in a higher probability of success so that the return on investment is above the cost of capital and an important driver for our business.

So when you look at our progress over the last 2 years, I think this slide says a lot. We've had 30 major unblindings, 30 trials that have generated data over the last 2 years. And 25 of the 30, as described on this slide, in fact, were positive. So that's, basically about every 3 weeks or so, we've unblinded data with the success rate of around low- to mid-80s. And it's this kind of success rate that results from all the things I just described, and it allows us to really deliver truly transforming medicines to patients. They're medicines that regulators will approve, that payers will reimburse and that ultimately will benefit patients. And it's this model that we continue to focus on as our innovative strategy.

Today, and this is the late-stage portfolio, there's 11 molecules, 11 new molecular entities that we call them, that -- molecules that have yet to be approved, that are in development. These are some of the most promising molecules in the portfolio. I'm going to cover mostly the oncology and metabolism areas, because the neuroscience is going to be covered in the breakout session with Mike and Luca, and Richard and Andy are going to be covering the immunology molecules in the gRED breakout.

Now it's important that in addition to the late-stage portfolio, as we described it, by new molecular entities, it's important to realize that in the product development area, we spend a lot of time focusing on making sure that each of the molecules that once they get approved, we don't ignore them. We continue to explore opportunities to help patients by looking at other indications, looking at the science and seeing where it tells us to go. And in fact, we have 11 molecules in which we're actively developing molecules either that were recently approved or some that were approved for a long time but where we think there's still opportunities to help patients and therefore, grow the market. And those are described here.

Now in oncology, where we have about 60% of our investment, we have taken a very broad approach. Not only are there a lot of different potential therapies, but we have a lot of different strategies in developing these different therapies. We have a very, very exciting platform technology called antibody-drug conjugates, which I'm going to spend a few minutes on in a second, where I'm going to highlight both T-DM1 and a new antibody-drug conjugate that's in Phase 1 testing that we believe is very promising called anti-CD22.

Another competitive advantage that we have and an advantage that hopefully will translate into real significant benefit for patients is the opportunity to combine 2 molecules from our portfolio. And if you follow the scientific literature, it's becoming increasingly clear in some of the clinical literature, that to really make the biggest impact for patients, it's not going to be just one reagent. We're going to have to combine things that hit several pathways so that we can either overcome the resistance that's engendered by the first one or actually -- to actually outsmart the tumor and be able to have more cell damage and actually, hopefully, allow the patients to live longer.

And one of the real advantages to having so many different therapies in your toolbox, if you will, is the ease with which we combine them. You can do the basic science, find the combinations that are most likely to help patients and then by having so many of them, combine them relatively freely. It's very hard sometimes to form collaborations with other companies to develop these combinations, and this gives us a real competitive advantage. And I'll be talking about a couple of those in the slides that follow.

The third strategy, which we're very excited about, and we have 2 molecules in this area called glyco-engineered antibodies. And this is a technology that essentially enables us, through a modification of the genes in the cells that are producing these antibodies, to produce the antibodies that are so-called afucosylated. They have a different glycosylation pattern. And that change in glycosylation of the antibody itself, while not affecting the binding area on the antigen that it binds to, affects the Fc portion, the bottom part of the antibody that engages the T cells, the natural killer cells and allows the T cells to attack the tumors in a more effective way, at least that's the theory, by increasing so-called ADCC. And this technology is being explored with 2 different molecules, GA101 and GA201.

Maybe one of the most promising areas in all of oncology is going to be the opportunity to use our own immune system to actually attack the tumors. And one approach, which is very exciting, is an antibody to -- an antigen called T-DM1. I'm not going to spend time on that, but that represents another strategy we're taking to will really make a big difference for patients with cancer. And of course, we have a very large -- no pun intended, a very large small molecule program that really focuses on all of those targets that are inside the cell that antibodies that target the express antigens couldn't be able to attack. So very excited about this very deep portfolio as well as broad in terms of its approaches to oncology.

So let me move to the antibody-drug conjugates. I leave this image for you to think about. I'm sometimes focused on ensuring people really understand the impact that the antibody-drug conjugates are going to have not only on our company but on patients and medicine in general. This is really -- I think we're in the beginning of a very transformative period, where antibody-drug conjugates are going to be a really important therapy in oncology. And if you think about the evolution in oncology, we've gone from chemotherapy, which has been helpful in attacking the tumor cells but, in fact, is very non-targeted and has many systemic complications, much toxicity to the targeted therapy area, the personalized health care, the area that we helped ushered in and are continuing to advance in incredible rate, Herceptin, Perjeta, Zelboraf. And what that does is the antibodies are specifically targeting the tumor, but they don't, in effect, avoid the systemic toxicity from the chemo. Patients are still suffering from all the toxicities. In many respects, the holy grail is to be able to treat a patient with a targeted therapy that doesn't have these systemic chemotherapeutic side effects. And as I'll describe and as you probably well know now, T-DM1 is a molecule that is allowing us to deliver the chemotherapy selectively to the cancer cell to minimize some of the side effects associated with systemic chemo. And this highly targeted antibody-drug conjugate concept is really going to take us to the next level.

Now the concept behind the antibody-drug conjugates is relatively simple, but it's the simplicity that is, in some respects, very complicated. All we need to do is find cells where the antigen is expressed, hopefully, exclusively on the tumor cell, that the antigen that the antibody binds to has to be internalized to bring in the chemotherapy. We have to make sure that this antibody-drug conjugate has a stable linker so that the chemotherapy that's linked to the antibody doesn't fall off in the circulation but falls off once the antibody is internalized into the lysosome. And we have to tag the antibody with these very potent chemotherapeutic regimens, sometimes, chemotherapy that wouldn't be tolerated, in fact, if given systemically but when delivered selectively can have a really potent effect on killing the cell. And when you marry those 3 concepts together, you actually find that, in fact, you can expand the therapeutic window. You can -- by delivering more chemo, you can actually have greater efficacy but, because there's no systemic toxicity or limited or less, you can actually give more. And it's that dream, if you will, that led us to hope that maybe we could develop molecules that not only were better but at the same time, safer.

And that's exactly what we saw with T-DM1 in the EMILIA trial. And this is a study in second-line breast cancer, as you know, where patients who were refractory to Herceptin were randomized to either T-DM1 or the standard of care lapatinib, and the data is shown here. Not only was there a marked improvement in the progression-free survival time, an improvement in quality of life, but there is now, as we recently noted, an improvement in overall survival as well. And this was associated with less toxicity, so better response rates, better progression-free survival, better overall survival, better quality of life and less systemic toxicity, and this is not likely to be isolated to this one molecule. This is a platform technology that we're really going to be able to take to the next level.

We have the opportunity, now that we've demonstrated such a significant benefit with T-DM1, to move in into the front line setting, which we will do with, hopefully, MARIANNE, which hopefully will be positive and then also move it into the adjuvant setting. But in the adjuvant setting, it's very challenging to actually be able to improve upon what we think will be the standard of care someday, which is Herceptin plus pertuzumab. We already have a terrific response with Herceptin. With pertuzumab, we expect the APHINITY trial hopefully to be positive and increasing the bar further. And so to actually add on to this, we are trying to be very innovative in our approach. We need to show that T-DM1 benefits, we believe, over Herceptin and pertuzumab.

So to do that, we actually have a three-pronged approach. One is to do what's called a non-pCR study. And what this means, essentially, is that some patients who undergo neoadjuvant therapy will actually have their tumor essentially eradicated based on the observation under the microscope that there are no tumor cells when the surgeon goes in to take out the tumor. They get their upfront chemo. They go for the resection, and there's no evidence of tumor at the time of the surgical resection. Those patients do extremely well. The converse is those patients who do have evidence of disease don't do well, and it's in those patients where we think we can make a big difference with T-DM1. Now that's a relatively smaller population, but we're also going to try for the traditional adjuvant study as well in a high-risk all unselected group, and that's this trial 2, where we think we might be able to provide a benefit.

And the third is a very innovative design that I'll talk about at the end when we talk about the ReThink D initiative, where we're trying to see is there a way of working with the regulators to design a path forward that really speeds up the approval of molecules into the adjuvant setting. One of the most important things in oncology is taking these great drugs that work in a metastatic situation and moving into the adjuvant, where you can really cure lives. And oftentimes, this takes years and years and years. And it will be terrific if we can use some of the accelerated processes that the FDA has been talking about and develop innovative designs to move this forward, and I'll talk about that at the end.

Now as I said, we're not -- the antibody-drug conjugate is not unique to T-DM1. In fact, we have 8 other antibody-drug conjugates in the clinic, 9 total and 16 other ones that are in research for a total of 25 antibody-drug conjugates. And of course, not all of them are going to be as spectacular as T-DM1, but we really believe that by combining all the attributes that I described earlier with a stable linker, with the antigen expressed internalized, that we can even leverage the PHC, the personalized health care approach. Because each time we develop these antibody-drug conjugates, they're usually uniquely identified by the antigen that's over-expressed.

Sometimes, diseases are very homogeneous, like some heme malignancies, where CD22 is almost universally expressed. And in this disease, we have developed an antibody-drug conjugate to, where the antibody, being CD22, is a stable linker and a very effective chemotherapeutic reagent called MMAE. Now in -- with this molecule, we are in Phase I testing, and this is just a representative patient who you can see has -- this is the right side of the body. You can imagine the person's feet towards you and the head in the back. And this is a right inguinal lymph node. And you can see that, basically, after 4 cycles, there's essentially no evidence of disease. This is a 79-year old patient with a diffuse large B cell lymphoma. End of one [ph], but with T-DM1 and 23 others behind it, we do really believe this is going to result in the transformation in the treatment of oncology.

Now there's another antibody-drug conjugate that, again, we're not going to disclose the antigen but is also demonstrating activity. This is also in heme malignancies. And in this patient, we have an 85-year old with a mantle cell lymphoma. And this is a patient who had been previously treated with Rituximab and Bendamustine, and they had rapid progression after 3 cycles. You can see that they have an over 80% reduction in their tumor mass after this antibody-drug conjugate. And not that we make decisions on end of ones [ph] and end of twos [ph] and things like that, but this is likely, as I say, to represent a really significant shift in how we think about oncology.

And here's the slide that I hope amazes you. This is what the future holds. This is the ones in the clinic. This is the ones behind that are in late-stage research and in early-stage research and again, represents the future of what we're going to be doing.

Now we talked a lot about our strategy to take a great drug, Herceptin and make it even better for patients with pertuzumab and T-DM1. And we need to do the same thing with Rituximab, and we have a very solid strategy to do so. As you know, the bar is very high. Rituximab is a terrific drug. It's really made a big difference in standard of care and the treatment of lymphoma and CLL. And getting a better drug is very, very challenging, but we have 3 approaches: GA101, which I had briefly mentioned; the Glycart technology, the afucosylated antibody, that also has a different antigen that it binds than Rituximab on CD20, and it binds in a -- to a Type 2 region, it's called. And not only does this antibody block the signaling of CD20 but also, by binding a different epitope on the CD20 antigen, it actually induces a apoptotic process, a caspase-independent apoptotic process that we believe, and preclinically is supported by, could have incremental benefit over the CD20 that binds in the so-called Type 1 spot, liked Rituximab. In addition to GA101, we have a very exciting molecule that I talk about, Bcl-2, an apoptosis inhibitor and the anti-CD22 that I just went over. Now we have a number of programs with GA101, the first of which we'll read out is the CLL trials. But we also have trials in refractory indolent lymphoma, aggressive lymphoma and the first-line indolent NHL. And you can see the names and the dates in which those trials will read out.

We also put, for your reference mostly, what we're hoping for in terms of the target effects that we're anticipating or at least striving to achieve. And the first one to read out, which will be in 2013, is in fact the front-line CLL study. And we have -- that is a study, at least one of the arms is comparing chloramiocel versus Rituximab -- sorry, versus GA101. So this is where we expect a pretty pronounced effect over doubling in the progression-free survival. In the third arm of this, we're actually going to be comparing whether GA101 is actually superior to Rituximab in this setting and higher bar, as I said, the one we're again hoping for a significant benefit. And based on the Phase I data, where we saw a pretty significant number of responses in Rituxan relapse refractory patients as well as some interesting data from the Gow [ph] study, we're cautiously optimistic that this program may deliver.

Now a program a little bit further behind but one that represents really exciting biology and I think based on this Phase I study, really clear evidence of activity, is a collaboration we have with Abbott on a series of molecules. This is the most promising that, basically, are targeting the pro-survival proteins called Bcl-2, Bcl-XL and Mcl-1. These are proteins that are over-expressed in heme malignancies and some solid tumors as well and are thought to possibly be upper regulators in the mechanism by which patients become resistant to chemotherapy. So in theory, this could be a very, very important pathway in oncology. They're basically small molecule, BH3 mimetics, that basically neutralize the pro-survival effects of Bcl-2. So Bcl-2 keeps the cells alive, and by blocking that, you can induce an apoptotic process.

Now the collaboration with Abbott started with a molecule called ABT-263, and in fact, that molecule was an inhibitor of both Bcl-2 and Bcl-XL and had a side effect of thrombocytopenia. This new molecule, 199, actually is much more specific for Bcl-2 and doesn't have the thrombocytopenia as major problem. And what you can see here in this waterfall chart is that when administered in the Phase I trial to these CLL patients, you see some pretty impressive results: large number of responses; responses are upwards of 6 weeks and longer. And in fact, when we looked at the total lymphocyte count, about almost 80% of these patients had a normalization of the lymphocyte count. So clearly active, very exciting, and there could be a lot of synergy with this molecule and anti-CD20. In fact, we have our CLL Phase Ib study with GA101 going to start this month.

Let me move on to MetMAb. We've talked a little bit about this last year, and we have some pretty interesting data as the field evolves that I think makes this an even more exciting molecule than it was a year ago. As you recall, c-Met is a receptor that's the ligand for hepatocyte growth factor. The ligand binds to the receptor and dimerizes it and signals, and what we've developed is a one-armed antibody so as to not actually stimulate the receptor, but which happens actually with a full length monoclonal, but by using a one-armed antibody, we can actually block the ligand's dimerization. And what's pretty evident from the epidemiology of Met expression in tumors is that in many tumors, the more -- the higher the Met expression, the worse the outcome. Preclinical studies suggest blocking Met actually helps reduce tumor size, proliferation, survival, et cetera.

So we're very excited about this target. The rationale for our Phase II with Tarceva in second-line lung was that it looks like Met might be a mechanism by which patients become resistant to Tarceva. We moved forward on that trial. And as you know, in aggregate, the Phase II trial was negative. But in the subgroup of patients who over-expressed Met, which was a predefined endpoint, we found that not only was progression-free survival dramatically improved, over a doubling in PFS or about a doubling in PFS, we also saw, and this is very, very interesting, a very pronounced effect on overall survival. It's very rare that the effect on overall survival is larger than the effect on PFS. In fact, it's kind of hard to understand how that might even happen, and you have to sort of suggest other mechanisms by which this would be working outside of the typical progression. And if true, this represents a really exciting opportunity to make a big difference for not only patients in second-line lung, but as you'll see, a robust program.

Now when we presented this last year, many of you or several of you, actually, were concerned that when you have an overall negative trial, the subgroup analysis, particularly retrospective, even if prospectively identified, could be spurious, could this be just a chance finding. Because, in fact, if it's benefiting the Met high, it has to be harming the Met low, and that's a little unusual. And so one of the things we were interested in seeing is how do the other companies that have -- that are blocking this pathway in one way or another -- and there are 2. ArQule has a small molecule c-Met inhibitor, and Amgen has an antibody which has a ligand HGF. And in this graph, just to quickly walk you through this, this is our data in blue. This is the Amgen data in gastric cancer, and this is the ArQule molecule in gray. And what you see here in the Met high and the Met low is exactly what we saw. Not exactly but very, very similar to what we saw. We see an overall trend towards benefit in the aggregate population, but what's more pronounced is, of course, the benefit in the Met high. But instead of the lack of benefit in the Met low, we see in all 3 of these trials either a strong 10 or a statistically significant worsening in the outcome in the Met low. So this gives us some confidence that what we observed is real, that this is a biologic interaction, something very important. And we believe this proof of concept in 2 other diseases actually gives us a lot of confidence that this could be a very important molecule in oncology.

So we had originally launched a number of programs when we got our Phase II data, the -- of course, the Phase III study and second- and third-line study in orange. But also, as we've subsequently initiated a couple of Phase IIs in the front line, both squamous and non-squamous, separating them out, we're looking at gastric cancer moving directly into Phase III. We have a study in Phase II in triple negative breast cancer, where we think that there might be some compelling reasons to think Met's involved as well as trials in metastatic colorectal cancer, where the data's pretty interesting pre-clinically and glioblastoma, where again, Met over-expression has been identified and seems to be prognostic. So a pretty robust program. We believe this is an important pathway. We believe we have a really excellent reagent that could really make a big difference for patients.

So I mentioned synergistic combinations. We have a lot of different molecules in the pipeline, a lot of different possible combinations. Our combinations are guided by preclinical data, telling us which approaches might be most fruitful. This is a list of a number of different combination trials that are underway. MetMAb plus Avastin, the preclinical data is pretty interesting there. We have a program with an anti-EGFL7 and Avastin. Of course, we've talked about T-DM1 and Perjeta as being, for the MARIANNE study, is the combination that in the future might completely transform the landscape to make even Herceptin not needed in the treatment of breast cancer and some other trials.

Now as I said, we're guided by the preclinical evidence, and here's the kind of data we look for. We look for true synergy, where you see that in this model, Avastin or an anti-VEGF was active. You see activity of Met inhibition. But you see a dramatic impact. In fact, almost a flatlining of the tumor cells when you combine the 2. And this is the kind of data that lends itself well to telling us maybe we should test this combination out on patients. Obviously, when we're looking at synergy, we have to make sure it's not synergy in a toxicity way but hopefully, just with the efficacy. And so we slowly but surely move these through development. We've actually had very, very interesting and fruitful discussions with the regulatory authority on how to develop combination NMEs, and I think we have a very nice path forward for a number of different programs to really see if we can make a big difference by this synergy. We also -- I'm not going to spend time on this but similar process, as I said, with Avastin and anti-EGFL7 and many other combinations that we think are really going to drive the future of oncology.

So I know that was a lot to digest. Oncology is a complicated area where we think we're really well positioned not only in the -- just the hemo indices but broadly, with the anti -- the ADCs. As I said, the Glycart technology could be very fruitful in the combination.

Let me move to metabolic diseases in the last 5 or so minutes and tell you about 2 molecules. You know about aleglitazar. It's in Phase III. It's a mixed PPAR alpha/gamma agonist. That means that it has properties that are both focused on insulin sensitivity and reducing hemoglobin A1C but also the alpha effects that are modulating lipids in a beneficial way. And it's the balance of these 2 that we believe is important in developing a really important therapy for diabetes.

Now one of the interesting observations from a number of studies has shown that when you look at the data for pioglitazone, which is a relatively good drug, that if you try to predict which patients are really benefiting from a cardiovascular perspective, it turns out that the strongest predictor of atheromatous volume reduction, which is a surrogate for cardiovascular benefit, turns out to be how effective the drug is at lowering triglycerides and raising HDL, that ratio. Some patients have a pretty pronounced effect on that, some less so. And in this study done by Nicholls et. al. published last year, they identified a group of patients who have a really robust effect on volume of their atheroma that are being reduced, and it being associated with this change in lipids. Why we think that's interesting is because when we compared pioglitazone and aleglitazar in our Phase II study, we found that when you look at the triglycerides, you look at the HDL, you find that the ratio is almost twice as much reduced by aleglitazar compared to pioglitazone. In other words, the ratio goes down about onefold with pioglitazone and over twofold at the dose we chose for Phase III. And we think that sends a signal that it's modulating the lipids in a way that we think could actually further benefit the patients with cardiovascular disease.

And again, why are we excited about this? Well, one reason to believe that this trial might work is the fact that, in fact, pioglitazone at dose has been studied in cardiovascular disease, and it was negative. But what people don't realize is if you dig deeper into the details, while the overall trial had a hazard ratio of 0.9 with a p-value of 0.9, therefore, negative, if they had used the same endpoint that we're using and the endpoint that's traditionally used, which they didn't use, and you redid the analysis using that endpoint, in fact, the proactive study would have been positive with a hazard ratio of 0.84. And this, again, is the endpoint we're using. And if one assumes that not only do we have the same hemoglobin A1C reducing effect, but we have a better effect on HDL and triglycerides and for that matter, a better effect on LDL, that gives us cautious optimism that, in fact, a trial like this could be positive. And it would be the first study of a drug in diabetes that would actually really reduce cardiovascular events and a big landmark for an epidemiologic phenomena that's just really sweeping the world.

So we have the ALECARDIO study. This is the randomized Phase III. A 2.5 year follow-up to ACS patients, acute coronary syndromes, to see if we can reduce the events. One of the gates in our program, our concerns about this molecule, is that in the Phase II studies, we saw a reduction in creatinine, which reflects, usually, a reduction in glomerular filtration rate, which is not a good thing because it usually represents renal toxicity, and that's bad. And in fact, one of the PPARs was actually killed because of that. We believe that was an effect due to the alpha -- PPAR alpha, which simply reduces blood flow to the kidney which is, in some studies, actually a good thing to do. And if that was the case, the reversibility should be demonstrated in a trial. In other words, it should only happen when the drug is on, and when you stop the drug, the creatinine and the renal function return to normal. If it was due to nephrotoxicity that doesn't return.

So we did a study called ALENEPHRO. It was a Phase II study, head-to-head against pioglitazone, looking for that very phenomena. Is the reduction in creatinine suggesting a -- or an increase in creatinine or a reduction in GFR due to blood flow or due to some other mechanism? Or is it renal toxicity? Well, we're pleased to announce today that, in fact, the trial hit its primary endpoint. In patients with moderate renal impairment and type 2 diabetes, the average decrease in the estimated glomerular filtration rate within 8 weeks of discontinuing therapy basically came back. So that's very good news. It somewhat was predicted by the data but very comforting to know that we're, in fact, not having any deleterious effect on the kidney. And in fact, the rest of the safety profile looked very consistent with the PPARs. So that was a very important gate for this ALE study. It's that, in conjunction with the constant monitoring by the DMC, gives us, again, reassurance that this is an important trial to finish.

Now lastly, we talked about PCSK9, a molecule that we're really excited about. It's one of the hottest areas in cardiovascular medicine right now. And that's because of the 71 million people with a high LDL, there's about 12.5 million people who are inadequately treated by statins. Either they get a statin and the effect on LDL is not adequate enough, that's 11 million or people who actually are intolerant to statins.

The reason we're excited about PCSK9 is a number of reasons. First of all, PCSK9 is a protein in the blood that actually binds the LDL receptor and internalizes the LDL receptor in the liver. And the LDL receptor is important as the body's own way of removing LDL. So your LDL's floating in your blood, it binds with the receptor and gets internalized and removed from circulation, therefore, prevents it from causing atherosclerosis. PCSK9 binds that receptor, and then the receptor is not around to bind LDL. Therefore, PCSK9 is essentially raising your LDL. And in fact, when people are given statins, their PCSK9 levels go higher to compensate for this phenomena. So by blocking PCSK9, we think we have more LDL receptor available to bind LDL.

Now in addition to that sort of biologic basis, we actually have genetic data that's very compelling. In people, particularly in African Americans, there's about a 2% to 2.5% rate of a genetic abnormality where they have a loss of function. Sort of a genetically PCSK9 low. And in those patients, they have about -- in the African Americans, about a 30% reduction in LDL. In the Caucasians, about half that. And in the African Americans, those patients who have that LDL reduction have about an 88% reduction in cardiovascular risk, as shown over here. In fact, it's about a 3:1. For every percent reduction in LDL, there's about a 3% reduction in cardiovascular risk. And a similar phenomena in Caucasians, where there was about a 15% reduction in LDL and about a 45% reduction in risk. So we really believe that this is hemo causal pathway of atherosclerosis. Given the unmet need, we think this is likely to be effective in this disease. And as this graph on the right shows you that in general, LDL lowering is about a 1:1 ratio. 1% LDL reduction results in 1% reduction in CV risk. And what this genetic data tells us is that, that association is even more dramatic in the PCSK9 modulation.

Now we've done Phase I. We exposed a number of healthy controls, individuals who had elevated LDL. And they experienced, as we expected, a significant decrease in LDL. The PK results were actually very important, and they suggested to us that there may be a favorable administration profile that, in theory, could give us a competitive advantage. We have a current Phase II underway. We'll know the results in early 2013. And if they support our belief, we'll be ready to move quickly into a Phase III program and hopefully, have what might be best in class in one of the most important new developments in the cardiovascular area.

So lastly, ReThink D. To stay up in the upper left corner, the place that Severin showed you where we'd love to be in terms of our efficiency, we need to do 2 things simply really well. We need to make sure that we're better than industry averages on our probability of success, and we need to make sure that the development programs that we do undertake are less expensive than they've ever been. By increasing the probability of success, reducing cost, the cost to us -- develop a successful molecule goes down, and that's our goal. It's really a mind shift in the product development organization. We want people to focus on ensuring we develop drugs with companion diagnostic to up the PTS. We want people to be more efficient at their execution, really ask ourselves how many things in a clinical trial do we -- can we not measure because they really don't add value. Being much more efficient with that can save a lot of money. Can we take some smart risk? And can we come up with some innovative trial designs to actually go faster?

Let me just give you one quick example of this. We've been working with I-SPY investigators, academics, the regulators to ask the question how can we get a new therapy to patients in the adjuvant setting faster? It's such an important benefit to, particularly, breast cancer patients who could be cured if we could get these great drugs to them sooner in the adjuvant setting, because it takes so many years. And one of the most compelling findings that we believe is predictive of whether something is going to work in the adjuvant is this pathologic complete response rate that I just described. With pertuzumab, we saw a really dramatic benefit in that, and that made us confident that we should start the APHINITY trial. We asked ourselves, if we're so confident, can we convince regulators that they should be confident and maybe provide a pathway for accelerated approval? And the FDA is looking into this. In fact, they expect, by the end of the year, to put out a white paper with guidelines after they analyze many, many different trials to see how tightly correlated is pathologic complete response with outcome. And if their data is what we think it might be, there's a possibility that we could work together to come up with a design where we can look at this very short endpoint, sometimes as soon as 12 weeks after chemotherapy, we can do the surgery and see, in fact, if there's a pathologic complete response. And if that can be used as a surrogate for benefit, its transformational in how you can develop drugs in the adjuvant setting. It's risky, but we think it's a smart risk to take. We're focused with the FDA and other regulatory bodies around the globe to see if there's a way of making this happen. We hope we can do it with T-DM1, and that was the third trial I described that I get back to is using this novel paradigm that we're hoping we can move forward.

So that's about it for me. The upcoming news flow over the next few years is pretty dramatic. I'm not going to go through all these. We have a number of additional life cycle options that are going to read out, as you can see on the bottom, a number of new molecular entities coming over the years. We think we have a really robust pipeline. The pipeline is going to make a big difference for patients. They're innovative products. We're going to do everything we can to ensure that they're successful in maximizing the effect we see to make it good for patients, for payers, for regulators. And we're going to be rigorous about getting rid of those programs that really don't cut the bar, meet the bar that we post for the Reds so that we can really develop best-in-class, first-in-class drugs and really make a difference. This is some of the big decisions, it's mostly for your reference, that are coming up. As I said, we're very optimistic that this is one of the best portfolios in the world, and it's continuing to grow in an impressive way.

So thank you very much. I am now going to ask my colleagues, Severin and David, to come up, and we're going to do a little Q&A.

Severin Schwan

Thank you, Hal. Thank you, David. With this, we have another 20 minutes or so to go for the first Q&A session here in the [indiscernible] before we then go into the break. I'd like to ask you to really focus on questions which are somehow related to the areas which were covered by Hal, Mike and David and myself and keep the questions on other topics for the respective presentations later today. Can we have the first question, please? Take here, the question in the first row.

Question-and-Answer Session

Steve Scala - Cowen and Company, LLC, Research Division

Steve Scala from Cowen and Company. A couple of questions on aleglitazar. In ALENEPHRO, was there a trend towards reduction of cardiovascular events? And in ALECARDIO, I think the slide said data in 2015. I could be wrong, but I think in the last update, you had said data in 2014. Secondly, Severin had mentioned early on it's critical to understand disease. Does Roche think it understands Alzheimer's disease sufficiently to have 2 antibodies in development? And then lastly, may I ask a question about dal-OUTCOMES? Was there any numerical trend in cardiovascular events, either positive or negative, in patients treated with dalcetrapib?

Hal V. Barron

I should have written that down. Okay. I think...

Severin Schwan

I think the first one was on aleglitazar.

Hal V. Barron

Remind me, so you said in...

Steve Scala - Cowen and Company, LLC, Research Division

ALENEPHRO, any trend in cardiovascular events?

Hal V. Barron

The ALENEPHRO data, we're just going to give you the top line results today. The full dataset will be presented at an upcoming meeting, and that data should be in there that you're looking for.

Steve Scala - Cowen and Company, LLC, Research Division

ALECARDIO. A delay in the data?

Hal V. Barron

I don't think it's a shift, is it Karl? No. Part of the challenge with these programs is that they -- upon completion of recruitment, in order to satisfy some of the regulatory concerns about safety, every patient needs to be followed for 2.5 years so you can kind of march out when the last patient ends 2.5 years later, data cleaning. Unlike some of the oncology trials which are event driven, we pretty much know when this is going to -- so that's correct.

Severin Schwan

There was a question on the timing also of 2014, '15.

Hal V. Barron

That's what I just -- yes.

Severin Schwan

Okay. No delay there. Okay, good. Dalcetrapib, was the other question.

Steve Scala - Cowen and Company, LLC, Research Division

Was there a positive or negative trend in patients that were treated with it?

Hal V. Barron

Again, I think the full dataset will be presented. We were confident in the decision we made was the right one to stop.

Steve Scala - Cowen and Company, LLC, Research Division

Alzheimer's understanding?

Hal V. Barron

My suggestion is that we defer that to the -- to Luca and Mike's discussion, where they'll actually go in great detail on exactly that question.

Severin Schwan

But perhaps on a high level, I can comment that we look at Alzheimer's in a broad way. We looked at different mode of actions not only at removing the plaque but also other mode of actions, for example, the base inhibitors and others. And as far as Alzheimer's is concerned, we have 2 antibodies, but they are used in different settings. So the one which is developed in chivad [ph] is really for mild, moderate Alzheimer's patients whereas the one which is pursued in peret [ph], and which is actually a bit further advanced, is on Potolmin [ph] Alzheimer's. So it's an earlier setting. And then we have an additional study running with this Colombian group of patients, which actually goes even more earlier into preventive settings. But again, this will be discussed in much more detail this afternoon in the breakout session. Okay. Let's go along in the first row here. In the middle, please.

Sachin Jain - BofA Merrill Lynch, Research Division

It's Sachin Jain from Merrill Lynch. A couple of product questions and one broad question. On T-DM1, you've discussed the various filing potentials and discussions with the regulators. Just what's the bandwidth of filing time lines dependent on outcome if the FDA accepts your surrogate or will you have to fully proceed with the full study? Secondly, on PCSK9, just what's the maximum dosing interval being investigated in Phase II? And then finally, for Severin, you discussed in your introductory slides your R&D productivity, 4x industry average on the metrics you've looked at. Just wondered what data you have further down in terms of pRED versus gRED or across therapeutic categories. What data is driving your allocation further down the chain?

Hal V. Barron

Well, your first question about T-DM1 in the adjuvant setting, just to be clear, I think I know you meant that. We have 3 strategies, because we're not exactly clear how the pathologic CR innovative neoadjuvant program will actually evolve. We're hoping that guidance is clear exactly what's needed, the population, the treatment effect on pCR, how robust does that need to be and how much information do you need on disease-free survival in a study like that and whether the treatment should stop at the time of surgery or continue beyond. There's a bunch of unknowns that we're going to need to work out with the regulatory authorities. But we're optimistic that we can come together with a plan, what -- try to speculate what that plan might be and the time lines for doing a trial and then the time lines for filing would be too premature though. Sorry...

Sachin Jain - BofA Merrill Lynch, Research Division


Hal V. Barron

PCSK9, yes. It's a good question. The Phase I data will be presented soon. And from that, you might glean some insights as to what that might be, but we're not disclosing the treatment intervals yet.

Severin Schwan

Okay. On the R&D productivity, that was really done on a group level. So we looked at total overall sales derived from new molecular entities. And if you look over the last 10 years, it was very much also driven by the oncology portfolio and by gRED and to a lesser extent, by Chugai and by pRED. But the data which we used are really public data which you can easily reproduce. Can we have the next question here, please?

Andrew S. Baum - Citigroup Inc, Research Division

It's Andrew Baum from Morgan Stanley. It's Andrew Baum from Citi. Three questions. So firstly, on the subcutaneous Rituxan. In the environment when we might anticipate biosimilars with a 30% to 40%, let's say, price reduction, do you think the structure of gain sharing or of reducing the infusion cost is sufficient to offset just the mere headlines, bigger price reduction a drug budget can be -- can address and how that's going to play out in the various countries? So that's one. Second, ArQule recently suspend their Japanese trial with MetMAb, with their Met small molecule, the TKI, due to lung -- interstitial lung disease. Is there any read-through or any impact as you think about how that may impact MetMAb in lung? Have you seen anything in combination with Tarceva that may raise that concern? And then finally, just on the idea of using neoadjuvant data as it reached approval, I remember that there was one of the NSABP trials, where it clearly went in the other direction, where there was no read-through between pCR and PFS. Surely, that's going to be a significant hurdle to the FDA signing off of this as a surrogate endpoint for the adjuvant setting.

David Loew

I'll take the first one. Regarding the pricing strategy and the differential, we can come in to biosimilars. First, I think important is to know that the biosimilars have been delayed on [indiscernible]. So we're going to bring subcu onto the market well in advance of that, hopefully. And so I think that provides us the opportunity to establish, actually, the drug into the market, and clinicians, pharmacists will see the benefit. Nurses will see the benefit of what it means to shrink massively the administration time and alleviate the whole infrastructure problem for the hospital. Now having said that, of course, we need to remain competitive. So we will have to observe very clearly what is going to happen. The 40% assumption is an assumption that you throw in. I think it depends on country by country, how they're going to price it, what the local regulations are, but be ensured that we will behave competitively.

Hal V. Barron

So the MetMAb question, as you know, our Phase II study with Tarceva and to the best of my knowledge, that we haven't seen anything that we would consider a signal for interstitial lung disease. Now we have observed that in certain countries, like Japan, the incidence of interstitial lung disease is clinical trials is higher. And so there is a regional effect, a diagnostic sort of a proclivity to diagnose that more commonly in certain areas. Whether it represents a real signal for them, I can't comment, but we haven't seen that as to be a rate limiting step in our development program at all. And your other question, one of the issues in why the FDA is taking their time and doing actually something like a 12,000-patient net analysis of all the data in adjuvant and neoadjuvant trials to see whether there's a correlation, is not just to see whether there's a correlation but to look at the subgroups for whom there may be less of a correlation, like ER-positive versus ER-negative or other [indiscernible] markers that might suggest a group of patients for whom it either is or isn't as correlated. And we're hopeful that we can find a way forward, as I said, but you're right. We're waiting for their analysis of all this data and to see how it shakes out in terms of what the implications are for T-DM1. Sandra, I don't you if you have anything to add to that? Or...

Severin Schwan

Perhaps we switch now to the back. We have a question in the last row.

Unknown Analyst

[indiscernible]. A few questions, please. One is on ALECARDIO. Could you just comment on whether you're doing any patient stratification based on metabolic risk? And also, what is the earliest date that the study could be terminated prematurely for efficacy? And a question also on the surrogate markers in the adjuvant breast cancers or the adjuvant cancer settings generally. Did the FDA indicate at all that if they go ahead with this, it's only going to be applied to highly innovative new compounds? Or could that actually facilitate the work for the biosimilars companies? And the final question is just on your PCSK9. I was just wondering if you can comment a bit more on the actual molecule and any advantages it may have over the competition in terms of potency. Or do you think that the dosing is going to be the main advantage?

Hal V. Barron

Okay. ALECARDIO patient stratification. I mean, I'm not 100% sure I know what you mean. But in terms of a baseline covariance that we collect and are going to analyze, we certainly will look at lots of different predictors to see if the treatment effect varies by various things that one might think a treatment effect could vary by, including basically lots of different things I won't go into. But the primary endpoint of the studies are [indiscernible]. So if your -- if that's your question, yes, the primary endpoints are [indiscernible]. Premature termination, there's a Data Monitoring Committee that's looking at this. We haven't disclosed any kind of interim analysis plans. So I wouldn't plan on those. PCSK9, it's an IgG1. The data will be out soon. You can look at that to see if you think there's any difference from that. But what we're commenting on today is that we believe the PK difference or the PK attributes are what we think is our competitive advantage there. As far as the adjuvant stuff and where we're thinking, if it's possible, maybe I can just ask Sandra to comment on that and any other aspects of this pCR strategy. Is there a mic you can give to Sandra?

Severin Schwan

Do we have a mic in the second row? Yes, we have one. Thank you.

Hal V. Barron

Oh, she has one, sorry. Sandra Horning is our Head of Oncology and always around to bail me out.

Sandra J. Horning

Let me comment on the prior question about subtypes of breast cancer that are being targeted, if you will, for this potential pCR status and early approval on that basis. And it is very clear that the correlation between pCR and disease-free survival is different among different risk groups. And specifically, in those patients that are HER2+ and/or triple negative, that appear to be the groups that will most likely be amenable for this procedure. And we will, of course, await the FDA guidance and the results of the meta-analysis to know that further. I might need to have that second question repeated.

Hal V. Barron

The question was more about will this result in neoadjuvant claims or adjuvant claims and how might the FDA sort of interpret what you can do with this, as well as will it apply retrospectively to other molecules or just innovative ones. Or could biogenerics use this approach?

Sandra J. Horning

Well, again, we're awaiting what the actual guidance will be. But at this point in time, it would appear that we're talking about the approval for neoadjuvant use and that on the basis of understanding what is necessary in terms of identifying the population and identifying the difference, the delta in path CR, that would presumably be something that any therapeutic or any therapeutic regimen could qualify for.

Hal V. Barron

But again, I think the key thing is there would have to be a substantial increase in pCR. So it would seem very hard for a generic to do that.

Severin Schwan

Thank you. Let's go forward again into the second row here, please, if we can have the mic. Yes. Thank you.

Alexandra Hauber - JP Morgan Chase & Co, Research Division

Alexandra Hauber from JPMorgan. Two questions. Firstly, on the capping concept. So for Avastin, the concept is gram per patient per year. But then HER2, it's going to be more like a cost per patient, and I assume that's also going to be per year. So the question is do you need -- before you figure out where that cap roughly should come in, do you need to know how T-DM1 fits in? Or is it just going to be 2 different capping programs, one for Herceptin plus Perjeta and one for T-DM1 plus Perjeta?

David Loew


Alexandra Hauber - JP Morgan Chase & Co, Research Division

So 2 different ones. And does that have within a relatively homogeneous geography, such as Europe, will that have to be very similar caps? Or could you see that there -- I mean, of course, I understand the emerging markets are going to have innovative pricing models. But because we're moving away from the gram per product, is that still going to be very homogeneous in comparable geographies?

David Loew

Yes, you know that all the governments talk to each other. So even if you try to have confidential deals, eventually, people will calculate what have you offered, et cetera. So there will be homogeneous offerings coming out at least for certain regions. You have still a different setting in the U.S. than, for example, to Western Europe, it's a very different setup, different pricing mechanisms. But inside of Europe, you're basically forced to have extremely similar setups.

Alexandra Hauber - JP Morgan Chase & Co, Research Division

Would the European reference pricing system adapt to that, sort of like a 20% discount to the cheapest and so on and so forth? So could you completely operate outside that framework?

David Loew

If you do a very specific commercial agreement. That is actually outside of the typical price comparison.

Severin Schwan

Yes, and if I may add on that. Of course, the reference system in Europe is a national matter. So every nation in Europe decides on pricing and as such, on reference pricing. There is parallel exports and parallel imports are something which is an institutional matter. On a European level, pricing is a national matter, and we do see very different reference pricing schemes in different countries. So per se, that would allow for some flexibility and some differences within Europe. But again, as we all know, especially in Europe, schemes tend to harmonize across countries. And then it's more difficult to differentiate, which actually, I believe, if I may make this comment, it's a pity, because eventually, it reduces the possibility to give broad access to patients. And what we see and will increasingly see with enormous cost constraints, in particular in Southern Europe, actually would be beneficial for patients if we had more of a differentiation, if there was more of a solidarity between the rich countries and the poorer countries. That would be beneficial from a medical point of view in terms of patient access, and it would certainly be beneficial from, if you like, a macroeconomic European point of view, because it will force the innovative industries, like the life science industry. It's more a prolific topic for the longer term, I guess.

David Loew

Just one thing to add. I mean, there will be, of course, still a list price for a vial. So when you allude to parallel imports or exports, the list price is still going to matter.

Alexandra Hauber - JP Morgan Chase & Co, Research Division

My other question is on the antibody-drug conjugate platform. So if I understand correctly, so the key really here is the specific antigen that is over-expressed. So unless the antigen is expressed universally, like CD22, that almost certainly will require a companion diagnostic then. So how many of the 7 in the clinic have a companion diagnostics? And how many of the 25 have?

Hal V. Barron

I'm not sure we've disclosed it. It might be on the slide. It was on the slide. I think there's -- each one that has a circle around it, the circle implies companion diagnostic. I don't remember the exact...

Severin Schwan

I think it's the majority.

Hal V. Barron

It's going to be -- most diseases are heterogeneous. They're -- yes.

Richard H. Scheller

It's essentially all of them, because the antibody itself is the diagnostic. We're working with VENTANA very carefully on that.

Hal V. Barron

Yes. CD22 is sort of the exception.

Severin Schwan

Okay. Good. If we move on in this row, if we just can move on the mic.

Unknown Analyst

Sandy Planik [ph] from Sencos [ph]. A couple of questions, one following on from Alexandra on the ADCs, antibody-drug conjugates. You're using a stable linker between the antibody and the drug so it's toxic. And as that goes into the cell, I'm just a little bit surprised that that strategy is the only one that you're potentially looking at into clinical development, because if you look at the drug preferentially, surely the drug should be released through the endosome, for example, to get into the nucleus if it's hitting a nuclear target. If it's going through the lysosome, which is where the antigen will be marshaled, then eventually, that will be recycled to the surface. And hence, the antibody will be taken out of the cell with the drug that's conjugated to it. So what's the rationale for using a stable linker?

Severin Schwan

Richard, if you can take this question. If we can have a microphone in the first row.

Richard H. Scheller

So the mechanism is that the antibody is taken into the lysosome and completely degraded, which releases the cytotoxic that then diffuses out of the endosome and into the cytoplasm. So the antibody itself is not recycled. It's degraded, and that's the mechanism of the release of the toxin from the antibody. The Seattle Genetics' platform is different and that is a cleavable linker. And it depends on which antigen and which cancer, which linker you use, and we're developing a series of our own internal linkers and different sets of toxins that not only inhibit the polymerization of microtubules but also damage DNA and so on.

Severin Schwan

Thank you. Do you have another question? Yes.

Unknown Analyst

Just a final question. On the sort of top level for you in terms of the oncology franchise. You talked eloquently about the use of immunotherapies. I'm just wondering whether, going forward, you have a vision of getting into the area of regenerative medicine and more broadly, into immunotherapies. Because you talk about reducing side effects, and you're talking more also about combining antibodies. But it seems to me, the next step is the third wave of biotech, which is effectively regen medicine and immunotherapies, which deal with those 2 issues very well.

Hal V. Barron

Andy, you want to take that?

Severin Schwan

We have another mic here.

Andrew Chan

So as Hal noted, we do have one program, anti-PD-L1 that's in Phase I. We have a number of research programs. We're exploring other mechanisms by which we can activate the ones -- the patient's own immune system, to harness their immune system for a curative outcome. And with respect to regenerative medicine, we have several earlier programs in research more on the information side first with respect to regeneration of epithelial cells for a number of inflammatory diseases. And if that holds up and we can demonstrate proof of concept, we can easily move that into a number of different fields, including oncology, neurosciences and the like.

Severin Schwan

Thank you. I'll take one more question, more in the middle here, if we can have the mic here. And then we go into the break. Thank you.

Vincent Meunier - Exane BNP Paribas, Research Division

Vincent Meunier from Exane BNP Paribas. Two questions, please. The first one is on aleglitazar. Now that you have been reassured on the safety of the drug, can you address the diabetes indication by itself beyond the cardiovascular risk reduction in diabetes? And for this, do you need to wait for the readouts of ALECARDIO in 2015? The second question is more general, regarding the biosimilars and the protection of the anti-CD20 franchise. So you have GA101. You said that some biosimilar projects have been delayed. Can you elaborate on this? And beyond GA101, the difference versus the L2 franchise is that you just have one molecule to protect the franchise beyond Herceptin -- sorry, MabThera. And then, what are really the options for you when we are talking about the end of the decade?

Hal V. Barron

So let me take the aleglitazar one first. It's an excellent question. If everyone understood that, it's whether or not -- I think, if I understood the question, whether or not to pursue just a general diabetes claim in that program. And in fact, what we had as a company decided is that once we got the ALENEPHRO data, that we would digest that and actually have the Late Stage Portfolio Committee review with the teams what we should do next given that finding, whether we should pursue additional indications like diabetes or wait for the data. And we haven't made that decision yet, and you can expect us to do so soon. And we'll get back to you once we've made those decisions. I think that was the...

Severin Schwan

I think the second question was on the biosimilars. And perhaps what I should say is that -- I mean, it's clear that Rituximab is a tough hurdle to take, and that also applies for us, for Roche. And we are certainly less advanced as we are in the HER2 franchise. I mean, it reminds me a bit of the situation we were 2 years ago in Herceptin, where we said Herceptin is a very tough hurdle to reach, and we went for T-DM1 and Perjeta. Today, we have astonishing data, perhaps even better than we would have thought. Now we can't be so confident yet at this very point for the anti-CD20 franchise due to the fact that the data are not yet in-house. But it is not only GA101. We have 2 other opportunities lining up here with Bcl-2, the collaboration with Abbott as well as the anti-CD22 ADC, which Hal has referred to. In terms of biosimilars delays, I think there is some competitive intelligence which indicates that they would only come begin...

David Loew

Q1 2015 is kind of the latest assumption as it looks like. But I mean, you know yourself, I mean this is a field which is evolving. I think the hurdle is high to prove that it's the same. So we observed it very closely what is happening there.

Severin Schwan

Okay. Thank you very much. With this, I suggest we go into the break, and let's try to be back in about 15 minutes. Thank you.


Daniel O'Day

Very good. Thank you. Welcome back from the coffee break. Good morning, everybody, from my side. It's a great pleasure to have a chance to present in front of you here today. I'd like to thank you also for the discussions we've already had around the coffee breaks and last night. You have certainly provided me with a warm welcome, which I appreciate very much. I realize in my new role, I was asking more questions than giving you answers. You can understand that after Hal and David's presentation, I still have a lot to learn having been away 7 years from pharma. But it is sinking back into me very quickly. But I just want to say right up front that I consider this audience to be a key stakeholder for me as I start my new role in the organization, making sure that we're good communicators on our portfolio moving forward. Our business model, and certainly look forward to your ongoing interaction with you about the ideas you have about our business as well. So thanks a lot for that.

I come up here with 2 minds today, really, one, very excited about the new opportunity in pharma. It certainly is an exciting time to be joining pharma, to be leading pharma. The momentum we have and my goal will clearly be to keep that momentum going and to continue to drive our continued success in the pharmaceuticals organization. On the second hand, though, I have to say this is the -- perhaps the last time I'll have the opportunity to talk to you significantly about diagnostics, and so a bit sad that this is the last chance I get to talk to you about the value of diagnostics at Roche, the value that I believe the industry brings to health care, the value that it brings to Roche, which I want to communicate today. And then thirdly, hopefully, as you know, it's been a common theme of mine, the value that I think it should bring to the valuation of Roche and the differentiation of Roche and eventually to your spreadsheet. So I'll continue to use that as a theme as I move forward as well. But with that in mind, just to give you all a little bit of a perspective again on what is Roche Diagnostics, why are we the market leader in diagnostics and why do we feel comfortable we'll continue to grow faster than the marketplace? And as you can see here, one of the key aspects of our competitive advantage is that we have a very broad range of technologies. And you heard already from Hal and others the importance of following the science. And I believe what this gives us is an ability to really truly follow the science. Whether it's in the case of genes, whether it's proteins, whether it's genes or proteins in the context of tissue, we are world leaders in each one of these technologies that allows us to leverage that to the advantage of patients, to the advantage of health care. And I'll be going over a number of different examples for that today.

There are 2 numbers I want to continue to remind you of, it's 2 and 70. Diagnostics today contribute about -- only about 2% of overall health care GDP cost, and they contribute to about 70% of decision-making. When I think of the disparity in those figures, it just screams potential for me. And I believe Roche has one of the best potentials to be able to demonstrate the continued value that diagnostics can bring to the efficient use of health care resources at a very efficient spend. And of course, we want to increase that 2% to something larger, and I believe we can do that by showing the medical value of our products and showing the reimbursement capability of our products as we move through our pipeline in our portfolio.

So a reminder on the marketplace, attractive market, it's a CHF 50 billion market worldwide. We represent 20% of that or around CHF 10 billion. It's been growing consistently over the past, and we continue to see that growing in the future. And you also see in those areas that are further taking advantage of the molecular understanding of disease, Molecular Diagnostics, Tissue Diagnostics, immunoassays, those are growing disproportionately in our portfolio, in many cases, high-single digits to double digits. And we see that type of innovation very much continuing as well in the future.

From a geographic standpoint, we have growth in all of the regions worldwide. And we're going to get into this in the afternoon session. Michael Heuer from our EMEA LatAm head organization will be in the breakout, combined with Luke Miels to talk about the strategies in our emerging markets and why we see this very, very strong growth in those areas. Having said that, we continue to see growth in the established markets as well, and I think that's driven by the uniqueness of our portfolio and what it brings to customers and patients.

So just to take a step back for those of you that may still be new to diagnostics. Just to explain, the business model is different than pharma. It took me at least 2 of my 7 years in diagnostics to learn this, so I feel compelled to repeat it. But it is a razor, razor blade model. It relies less upon the certainty of intellectual property, more upon the competitive dynamics that go along with having a significant large install base out there with our customers that's sticky, that's hard to replace, and then the reagent and consumable flow-through on that, which in that model represents the large majority of our revenues and profits as we move forward.

Roche clearly has the leading install base of any company in the world out there in the marketplace across 130 countries. And that gives us the footprint to be able to allow us, I believe, to put the medical content on our platforms that drives our business as we move forward into the future.

Now this strategy plays out in 2 different major ways, one is a strategy around the instruments themselves, the reliability of those instruments, the speed of those instruments. And that is contained in what we call the testing efficiency part of our strategy. This is, some people say, the basis of our business. It's important today. It'll be extremely important in the future with the changing needs of our customers. And so we have to focus and do that well. On the other side, we've got the medical value side, which again I think lends itself particularly well to the Roche strategy of innovation in health care because this is something that in the diagnostics world is still relatively new, and it's represented by this graph. And it's also what I believe will drive the value of diagnostics in the future. So I'll give some examples of that, but it's really about what value does a particular test bring to a patient, to a health care system in terms of driving the better management of disease, the better utilization of health care resources. And that's something that I believe that Roche is in a unique position to leverage the expertise of the group. Whether that's in clinical development, whether that's in regulatory, to be able to exceed the competition in this area.

I'll give you examples of each one of those to try to make them a little bit more real for you. Testing efficiency can manifest itself in many forms from our very largest labs to an individual device used in a patient at home. But just to show it in the context of our largest business, this is in the case, Professional Diagnostics, our Serum Work Area business. I mean, to give you an idea of what this means, literally you have labs at least the size of this room, sometimes manyfold larger that have instrumentation that would fill the entire room. And of course, the key is because our lab customers, as you know, are consolidating out there, their volume is getting larger, and therefore, they need an ability to automate their workflow and get a significant reliability in the results because also the ability to hire and to recruit lab techs is becoming more and more difficult out there in the marketplace. So we literally have systems, and you can see them here. I mean, they're fascinating to see, where you have everything from the sample coming in the door at the lab, to its storage eventually after complete processing in a refrigerated unit that is handled from an automated perspective. And this is done in our Serum Work Area or Professional Diagnostics area, probably to the most and the highest level of sophistication because you literally have thousands and thousands of samples a day running through these labs. And so we have to continue to excel in our product offering here to make our systems sticky, if you like, in the competitive advantage we have out there in the marketplace. Now there are literally hundreds of programs going on in Roche Diagnostics on testing efficiency. I just picked 2 to give you an example of this. One, a product that we just recently launched in our VENTANA Tissue Diagnostics area for digital pathology, VENTANA iScan HT. And if you like, this is at the -- probably the lowest end of automation of any diagnostics. If you walk into a pathology lab, you literally have people looking at slides in microscopes in a very manual way. What this instrument allows you to do is to digitalize those slides, which allows for better diagnosis because you can see things more clearly. It allows you to put algorithms on there, which takes away human error and allows you to store these slides or share these slides across the world. It's a center of excellence. So this is a very important advance that we just recently launched in a high-volume way.

The second example I have up here is a product that's rather in development right now. But to the example I had in the last slide about the Serum Work Area, Molecular Diagnostics is still actually a fairly young diagnostic science, if you like. It's only been around for 20 years. And there's still a lot of manually automated process, multiple instruments. And as the leader in Molecular Diagnostics, we feel compelled to take that to essentially the same level as we have in our Serum Work Area. So we'll be launching 2 new instruments over the next several years that will fully automate the molecular process, that will actually be designed to connect in from a lab perspective into the pre-analytics and post-analytics of the Serum Work Area environment and allow us to really take Molecular Diagnostics, if you like, into the 21st century.

So this is 2 examples of what we mean by exceeding and excelling in the testing efficiency side. For today, I wanted to spend a bit more of my time on the medical value side because it links in very nicely with the talks of my other colleagues. And again, it's intuitive in pharmaceuticals, if you would look at unmet medical needs and follow them. In diagnostics, it's still an evolving discipline, if you like. But we systematically do this. We also do this with our colleagues in pharma by utilizing their ad boards and other expertise we have within the organization to look at different disease areas and say, where within that context can we find unmet medical needs for diagnostics that have the potential to provide the most value? This is just a few examples, some of which I'll talk about today. It's certainly not inclusive. But you also, in diagnostics, have to think about the horizontal spectrum here, which is diagnostics can play a role, everything from early-stage screening, and therefore, potentially prevention of disease, to disease diagnosis, to therapy monitoring. And each one of these has different values depending on the nature of the need and the particular disease state.

So I'm going to go through 3 examples to demonstrate what we mean by medical value in diagnostics and why I'm so bullish on the future of diagnostics being able to contribute broader to health care society. The first one is an example also of how we lifecycle management a project. NT-proBNP is a marker for heart failure. Today, it's used in the emergency room setting, if somebody comes in, to rule or rule out heart failure in an acute situation. What we have now an initial study on, a small study that's been published, and now we're entering into collaborations with Duke and the National Institute of Health, is to take this context out of the acute setting and more into guided therapy management in heart failure patients. Two things can happen from this depending on the outcome of the study. First of all, it takes an established test and an established platform and allows you to get into a much broader usage of this. And the other thing is, it's not just a onetime test like it would be in the acute setting. You would use this repeatedly over periods of time in larger patient populations to use this to better guide therapy to reduce hospitalization in cardiac events and improve overall health care. Again, for a test, NT-proBNP, that is one of our higher-priced tests in terms of our immunoassay. It's probably more around the CHF 8 mark versus the CHF 2 mark. But remember, we have more than 1 billion immunoassay tests a year that are produced on our platform. So the volume that goes through this and the commensurate value can be significant as we expand these indications.

A second example, also in the cardiac area, is troponin high sensitive. Again, today, predominantly used in the acute care setting, this time for myocardial infarction and now taking it into a much broader population and in much more frequent use. Very interesting study, just recently published in JAMA, still an ongoing trial, but an interim analysis of more than 15,000 patients in a completely different setting. These are noncardiac patients, and what was looked at in this study is 30-day postsurgical noncardiac mortality. And what's impressive here is that you see a pretty significant correlation between the troponin high sensitive level and the 30-day mortality area. Now clearly, we need to do more studies in this area to further differentiate what can be done with measuring this assay. But if you can imagine, the population that exists for noncardiac surgeries and the ability to actually significantly reduce or triage and identify those patients that should be followed versus those that shouldn't, this represents I think and why, again, some facts why I'm so bullish on the ability to take these tests, take them into new uses and demonstrate the medical value.

And then the final example that most of you have heard me talk about before, but something I'm also very passionate about is cervical cancer, the second-leading cause of death in women. We know if you find it early, you have a very good survival rate. If you find it late, a very bad survival rate. We know the Pap test that exists today is about 50% sensitive, and that presents a real opportunity, I think, to continue to improve the patient management in cervical cancer.

We have demonstrated in the largest-ever clinical trial done in diagnostics, 45,000 women, the ATHENA trial that when you identify HPV 16 and 18, you can identify 1 out of 10 women that are missed with Pap plus pooled high-risk genotyping in a statistically proactive way.

Now obviously, we've already begun to change the guidelines in the United States. Changing screening protocols around the world takes some time, and it takes strong data. My point is, I believe we have the strong data. Now we have to get it out there into the thought leaders' area and really change the way that cervical cancer is screened out there. This combined, and again one of the reasons why I talked about the multiple technologies, because you attack these disease states with multiple technologies. Within our VENTANA Tissue Diagnostics area, we acquired an asset around the p16 biomarker, which has very high sensitivity, very high specificity. So once you've screened the patients, you can literally triage to a certainty in terms of diagnosis and therapy monitoring. And we're developing these products together in common clinical trials to really essentially, eventually replace Pap smear overtime with much more accurate diagnosis to make a difference in health care.

So a few examples on that. From the clinical work, which I think was the entry point to be able to revalue diagnostics, we are now at a stage of using the pharmacoeconomic data associated with that to provide greater value to the health care system and of course, back to the pricing and the ability for us to get reimbursement for products with this high medical value network.

So it's nowhere more obvious than in personalized health care that you're going to have a high level of medical value and medical differentiation. And one of the great benefits that we have within Roche is that we clearly have, as you saw from the portfolio, some of the best opportunity to be able to drive medical value with diagnostics that are specifically used with therapies. This is a picture I always like to look at. But as of the end of last year, we had more than 200 collaborations within the Roche Group. I think it gives fact to what Severin had talked about, about this concept of being able to freely work with scientist to scientist at an early stage, not be encumbered by IP, by contracts. And although we work with external pharma, we clearly don't enjoy that same level of freedom, that same level of cooperation and the same level of output and result, as you see here.

The pipeline, as we know, is rich. And more than 1/2 now, 11 out of the 18 have companion diagnostics associated with them in the late-stage portfolio. I thought I'd take 4 examples. And I think the thing to think about with these 4 examples, and my colleagues will get into these in even more detail in some of the breakout sessions, is remember the concept of having multiple technologies go to work here. And also very importantly, we often think of companion diagnostics only in the oncology setting. I also picked 2 examples outside of oncology that we could see as well that this understanding of disease is permeating many different disease states now.

So we talked a lot about the HER2 franchise. And of course, the most important thing about being able to target patients when you get the patients is making sure you're getting to the right patients. We recently launched with VENTANA, the world leader in Tissue Diagnostics a dual assay, an assay that allows you to look at both IHC and ISH in one assay in a very clear and definitive way. And we presented some abstracts on this. There'll be publications later to demonstrate in those patients that have -- we have difficulty in determining the HER2 status, this test has been more successful in identifying HER2 patients accordingly. To the degree that we identify about 4% more than FISH alone. 4% on a population-dynamic basis makes a big difference. 4%, if you're one of those patients in the 4%, can mean the difference between life and death. So it's extremely important that the quality of the test and the reproducibility of the test is at the highest level as we continue to penetrate patients with HER2+ breast cancer in this case.

That's an example from our VENTANA organization. The second one is one in our development, and I'm going earlier and earlier into our portfolio. You saw the data that Hal presented on MetMAb. And what's extremely important here, at least the hypothesis is, it's obviously very important to identify Met high and Met low, not just because you're identifying patients that can respond, but you may also be identifying patients that actually can have a deleterious outcome from the intervention with MetMAb.

So having a highly reproducible assay, again, with our VENTANA organization in this case combined, because their trial is combined with Tarceva to also look at EGFR mutational status, which we will look at in terms of, do you see a difference between patients with EGFR mutant status or not. Again, 2 business areas working together. If you had to do this with another company, you'd have to work with 2 different companies. In-house, we have the ability to freely sort of collaborate, connect, have highly validated assays and bring these to work quickly in the clinical development process.

Moving earlier into the value chain and outside of oncology now, we go away from our 2, if you like, gene-based businesses to our protein-based business, our Roche Professional Diagnostics to the asthma compound lebrikizumab. And looking again at the early-stage collaboration that went around identifying the Periostin assay, being able to file the IP on that assay and then be ready to move that program into late-stage development was, I think, integral to the 2 companies working together. Also we are now looking at much larger potential disease population of patients out there, right, from the thousands associated with oncology to very large, broadscale primary-care diseases. Again, that lends itself, I think, to the strengths of Roche Diagnostics. Because as you can see, in this particular business area, we have more than 40,000 instruments around the world, and we have different volumes of instrument. But the important thing is, that every single assay, the Periostin assay that were to go out there on our instruments would get the very same results regardless of where you did it. And of course, from a patient access perspective, having 40,000 instruments around the world assures that we won't miss patients, and that we'll be very convenient for patients to be potentially diagnosed on this depending on the outcome of our Phase III trial.

And then perhaps even going a little bit earlier into the hypothesis generation, looking at 2 new immunoassay markers in Alzheimer's. I'll leave most of this for Luca and Mike in the breakout session. But just to say that, again, this would be the identification of patients potentially with early Alzheimer's disease, which allows you to both identify patients out there at an earlier-stage disease that can benefit from patients and then also potentially link it in to the action and mechanism of action for our products. So more on all of these things, but I wanted to give you an understanding for why I always talk about breadth of technologies and the value of working within the Roche Group on these.

So finally -- I mean, as the #1 company in diagnostics, and as Severin said, not all innovation occurs within Roche. We're well aware of that. And also from a diagnostics perspective, we need to be collaborating with multiple companies out there to make sure that we have the strongest portfolio on our platform and also, frankly, can still be a good partner to internal Roche. And we have more than 30 collaborations out there today. Some are public, some are not public. But the very same benefits that Roche pharma sees in working with Roche Diagnostics, the other partners see as well. And of course, we amortize that, as well as we explore the external environment.

And the final question that I wanted to try to capture on one slide. I often get the question, yes, Dan, but isn't all the value obtained on the pharma side for a companion diagnostic? You have small reimbursement for your test, an individual test, and then maybe a small patient population. Again, think back to that large lab that I talked about that has Roche instrumentation throughout this lab. If you're a lab director, the last thing you want is to bring another piece of equipment into your lab to have to run an esoteric test that may be needed from time to time. So even though these may be smaller volume tests, in some cases, these companion diagnostic tests, although as I mentioned as we go outside of oncology, they may be larger volume, it's extremely important to provide a complete menu to labs out there in the environment because they don't want to train new techs. They want to be able to take that same large instrumentation, take the vial, put it on their instrumentation and get a good reliable result whenever they have it. This is extremely important to our business model. It allows us to differentiate from the competition. It allows us to get out of any commoditization that could go on within diagnostics, because the workflow, because the need to have this complete menu is critically important.

So another reason why I also think and a fundamental reason why we are able to continually stay ahead of the competition in terms of our growth rates overall.

So with that, I just want to close by saying I believe we have an extremely strong business in diagnostics. I think there's tremendous value still to come for the health care system and within Roche. And I hope you see that, particularly with where our strategy is going. We enjoy unique benefits that allow us to advance programs faster, get them to the market faster and penetrate them better into the marketplace as well.

So thanks a lot for your attention. With that, I'll turn it over to Alan to take the next section as well and be back up for questions later.

Alan Hippe

Wonderful, Dan. Thanks a lot.

Daniel O'Day

Over to you for diagnostics.

Alan Hippe

For diagnostics, yes, just a part of it here. I have one slide on diagnostics, too. Yes, hello to everybody. Great opportunity to be here and to talk a little bit about, let's say, the financial view. And you will see I will close the loop on one hand certainly to what Severin has said in his introduction and what he talked about when it came to R&D productivity and R&D funds allocation. But it will also refer to things that were said from the other speakers, just to show you that we have quite some examples on hand for the specific topics I'm going to talk about. And this is my agenda. I will talk about capital allocation and R&D and a little bit how we distribute funds, yes, in the long and in the short term and also across disease areas. But then as a major point will be how we monitor the value creation of the pipeline, yes, and how we look at this, and I will give you a conceptional view. And the other point is continuous productivity improvement, and here it will be a little bit about innovation and efficiency. And as we all know, these guys are not enemies, we have to bring them together. And in best case, they marry and we will see whether we get there. So I will have a couple of examples that we have really done here major efforts, yes, to bring that really to life. And last but not least, a focus on cash generation and allocation, and I have also included the famous dividend chart.

When we really look at the core operating margin, the profit margin and also the degree of diversification, it is very clear, the more focused you are, the better your margin is in our industry. And that comes across very well, and we have here in this slide compared that with our peers. And we have also put a regression into it, and there is some, how should I say, some space to maneuver. But very clear what comes across is that we are well positioned. We know that we have to really deliver more here compared to companies which have a higher degree of diversification, but we do well. And even when you really look at the regression line, we're well positioned and come up with a good return.

The other point, and I cannot mention this often enough. I have on a regular basis, a lot of discussions about our R&D and the R&D on sales ratio. And as you've seen today, our company is really about differentiating products. We want to come up with good data, convincing data, and we want to help patients and improve their lives. And if we achieve that, you see there is a very good result in case of M&D and G&A on sales. And when you put these 2 things together, we come up with a fantastic margin, and this is what our business model is about.

Now I come to that point about R&D and value creation in R&D and how we measure that. And first, I have to explain a little bit the conceptional basis I'm going to talk about. You see here a redline, a thick red line, and what is that thick red line really all about? In fact, it's an efficiency line and it represents a breakeven. What's in that line is, yes, it's constant costs, and it's fully loaded cost. And fully loaded means, says R&D in, says M&D in, so G&A. But on top of that, there's also the cost of capital. So whenever you're located on that line, and then it represents really a combination between the number of NMEs launched in a year, including line extensions and the respective peak sales. So the conclusion is, whenever you're above the line, you create value. When you're below the line, you destroy value. And to validate the model, let's do a projection into the past. And you see the projection here is about the last 10 years, 2001 to 2010. When you look at the facts, we have launched in the last 10 years, 2001 to 2010, in that time frame 1.1 NME per year. The related peak sales, certainly not risk adjusted, these are actual figures, are CHF 2.1 billion in average. And with that combination, we were very successful in case of value creation. I know you're not very interested in the past, you're much more interested into the future.

And here we go. Certainly, our costs have increased. We spend today much more on R&D than we did in the last 10 years. We have also in other areas higher costs. And certainly, we also have a little bit higher cost of capital. So what happened is that thick red line has shifted. And today, we are aware of the fact that we have to come up with a higher combination of peak sales and NMEs. I think that's a relatively trivial outcome. But the major point for us is, certainly, we are also projecting our pipeline into that concept. And you see here that green space, where the planned launches 2011 to 2016 are, let's say, mentioned in. And in this area, this is where we are located at the moment. Certainly, we do that with all kinds of scenarios. So whenever we have a failure, when we have a success, we can adjust the model, and we can make sure whether we are still really in that green space or whether there's a certain risk here that we fall below the red line. And that's really how we monitor the value creation that comes from our pipeline, and so far as you have seen quite successfully.

With that, let me make a couple of remarks about how we distribute our R&D funds on the pharma side. And what you see here is on the left-hand side, on the Y axis, the Roche budget for Roche pharma. And you see on the other hand here, on the X axis, you see really the probability of success. And it's an industry average from Phase 0 to registration. And what you see is the vast majority of our funds, as Hal mentioned, is spent in oncology. It's around 50%. That's what we are spending in oncology. And you see the probability of success in that area, it's an industry average, is relatively high. So it's really our bread-and-butter business, and really getting with that into the future. And you've seen that our Roche success rate compared to our peers is pretty good, and I think that's one explanation for it. But the other point is, certainly, we are also investing into disease areas like CNS and metabolism, which have a much higher risk related to it. But that makes sense. I think this is really where the science comes through and where we have to ask ourselves where is the next cycle coming. And it could well be that CNS is the next cycle and Luca will talk about it in his session and hopefully, we'll convince you that this is the right area to invest in. But the major point for me is, when you look really from a portfolio view on our spend, I think we are well balanced. We are really well balanced here for the short term, as well as for the long term when it comes to disease areas, and we monitor the risk for that.

Another view is certainly when you go through the value chain. And when you ask yourself, how much do we spend really in early, as well as in late stage. And you see the relation is roughly 50-50. That's what we can come up with. So 50% of our funds is really spent for the future and 50% is spent really for the near term. That's how we look at it. And another takeaway from that picture is, that the vast majority of our spend is done in Phase II and Phase III where, in fact, the de-risking has happened to a certain extent already because we all know the success rates go up in these phases. Pricing is really also here, I have to say, when I looked at it. To me, intuitively, it looks really well balanced, and I think it really shows that we do the decisions at the right spot and at the right time. And Severin has talked about our governance structure and certainly that has contributed to that picture.

Good. With that, I would like to talk about innovation and efficiency and these 2 guys and how they do. And this is really also, let's say, a more conceptual picture with an illustrative return. And we start really with the base, and then we increase the probability of success, and then really we reduce the time-to-market. That's the next step, and that's why we call, let's say, this innovation part. But also there's an opportunity to reduce cost, and we do that. And when Hal talks about development and how he would like to bring down the development cost per LIP, I think really, well, that's great and really amazes me, and that this concept is so embraced. When you look really at specific examples, Severin has raised it, I think really this decision on Nutley was really triggered by 3 different layers. One layer was really reduce the complexity and co-locate the management. A little bit more of a focus to Europe. The other point was really about resource allocation. We had an R&D review 2 days. We came out with conclusions out of this and brought this into the game. And last but not least, it was also about the infrastructure, because as you might know, Nutley hosted in the past roughly 5,000 people. Before we took the decision, it hosted roughly 1,000 people. And you can imagine what kind of infrastructure we had to feed, and there was no opportunity for us to bring that to a reasonable cost level. And when you take everything together, I think we have taken a pretty good decision. And what did that lead to? You know all that chart because I presented it and explained it in detail at half year, but you see really one point, the cash out and the related net savings. And when I have talked about Nutley, you see that was roughly -- or will be a CHF 510 million cash out. And we will have net savings of CHF 370 million. When you put that IT thing on top, this small gray bar -- this also applies pretty much to pharma -- then you have roughly CHF 620 million cash out and a saving of roughly CHF 400 million. I think this CHF 20 million saving in IT can go up, so roughly CHF 400 million, which really leads you, when you do the math, to a payback period of about 1.5 to 1.6 years. Well, the caveat is, this net savings will ramp up over time. So -- well, give us some credit. I think the payback period is more about 2 years. But I can tell you based on the experience that I have in other industries, I think this is a fantastic payback period. So I think that's really a measure that we have done perfectly well. And I think the same applies to diagnostics. When you look at the diagnostics measures, with a cash out of CHF 190 million and net savings of CHF 190 million, looks like a payback period of 1 year. But also here, we have a ramp up of the net savings. So also here, I would say it's around 2 years. But also here, we have a very, very respectable outcome and a very respectable and good payback period.

Just to remind everybody. Why did we do that? To really reallocate funds. You see we have feeded the pipeline with CHF 240 million and brought more into late stage. And the other point was really in Diabetes Care, we really use it to support the business, and we will spend some money in Applied Science.

Good. With that, I'm back to the example of how, yes, and here really I brought in even the target that we're having to bring really the cost per LIP down until 2015 by 15%. And why can we really bring that to your attention? Because we have a very, very substantiated plan, and we're pretty confident that we can deliver that number. And as Hal said, I think he went through all the points. It's really about making the processes better, streamlining the clinical trials, taking smart risks. I think everything comes in here and really leads, I think, to a fantastic result.

Another point is really Diabetes Care, another example of what we did, and you have seen the numbers. And one part of these numbers had come from diagnostics based on all the actions in Diabetes Care. And you have seen what happened to the market. It was quite evident for us that we had to adjust. So 3 major actions were implemented. The one was the restructuring and consolidating of the R&D organization, and what we are doing here is the blood glucose monitoring activities are focused on Indianapolis, whereby all the pump-related activities really will go to Mannheim and will be focused over there. And on top of that, we are prioritizing the portfolio in R&D, and that gave us significant savings to move on, and as I said, to support the margin in that business.

The other point is about One Global Operations structure, which really means and implies that we consolidate certain activities, but also is that we leverage best practices along the organization in a much better way than we did in the past. And last but not least, optimizing the M&D investment. The example I would like to raise here is we are the U.S. organization, which really brings together now the glucose monitoring sales team, as well as the insulin delivery system sales team. We bring that together and generate quite some savings and efficiency with this approach.

The last point is about cash generation here. And you can imagine, I think that's a great section to talk about. We have delivered. You know all the chart. You know what we are achieving when it comes to operating free cash flow. You know what we are achieving when it comes to the cash flow margin, and we have increased this margin quite significantly.

Let me talk about what we are now focusing on to make even further steps in that direction, and you also know the chart here, I present it at half year. We made a significant step in Southern Europe, and I will talk about Greece and Spain more specifically, but I can also say of today that we have also made a major step in Portugal to bring our exposure down, and it's really heading into the right direction overall.

Greece. What happened in Greece? And I go through these examples specifically just to illustrate that this is not a general approach we are taking. I think this is really a house-by-house battle. And we have to come up with specific solutions pretty much in every location. And it's not like that we have specific solutions for Greece and Spain, it even goes down into the regions, it goes down to the hospital level what we have to do. So very, very complex, pretty difficult and a lot of efforts to do. You see the history, you see the action, you see the impact. I would start with the history, and that's just one remark, I think in Greece, we were paid pretty much every 5 years. This was a constant framework we were following. So every 5 years, we had a reduction of accounts receivable based on the payments that we received. And during the financial crisis, you can imagine, this framework was not applied anymore, and so we had to take action. We had negotiations with the government. We got a government bond. Most of you know that story. We sold the bond right away, with a discount of 26%. If we were doing this today, I think the discount would be around 70%. So I think we did a great deal. But the major message here is, by moving forward and that's on the right-hand side when it comes to the impact, you see our accounts receivable situation since then is stable. Certainly, we have applied new concepts, like new payment terms and everything. But you see really that our approach was respected. And you might say even now, well, Alan, yes, for you it's simple because your sales go down the drain in Greece, and that's why your accounts receivables are stable. Not true. Our business is pretty stable also in case of sales in Greece, the only difference is we get paid now.

When you look at Spain. Spain, I think is another perfect example where you can implement really unpleasant actions. And in fact, the respect level increases. You see the history. Our business, high exposure to the hospital business, yes, also good sales, admittedly, and really accounts receivables piled. And you see it here on the action side really, the DSOs on the public debt side, even for the whole industry above our level. But nevertheless, we're clearly of the impression that we have to do something about it. So really, we implemented new payment terms. We went really with specific plans per region to the hospitals and implemented these actions, and we also applied certainly forfeiting. And you see what the impact is, and the impact admittedly was very much driven by the Montoro plan. But -- and you might know Mr. Montoro, the Finance Minister of Spain, and he had a great idea because he thought, okay, when I commit to the European Union, yes, that I pay a specific debt, I might get some money right away. And that was perfectly the case.

So what we had to do is, and you can imagine a lot of credit hurdles. We had to bring our 40,000 outstanding invoices on all these pooled credit lists for the Montoro plan. And that was a major effort we had to do, and we did it. We got the money, and you have seen how well we did in reducing our exposure. And also there, we see at the moment a stabilization of accounts receivable. So I think our customers, really, how should I say, it's not like that our business goes down the drain, I think we're getting paid now.

Good. With that, what is that leading to? It's really leading to the point that our cash generation is appreciated in the market. And it really leads to the point that our rating is significantly better, yes, than most of our financing counterparties we're dealing with, and this is just a selection. It's just a selection of -- by accident that it's here, of peers in this area. But I think what does it tell us? First of all, we have really access to cheap financing. We can really lower the interest expense and with that we can really increase the cash flow generation.

Yes, having said this, we have to look a little bit at the balance sheet and the net debt. And I got a lot of questions here yesterday about whether this de-leveraging is still something we're after. And I have to tell you, yes, we are because still we had a high level of leverage in the industry. And when you look really at net debt or net debt on total assets, and we have been at 25% at the end of 2011. At half year, we have been at 29%. So really, I think there's a way to go to get to that corridor between 0% and 15% net debt on total assets. And that's what we are heading to, and that's where we would like to get to. And we will see how it works and what opportunities are, yes, which might come along on the course of this development. But I think for the time being, I think that's what we are after and we have not reached this corridor.

The other point is the famous dividend chart that I've announced already at the beginning of my presentation. And you see the history of dividend payments. You see the history of payout ratios. And for the future, all I can say is, we are committed to an attractive dividend policy. Yes, and you might ask us later on, what an attractive dividend policy is. And we will tell you, it's an attractive dividend policy, just in case you've had already your question.

Okay. That brings me to the conclusion of my presentation. And the major point really is, well, we're an innovation science-based company. There is no doubt about that. But I think really where we can also make a difference, that's not really, how should I say, we cannot win that game just on the basis of efficiency. But we can free up funds and bring really cash to areas and spots where we have a better return. And I think that's what we should do, and I hope my examples really explained what we are doing, how we think about it and hopefully, it also increases the credibility. Innovation drives the sales growth. The efficiency drives the profitability. That all leads to good cash generation and that, hopefully, then also leads to a better value for this company, because this gives us the opportunity to invest for the future and for the patient. Gives us the opportunity to satisfy you, and certainly, that is a benefit for us. Thanks a lot for your attention.

And now I would like to ask my colleagues to come to the stage for the Q&A session. Thanks a lot.

Severin Schwan

So let's go into the next Q&A round. I suggest that we have a Q&A session until a quarter past 12 before we go into lunch, so another good 20 minutes. Let's start right here in the middle, if we can have the mic in the middle, please.

Michael Leuchten - Barclays Capital, Research Division

It's Michael Leuchten from Barclays. A question for Alan and Severin. On the patient-based payment models that were outlined this morning that seems to require quite a lot of patient information and a lot of infrastructure to gather that sort of information. So my question is, who is paying for that? And what sort of investment does that require? And then what does that mean for SG&A going forward relative to where we are today?

Severin Schwan

You are absolutely right. If you shift from a cost per vial to indication-based treatments or rather reimbursement or even patient outcome-related reimbursement, you need a certain infrastructure. You need IT systems which collect the respective information, and which feeds this information back to the payers for the various reimbursement schemes. And actually, it has been one of the biggest hurdles to make progress with more value-based pricing and purchase. Typically, you have this conceptional discussion with payers, and they are very enthusiastic about outcome-related pricing models. And then it goes back into the organization, and they look at it, and they see how they potentially can implement it. And typically, what comes back is we like the concept, but we can't implement it, and as a result of that, let's stay with the old procedures. Having said this, it's starting to change. Having said that, we see now real action. I think Italy is a good example and perhaps, David, if you'll step in here?

David Loew

I mean, it depends, it's country by country. Some of the countries have build up those databases and tracking systems already. Italy with Zeneca has a tracking patient per patient established. They have offered other countries that they can tap into that software. The U.K. has started to establish a system already. Germany has built up a system. You have seen it with Avastin, for example, on breast cancer where now, for the breast cancer patients they track them, but they have realized that they will have to expand this. So we are working together with governments around the world now to help build up the infrastructure around this. And it is very clear medicine is going towards that because they want to collect more information about the patients, not just in terms of what gets administered, but also tracking what kind of outcomes come out of this. Registries have been implemented by the scientific community in the past, but the link between the 2, so that the payer can have a look at it is not completely established yet, and that is a gap, which governments have realized that they need to fill. So I think it's a collaborative approach. They realize they will have to pay for this. If we pay for it, of course, then it's going to be only for kind of Roche efforts that we're going to implement, and that's not going to be the future. The future is going to be the whole industry will get drawn into this. If you look at oncology, you have so many combinations that Hal has shown on his charts is that the medicine is going to move to combine more innovative treatments in the future together, and then they will have to think, okay, how are we going to pay this? How are we going to track this? And what kind of results do we get?

Severin Schwan

Thank you, David, can we have the next question, yes, in the same row, please.

Unknown Attendee

Just looking back to the chart you showed, Alan, about probability of success and R&D spend. And if you look historically at the industry kind of maybe 10 years back, the probability of success for CV and metabolism inflammation might have been where oncology is today. As you look and investor today for the next 10 years, what gives you the comfort that your success rate is going to be as high as it has been in the last 10 years given the improvement in standard of care?

Severin Schwan

Do we want to hand this over to Richard for oncology and perhaps, Mike, you can join in for oncology as well.

Richard H. Scheller

Particularly, in oncology, it's the understanding of the mechanism of the disease, which largely derives from being able to sequence DNA, to sequence the DNA of a tumor and compare that to the DNA of a normal tissue and find the specific mutations in the DNA that make the cell cancer and being able to target those specific mutations, which is what our pipeline is full of. So the understanding of cancer has increased beyond any other disease area by an order of magnitude. So I'm quite confident that the probability of success in oncology will be at least as high as it has been in the past, if not higher as evidenced by the kind of data that Hal was able to show you in the kinds of molecules that we're moving into Phase III.

Severin Schwan

Perhaps, Mike, if you could also add on it and perhaps also add on the different emerging modalities and the new tools we have to really make tracks and target -- new targets, make targets trackable, if you like, in addition to the scientific understanding.

Mike Burgess

So just to build on what Richard said, because obviously an important aspect is understanding the underlying genetic defect. But it's also understanding that for cancer, the defect lies in the cancer cell, that a tumor is an organ. So it's the combination of therapies, which lead us to believe that we will be more successful, not only understanding what treatment is necessary for the cancer cell, but how we disrupt the tumor as an organ, and that continues to advance. Now clearly, based on this type of work, what we have to bring is a much more sophisticated approach to the therapeutic modalities that we deliver. You saw that reflected in Hal's presentation where he referred to the good work on T-DM1 and the antibody drug conjugates that are coming behind it. But also, there are newer technologies which better enable us to drug what has historically been undruggable targets, technologies such as DNA-encoded library technology, which instead of having a small molecule library of 1 million or 2 million compounds, you have 10 billion compounds that are available to you for you to use. We have more sophisticated approaches with respect to antibody engineering technology, not only being able to direct towards a single target, but to direct towards multiple targets. Then there are other disruptive technologies coming around behind it such as radioimmunotherapy, cytolytic fusion proteins. And this diversity of therapeutic modality, coupled with the more in-depth knowledge of what's driving the cancer cell and what's forming the tumor lead us to believe that we will be significantly successful in the future.

Severin Schwan

Thank you, Mike. If we can have the next question in the next row just behind.

Unknown Attendee

I've got a few quick questions, please. Firstly, on diagnostics, how much do you think we'll be seeing open architecture systems versus very specific systems that only run your tests? I would assume that payers are very reluctant to become too tied in, but you would like them only to be able to use your specific products. So I wonder if you could address that a little bit. Secondly, on the capping and the insurance, you mentioned in one of the slides the potential situation in China introducing insurance there. I wonder if you could talk a little bit about pricing in that market. If you're able to pay a $50 premium and get the whole of your family covered, I mean, they have very big families, some of the people out there, you could end up with some very expensive bills. So I wonder if you could talk a little bit about that. And just broadly, since you priced a lot of your new drugs a number of years ago, loads more, unbelievably expensive cancer drugs have been introduced by other people. Should we assume that your next-generation of cancer therapies will be looking to target prices in the unregulated markets like the U.S. at the same sort of level that we're seeing for some of the other more recent launches? I mean, really taking you up to the well over $100,000 a year time point? I haven't seen anything in your slides showing we have fantastic additional qualies and things like that, which we used to see a lot in your presentations, but we haven't seen so much now. And one final tiny technical question on the end, you've talked about a 15% reduction in the cost per project. Is that on the cost per successful project? Or are you including in that improved failure or reduced failure rates? So if your effective cost goes down or is it just on one project reduced cost?

Severin Schwan

Right. Perhaps we start with the diagnostics question, open versus closed systems?

Daniel O'Day

Yes, absolutely. Just to differentiate the 2 marketplaces. In the research setting, in the life science setting, we have very open systems. And I think that's very important because you have a lot of rapidity of technology and evolution there to be able to make sure we have open systems. As soon as you get into the diagnostic setting, not only our, but other competitor systems are closed. And they're closed for a very good reason. And that is -- these are very complicated instruments. They're very complicated assays, and you need to be able to demonstrate from a regulatory perspective that you get the very same result no matter where it's done anywhere in the world. And so if anything, we see regulation increasing in diagnostics, certainly in the United States, but also in some of the emerging markets, and we could talk a bit about that. And the reason for that is because the cost of getting an improper result is becoming more and more significant, particularly when you're talking about guided therapy for personalized health care and being able to reproduce that result needs to be done in a closed system that allows you to assure the reliability of the results. So from an IVD perspective, I don't see systems opening up. I don't think that's a trend we'll see. I think we'll see closed systems. So even more important that we have the most complete menu on our systems, and I think that also leads to the competitive advantage of Roche versus other companies.

Severin Schwan

Yes. I think it's important -- no, no, no. I just want to reemphasize, we have discussed this very often, this question, in which direction will the market go. I think it's a very valid question, and indeed, I just want to reemphasize, it goes into the other direction. It goes into closed systems. In certain Asian countries, in the clinical chemistry systems, there used to be more open systems in the diagnostics and the routine diagnostics and more and more we see those systems closing. So if there's a trend, it's in the other direction actually. Perhaps I can take a first run at the Chinese question in terms of insurance as a new segment and perhaps, David, you want to add on that. It's quite interesting. What you say is correct that there are very embedded family structures in China. And what typically happens for those drugs and many of our drugs are still out of pocket that the family is kind of the insurance. So if somebody in the family falls ill, then the people in the family would put their money together to provide the access to the medicine. And actually, what we have seen and there was one avenue we took, that there was a point -- and there was a -- like a growing segment of people who could afford an initial treatment for some time, but then they didn't have enough money anymore to complete, in particular, Herceptin treatment. When we saw that segment emerging, and then we understood the dynamics, that was the moment when we also introduced our patient access program that we provide the medicines for a certain period of time free after an initial period where the patients will pay. And that had a double effect. On the one hand of course, we increased the access to start with, but even more important were the physicians were more willing to prescribe Herceptin because very often the physician would know about the family situation. They would see the patient running into a financial problem and not start the prescription in the first place. With the introduction of these patient access programs, that actually turned out to be a real trigger to increase access and eventually our sales. Now the second avenue, which we have taken, I think it was about 2 years ago when we started to experiment with that is, why don't we take this concept of an insurance, which actually is a very embedded concept in the Chinese society. Why don't we take this on a more sophisticated level. Why don't we work together with insurance companies and provide insurance coverage for serious diseases like cancer? And the problem is that in China, foreign insurance companies cannot operate. So this is protected. So the insurance business in China is protected for local insurance companies. The market is not open. So it would have been difficult for us to kind of come with sophisticated concepts into the market. But what is possible in China is international reinsurance companies can come into the market. So whilst the retail business is in Chinese hands, at least today, the reinsurance business is where international players can participate. So what we did is we work together with reinsurance companies. And that was a very effective way to get into the market because now the reinsurance companies, they want to generate business, they were talking with the local insurance companies and said, we provide you a model of free insurance, so you don't have a risk, what you do is the retail of these policies. And it took a bit of time to get it going. And today, I think we have -- do we have 1 million policies target? I think -- Luke should know. Luke is sitting in the back. Perhaps, Luke, if you could make a comment on that? It's a real success story. Now 1 million people in China you could argue is still a small number, but it's getting material. It's getting traction. Luke, could you comment on that?

Luke Miels

It's around 4 million year-to-date.

Severin Schwan

Four million already, 4 million in China.

Luke Miels

And that's with 3 head count. So we work directly with the insurance companies. We provide the data so they can do their calculations. The $50 is for an individual in our company. So we provide that insurance, it's a retention tool. But we don't hold the risk. The risk is with the insurance companies. And again, it's a longer-term opportunity in China. It's not going to drive our business today; it's in the mid to longer term.

Severin Schwan

So it's a very interesting concept, I believe, and one of the pilots we are running to improve access in emerging markets. And I'm just looking on my -- yes, pricing for new products in oncology, where does it go? It's a very principled question. What is the right price for a medicine. And I don't think you can answer this in general terms. At the end of the day, it depends on the benefit you provide with the respective medicine. And if you have a cure for a serious disease, to take the extreme case, then I think it can have a very high price. And if you have a marginal improvement in PFS or in overall survival, you will have a harder time to negotiate a high price with the regulators, with the government and with the payers. So I really do believe that it is a matter of the additional benefit which you provide, and there is no absolute amount -- there is no amount where you say, here is the cut-off, and it doesn't go beyond that. It is a matter of the value which you provide. And the same is true for combination therapies. But having said that, we still need to remain creative. We still need to remain flexible because the purchasing power, the ability to pay is different across nations, across regions, and we try to adapt as much as possible as we can to that. But I wouldn't put in a specific number or a specific ceiling. Your very specific question, the 15% reduction, that is really including the attrition. So this is the complete picture. Good. Thank you very much. Let's take the next question. You already had a question in the last session. I have to be fair and go into the fourth row here. Yes, please. If we have enough time, I'll come back to you.

Seamus Fernandez - Leerink Swann LLC, Research Division

Seamus Fernandez from Leerink Swann. Just a couple of quick questions on the diagnostics portion. Michael actually mentioned a very interesting concept of what I believe would be tumor as organ system. How does that kind of work into your thought processes on processing of cancer-based diagnostics and tissue samples? Is there an evolution that we should be thinking about that Roche Diagnostics will be kind of pushing forward into? And how would the concept of sort of the broader pCR-based diagnostics that can get into physician's offices play into that over time? And then a separate question, just as we think about efficiency in R&D and to [indiscernible] earlier question, it does appear that as we go more towards personalized medicines that even though the bar for R&D efficiency may not be growing, the size of the market does appear to be shrinking at some threshold. So again, how do we kind of tie the 2 of those together or obviously you may believe that the size or the market opportunity may not be shrinking at the same time. So how do we play those all together?

Severin Schwan

Perhaps I can just give an answer on the last question regarding the size of the market. If you look at the products which are out there in the market, if you look into the products which are specifically in our late-stage pipeline, the facts speak against that. Herceptin is a highly targeted medicine. It addresses about 15% of women. Still, it's a drug which generates CHF 6 billion of turnover. I wouldn't call this a small drug. If you look at MetMAb, which was raised by -- which was discussed by Hal in his presentation, here we talk about 50% of all lung cancer patients. Just think a moment about this, 50% of all lung cancer patients, what that would mean if it works. If we talk about lebrikizumab, 50% of all asthma patients. So targeted medicine does not necessarily mean that this is a small market opportunity or a small number of patients. And it also works the other way around. Look at Erivedge. Erivedge is for all comers, and we don't have a companion diagnostic test with that and still it is a relatively small segment because it's a small cancer type. So targeted does not necessarily mean small. You have all comers, they have small and big opportunities, and you have targeted opportunities, which can equally be small and big. And then you had a question on diagnostics and how this will evolve in the cancer field.

Daniel O'Day

Right. Right. So I think the important thing about cancer diagnostics is, it is evolving and secondly, you need a breadth of technologies to be able to answer the questions that the clinicians have today. And I think that will change over time. And we've got to follow the science that Mike and Richard and Hal are producing. But today, what's true is that in the clinical setting, predominantly what's being used for diagnosis is pCR or tissue-based diagnostics. If you talk to Tom Grogan, the founder of VENTANA, he would be very clear to say, it's not one or the other, it's both. In many cases, you need to look at the genes in context in certain disease states to determine diagnosis. And in other cases, it's fine, as he would say, is you put the cells into a Cuisinart, you mix them all up, and you find just the base pair and the read length. So they complement each other. And today I think those are adequate for actionable decisions that we have in cancer today because pCR is pretty good or pCR panels up to about 3 or 4 or 5, 6 actionable mutations. Once you start to get much larger than that, then we start to see the role of sequencing coming in. But today, sequencing is predominantly a research tool. And it's a very valuable research tool in the future. And here, I mean, post 5 years as you have more and more actionable mutations that you can do something with, I think it will go into the clinic. But even today, where sequencing is being used in some of the major cancer centers, if you identify a particular mutational coverage, you're generally triaging to a pCR test to assure a certainty of that result, because of the sensitivity and specificity you get, and depth of mutational coverage that you get with the general sequencing. So we have plays in each one of these obviously, sequencing, pCR and tissue, and we've got to evolve those technologies to respond to the science is what I would say.

Severin Schwan

Thank you. We had a question here in the front. If we can have the mic in the third row, please.

Jeffrey Holford - Jefferies & Company, Inc., Research Division

It's Jeff Holford from Jefferies. Just on the financial side. Obviously, the transition to the leverage ratio that you're talking about is going to be relatively short term. So what can we infer about capital allocation post that period? Should we be thinking about other ways of allocating capital such as share repurchases, more aggressive M&A or is it just giving more freedom to be even more generous with your dividend policy?

Alan Hippe

I think it's very clear. I think I framed that a little bit. I think there's still a way to go. And as we've said in previous meetings, I think we will cross that bridge when we get there. I mentioned I think there might be a couple of opportunities, smaller ones. We are not big fans of big mergers and big M&A. But I think there might be a couple of supplements that we can really bring into our respective areas, yes, and we want to be focused. We want to stick to the business that we're having. So there might be some opportunities there. But besides that, I think we will see. I said the first target at the moment is to bring that leverage down to get to that corridor and everything comes next.

Severin Schwan

Thank you. We have time for one more question, perhaps in the back, in the last row, please. And then we go for lunch.

Gbola Amusa - UBS Investment Bank, Research Division

Gbola Amusa, UBS. I think something like 40% of cancer vaccines in the U.S. -- or actually something like 40% of all vaccines in the U.S. are cancer vaccines. And obviously, this concept is far from improving itself. But to the extent something or several products emerge in the coming years and people have increased confidence in cancer vaccine, to what extent does that threaten your overall oncology franchise and its sustainability?

Severin Schwan

Perhaps I can give you the general answer. We follow the science. If there are opportunities coming up in those fields, we would go for that. We would certainly be open for that. More specifically, for vaccines, I mean, we have focused on therapeutic vaccines. We have perhaps related to that a number of programs in the immunocancer therapies ongoing. We have not entered the field of preventive vaccines, and there are no plans at this very moment to do that. But we screen the market, we are open for such opportunities. And if the right one comes up, we would go for it certainly.

Right. Thank you very much, and we'll see each other later today in the breakout sessions.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to All other use is prohibited.


If you have any additional questions about our online transcripts, please contact us at: Thank you!