MorphoSys AG (MOR) 2003 Q4 法說會逐字稿

It's my pleasure to welcome all of you to our annual conference. My name is Claudia Gutjahr-Loser; I am Director of Corporate Communications of MorphoSys. I would like to thank you for your interest and participation at our conference today. With me are my colleagues, Dr Simon Moroney, our Chief Executive Officer, and Dave Lemus, our Chief Financial Officer.

Today we will present the company's annual results for the year 2003. At the end of our presentation, we will take questions from the conference audience and those participants listening in on the phone may have. We have planned approximately one hour for the presentation. Afterwards, we will answer your questions. After the conference, snacks will be served outside and we would be happy to invite you. There will also be a chance for some more discussions.

Before we start, we want to remind you that during this conference, we will present and discuss certain forward-looking statements concerning the development of MorphoSys core technologies and progress of its current research programs and the initiation of additional programs. Actual conditions differ from the company's assumptions, actual results and actions may differ from those anticipated. You are therefore cautioned not to take [indiscernible] undue reliance on such forward-looking statements that speak only as of the date hereof.

I would now like to hand over to Dr Simon Moroney, our CEO who will discuss the highlights of 2003 and the outlook of the year 2004.

Thank you Claudia. I would like to add my welcome to those of you joining us here in Frankfurt today and also to participants who are dialing in. Over the next hour, we will give you a comprehensive review of our year 2003, an outlook for the future, and a detailed presentation of all of the financial numbers.

We had a number of significant operational achievements in 2003. The therapeutic collaboration that we signed with Pfizer shortly before Christmas was certainly the highlight. We also made very good progress in our other existing partnerships. We generated promising pre-clinical data in our own proprietary product programs. We secured at the beginning of the year an exclusive worldwide license from Boehringer Ingelheim to antibodies against one of the central targets, ICAM-1, and we signed a manufacturing agreement with Lonza Biologics. This altogether points to successful execution of our business strategy.

On the financial side we had a very successful year. For the first time ever, MorphoSys was cash flow positive from operations. We significantly reduced our cash burn. We achieved an EBITDA positive result for the first time ever and the company's balance sheet was substantially strengthened. In other words, execution of this strategy is also bringing us short-term financial benefits.

2003 was a big year for antibodies in the biotech industry. Four new antibodies were approved for marketing and we saw a number of important deals in the sector. Indeed the respected US investment bank, Burrill & Co, quoted our deal with Pfizer as one of the three most significant antibody deals of the year and pointed to an increasing appetite amongst big pharma for large molecules and, particularly, antibodies. We see this as being also a significant trend for the future.

It is also a return to some of the cash rich deals that we have seen in the past but which dried up over the last couple of years. Our strategy, as you should well know by now, is based on our proprietary HuCAL technology and the commercial exploitation of that technology. We see that strategy very much as shown on this chart, namely that we have activities on the left hand side in the therapeutic field and on the right hand side, a new move for us in the non-therapeutic field. In the therapeutic field we have two types of activities - partner-initiated programs and MorphoSys-initiated programs.

On the right hand side in the non-therapeutic area we started a new initiative last year in the reagent area and we also in some of our collaborations have some diagnostic activity. What I want to do is spend some time on three of these areas - namely partner-initiated, MorphoSys-initiated and the reagent space and explain to you what we have achieved over the last 12 months in those areas.

I want to start by looking at the partner-initiated side of the strategy. The nine companies shown on the next slide are our therapeutic antibody partners. Of course a number of these partnerships have been running for several years. The most recent addition, as I mentioned, is Pfizer, since Christmas of last year. When you look at these partnerships, one of the striking features is the range of different diseases that we are working on. We have programs in cancer, infectious diseases, inflammation, in autoimmune disease and even in Alzheimer's disease. That speaks to one of the strengths or advantages of antibodies as therapeutics, namely that they can be applied in a range of different indications.

These deals are all different, but all have common elements as well. They all comprise an upfront payment of some form. In the case of Schering, up to €24m in an equity stake. They also all comprise of annual license fees in some form or other. The financials on a per program, and that means the per therapeutic antibody program basis, can be summarized as shown on this slide, namely that we receive research payment per program of between €0.5m and €1.5m. We expect to receive cumulative milestone payments per program of somewhere between €7m and €11m. When products come to market, we expect to receive royalties in the mid single-digit range.

I want to give you just one example out of the some 17 or 18 running programs with these therapeutic antibody partners and that example comes from our collaboration with Roche and is data that we published together with Roche in the autumn of last year. The Roche program is focused on making human antibodies to treat Alzheimer's disease. For the first time in publishing these results, we were able to disclose the identity of the target against which the program was directed, and that is Amyloid-ß Peptide, the key component of the plaque, the tangles of plaque, that form in Alzheimer patient's brains. When Roche originally came to us with this project, the challenge was not only to make an antibody directed against this target, but to have that antibody cross the blood-brain-barrier and somehow dissolve these tangles of plaque that occur in Alzheimer's brains.

This was an extremely interesting and challenging project for us and I should point out one that we would never have initiated without having such a collaboration with a company that has extensive experience in this disease. Our initial task was to make an antibody that bound to Amyloid-ß and that had sufficient affinity to lead to the dissolving of the plaque. The initial antibody that we generated had the right specificity, so it was able to bind to Amyloid-ß, but it had a rather moderate affinity. The number 200nM there is somewhat moderate in terms of affinity for target.

However, and this is the key advantage of HuCAL as an antibody technology, we were able to use the modular structure of HuCAL, which has been designed into the HuCAL technology, to improve that infinity, while maintaining the specificity. The affinity we improved over 30-fold in two steps, and then when we converted the antibody to an immunoglobulin, we achieved improvement overall of in excess of 1,400-fold - a very dramatic increase in the binding strength of the antibody for its target - while maintaining the specificity.

So step one of the project worked extremely well, and is a beautiful demonstration of the power of the HuCAL technology.

Step two, of course, was to see whether the antibody has the desired effects. I want to show you two pieces of data here, first in vitro data on the left hand side, in which Amyloid-ß Peptide, which is labeled, and which is allowed, in a test tube, to form plaques, is then challenged with antibody to see if the antibody can break up and dissolve these plaques.

Without going into the data in great detail, what you see here is, with increasing concentration of antibodies, moving to the right, the amount of aggregated Amyloid-ß is strongly reduced - the lines go down to the right. In other words, in vivo, it is possible to show that the Amyloid-ß plaques can be dissolved.

The next question, and this is perhaps the most challenging aspect, is would it be possible to administer the antibody to mice, in a mouse model of Alzheimer's, and have those antibodies cross the blood-brain-barrier into the brain and bind to plaque in the brain of these mice models of Alzheimer's disease. The figure on the right hand side shows that, indeed, this is possible. What was done here was the antibody was administered intravenously to mice, which were genetically modified in a way to represent human Alzheimer's disease. The mice were then sacrificed and sections of their brain was taken and then stained to see whether our antibody had reached its target.

What you see here are two colors - the red color is staining for the MorphoSys antibody, and the green color are those parts, which our antibody has not reached. So you can see that the majority of the plaque in this particular section has, indeed, been successfully labeled by the antibody. In other words, the antibody administered intravenously has crossed the blood-brain-barrier and has reached its target.

This was the data that we presented jointly with Roche in October, and which is the basis for ongoing work with this antibody at Roche. It is, in our view, a very nice demonstration that, not only can we make antibodies to precisely defined specifications, but that such antibodies can have in vivo effects.

That's just one example from the currently 17 ongoing partner programs, which are summarized on this slide. Six of those programs are currently at the pre-clinical stage, 11 are at the research stage, and what we indicate on this slide with the yellow diamonds are published milestones in each of those programs. If you were to go back and look through our press releases, you would see, for each of these diamonds, a reference to a successfully achieved milestone in one of those programs.

We are very proud of our success record in our collaborations and we think this speaks to the power of the HuCAL technology. We have a very good track record in achieving the objectives and the milestones that are set out at the beginning of the collaboration and again, this is not only a tribute to the people doing the work at MorphoSys, it's also a tribute to the strength of the technology.

I want to turn now to talk about our own proprietary programs, and summarize on page 12 the three programs that we are currently focusing on, namely MOR101 for burn, MOR102 for chronic inflammation, and MOR202 for cancer. Our strategy for our own programs is, as we have communicated some time ago, namely to take these programs to the point of proof of principle in vivo - in animal experiments - and then to seek a partner for subsequent development. In doing this we can minimize our risk and our expenditure while maintaining, through an option, the ability to join for co-development at some subsequent stage.

The first two programs, MOR101 and MOR102 are directed against the target ICAM-1. This was the subject of the exclusive license that we secured from Boehringer Ingelheim early in 2003. We chose this target, which we find particularly attractive, because it's a rare example of a clinically validated target. Boehringer Ingelheim worked with a mouse antibody called BIRR-1 and showed preliminary evidence of efficacy with this antibody in burn and in the treatment of rheumatoid arthritis.

However, they chose to stop development of that antibody because they encountered some toxic side effects that turn out to be related to the antibody and not to the target. We, with our technology, were convinced that we could use the validation that they had provided for the target, but by making a better antibody, we could avoid the problems that they encountered.

ICAM-1 is widely implicated, not only in the diseases which Boehringer looked at in the clinic, but in addition, in psoriasis, in Crohn’s disease, and in diabetic retinopathy. What we have done is we have looked at the antibodies that we have generated using our HuCAL technology in burn and in psoriasis. What follows is a short summary of data that we published in the autumn of last year in those two diseases.

The first study that we did was with a Fab fragment derived from the original BIRR-1 antibody to test the hypothesis of whether the antibody fragment - the Fab fragment - could reproduce the properties of the full immunoglobulin in burn.

The way this work is done is to use an established, which happens to be a rabbit model of burn, in which the animal receives a brand - essentially a small burn on the skin - and it is then, after three hours, challenged with the agent under test. Typically what happens in these types of injuries is that following the burn, there is an immediate up regulation of ICAM-1 in the thelium just below the site of the burn, leucosytes rush to the area and finish up clogging the capillaries around the site of the burn and thereby reducing the blood flow.

A test for the efficacy of a compound is to measure the blood flow in the capillaries under the site of the burn. The data you see here shows how, immediately after burn, the blood flow reduces, up to the three hour point and then, on administration of the agent in red, the Fab fragment and of light blue the immunoglobulin, the blood flow is restored. That restoration of blood flow is statistically significant compared to the untreated group.

Importantly for us, what we’ve found is that the two, the light blue and the red lines, are coincident. In other words, that a Fab fragment is as good as an immunoglobulin in this setting.

The other thing you can do is, with histology, you can look at the tissue to see if there any differences between the treated and the control group. In the control group at the top, you can see, if you look at the capillaries, that they are indeed clogged with leucosytes, which have come to the site of inflammation and blocked the capillaries. Whereas, if you look in the treated group and you look at the same capillaries, they are indeed free. So you can see visually the effect that is measured on the left hand side of the slide.

So there you can look at the surface of the skin, and a typical consequence of burn on the surface of the skin is so-called epidermolysis - lysis of the epidermis, which, in the control group, is very marked here. The skin has been badly disrupted. Whereas, in the treated group, the skin, although burned, is in a much better, more stable shape. That is, again, an indicator that the treatment is having an effect in this model.

The second set of animal data that we generated and published in the autumn was in a psoriasis model. Again we used an established model together with an external collaborator, here in Frankfurt, and this is a SCID mouse model in which human psoriatic skin is transplanted onto an immune deficient mouse. Here, what's done is that the mouse bearing this human psoriatic skin is challenged with the agent, and one looks at the reduction in the thickness of the psoriatic skin.

What we did here is we took a control on the left side, we took our MOR102 substance in green, and we took a positive control known to be efficacious, namely Dexamethazone, an anti-inflammatory agent. What we found was a statistically significant reduction in the thickness of the skin, both with positive control and with MOR102.

What we also found is that the quality of that psoriatic skin, not only was significantly thinner on treatment, which you can see in this figure, but also that this inter-digitation, these fingers of tissues that characterize psoriatic skin, are much less pronounced on the treatment than they are in the control group here. This is the papillomatosis, which is substantially reduced in the treatment group.

This model is very predictive of effects in humans and what we’re currently doing is we are comparing MOR102 with known anti-psoriatic agents, Enbrel and Amevive, to see how our compound compares to those known treatments in this model.

During the course of the course of the year we have become increasingly interested in the application of antibodies against ICAM-1 in eye disease, particularly in Diabetic Retinopathy. Diabetic Retinopathy is a common condition in diabetes patients, which is an inflammatory condition of the retina, which leads, in many cases, to loss of vision. It turns out that there is an increasing body of evidence implicating ICAM-1 in this condition. That evidence is based on a number of academic studies, for example, which show that mice that lack the ICAM-1 gene in which it is has been knocked out, also lack Diabetic Retinopathy symptoms.

You can see the same effect if you treat those mice with antibodies against ICAM-1. You can shut down Diabetic Retinopathy symptoms. What seems to be going on here is that, just as in other inflammatory indications, ICAM-1 is up regulated in the blood vessels of the retina, leading to recruitment of leucosytes and eventual destruction and damage to the capillaries themselves. This is an inflammatory condition that underlies Diabetic Retinopathy.

This is an interesting indication which is becoming more and more interesting to the pharmaceutical industry, the substantial market - $1b - and has come to the fore very much recently because of first, the deal that Eyetech did with Pfizer for ocular diseases, and secondly because of Eyetech's IPO which has been the most successful of the recent IPO's on NASDAQ.

So we feel here that we have a potential new indication in which antibodies against ICAM can be exploited, and that, therefore, the entire anti-ICAM program could be much lucrative than we previously suspected.

Very recently we have provided some data on our program MOR202, including announcing the target for that program. MOR202 is a cancer program, in which we are developing antibodies directed to the target CD38. CD38 is an antigen that is over-expressed on a number of human tumor cell lines. What we have done is generated antibodies using our HuCAL technology that are highly specific for human CD38.

What we have done is we have compared our HuCAL antibodies with an existing anti-CD38 antibody, which is currently in clinical development in the UK for the treatment of multiple myeloma. What you see here on the chart is that increasing concentrations of our antibody in blue kill cells bearing CD38 much more efficiently than does the reference antibody. The key thing to look for here is the difference between these two lines. You will see that the blue line has shifted some 10 to 100-fold towards the left, relative to the red line. That means that it works at a lower concentration, so it kills cells at a lower concentration. In fact, when we measured the improvement in effectiveness, it is some 5 to 100-fold more effective than the antibody which is in clinical development.

What is very interesting about this is that it seems to work by recognizing a unique epitope, a particular part of the CD38, which has not been known before, and we have submitted patent applications on that particular feature. What this means potentially is that we can patent not only the antibodies but the way they work by targeting this particular part of the CD38 target. We are currently initiating animal experiments with these antibodies and expect to have data in the summer.

I want to turn now away from the therapeutic side of our strategy and focus in on the reagents, which is a new initiative that we have announced early on this year and which has been running only for a few months now. That comprises a new business unit that we have set up in MorphoSys called Antibodies by Design. The objective of this business unit is to exploit the HuCAL technology in non-therapeutic applications, particularly in the reagent market.

For those of you who are not familiar, antibodies are widely-used research reagents in almost every life science laboratory worldwide. The total market for these reagents is about $800m and it's growing at some 15% to 20% annually. The fascinating aspect about this market is that it is completely dominated by old technologies, in other words, by mouse or rabbit monoclonal or polyclonal technologies. There has been as yet no impact at all of new library-based technologies in this market segment. We believe that the advantage of, for example, the HuCAL technology is a massive improvement in the speed and throughput at which antibodies can be generated for life science researchers.

What this means is that we can put antibody generation on the same high throughput footing as many other processes are currently done in life science research. We have hired two very experienced executives to the business unit. The first is Dieter Lingelbach, who came to us from Roche Diagnostics, where he was Senior VP with the responsibility for global sales and marketing of biochemicals. Secondly, Joanne Crowe, who came to us from Qiagen, where she had the responsibility of Head of International Marketing.

We got off to a very encouraging start here. Over 100 orders have been placed from life science research labs all around the world as you will see on this chart, even including South Africa. We have had some very encouraging feedback from some of our customers, including this very significant quote from Harvard Medical School, which is typical of what we hear, namely 'you have accomplished in the few months, what a year of rat immunizations has not'. We are very hopeful for this new business unit, and we believe that the HuCAL technology can have a big impact on this sector, which is currently served only by outdated technologies.

We haven't talked much recently about our internal capabilities, and in order to avoid them being taken for granted, I just wanted to talk to some of the improvements we have made and the increases we have made in our capabilities during the course of 2003. We have now a very comprehensive capability, not only for antibody generation and optimization, but also for early scale production and also clinical scale production.

Our HuCAL Gold technology is now extremely well established, and we believe it's a state-of-the-art, certainly for generation of human therapeutics. One of the aspects that we are often asked about is what is the difference between your library, which is a Fab library, and competing libraries, which are based on the single chain Fv format. A couple of years ago, it was hard to identify differences between those two formats. In the meantime, a lot of data has been generated, and one very significant point is that, when you convert a single-chain Fv to a whole immunoglobulin, in 25% of cases you lose affinity to the point where the immunoglobulin is not functional.

That is not the case with the Fab fragment. So therefore the evidence that is building up points to the fact that the Fab is a superior format for an antibody library than single-chain Fv. We are therefore delighted that we chose that format, and that HuCAL Gold is based on that format.

The other thing we are seeing more and more of, which plays the advantages of HuCAL, is that the therapeutic projects that people are pursuing have more and more demanding criteria. Typically we get five different criteria per project, or more, in terms of the properties that the antibodies should have, what specificity profile it should have, what signal it should turn on or turn off, what ligend it should compete with. We are getting very, very demanding projects. That plays to the advantages of HuCAL, because HuCAL, as I have said, is set up to allow you to make antibodies to precise specifications.

In the meantime, we have used the optimization facility of HuCAL many times. Our record so far is that we have improved the affinity of an antibody 28,000-fold and the best affinity we have reached is 1 pM. In layman's terms, extremely tight binding of the antibody to its target.

During the course of the year, we banded our deal with Bayer, during the course of which we secured access to a new human production cell line, called HKB11. I want to emphasize here HKB11 is a human cell line. That means that the antibodies that are produced in it will be in a glycosylation patent, as human as it is possible to make. It turns out to be a very successful cell line as well in terms of speed and yield of antibodies that it produces, and it is being used at Bayer for pre-clinical supply of HuCAL antibodies that have resulted from our collaboration.

Another very significant point is that the FDA has given approval for a drug and clinical development, which is being produced with the HKB11 cell line - very significant for pharmaceutical partners if this cell line has already been okayed by the FDA and that is the case.

The rights that we obtained from Bayer give us unlimited use of the cell line for research purposes, and an option to use it for commercial production. So a very important addition to our arsenal of capabilities.

I want to turn now to the future, and give you an outlook of what you can expect over the next 12 months or so. First, the situation in the pharmaceutical industry, productivity continues to be the major problem facing the pharmaceutical industry today. You have probably all seen this chart before, which shows the number of approved drugs - the blue bars - versus R&D investment - the red line. The red line continues to go up, and the blue bars don't continue to get much better. In fact if you look back 10 years, the number of drugs approved in 1994 was roughly the same as were approved last year and yet last year, two and a half times as much money was spent on R&D.

This is a major headache for the pharmaceutical industry and it is what is driving their need to collaborate with biotech. As I mentioned at the outset, we see a definite improvement in the climate for collaboration. Pharma seems to be returning to external investment. We think they definitely got their fingers burnt in the late 1990s and early 2000 in technologies that did not deliver and in genomics based on pro-genomics based technologies, where pharma spent a lot of money and didn't get a lot back.

Now they seem to have got over that disappointment and seem ready to invest again, but only where they see technologies or products that clearly lead to the market. As I said at the outset, there is an increasing interest in large molecules. Therapeutic antibodies continue to be a major growth area encouragingly. The approval of four new therapeutic antibodies last year only helps our case. We are therefore bullish about the number of partnering opportunities out there.

I want to just highlight that by looking at one segment of the industry, namely the top 20 pharma companies. This chart shows you those companies, with the ticked companies being those with which we have collaborations. We currently have collaborations with seven out of the top 20 pharma companies. The interesting point here is that many of those seven already had collaborations with other antibody companies. In fact if you look at Pfizer, Pfizer already had collaborations with Abgenix, Mederex, and Cambridge Antibody Technology - three of our competitors. We were nevertheless able to do a deal with Pfizer.

The significance of that is all of the companies on this list are potential partners for us, whether they have existing collaborations with other antibody companies or not. That, for us, is a very significant point when we look at our potential market for collaborations.

In conclusion, we believe we have a very compelling business model that is built around our state-of-the-art technology. Importantly for our partnering model, that technology can be a direct source of therapeutic products. Our Antibodies by Design initiative is taking us in a new direction and giving us the opportunity to exploit HuCAL in a new market where new technology has not had an impact at all to date. We believe the model is a balanced risk-reward model, combining near term revenues from our partnerships with the long-term growth that the therapeutic antibodies that we are developing with our partners will bring when they come to market.

Looking forward to the end of the year, how could our pipeline look? It could look as shown on page 31. The first HuCAL antibodies should go into the clinic this year, which will be a very significant highlight and milestone for us. We think that the further five antibodies currently in research will enter pre-clinic. We will start at least another five new therapeutic antibody programs with either existing or new partners, and of course we will generate the pre-clinical data on MOR202. So you can expect to see a significant strengthening in the quality and depth of our pipeline during the course of the year.

Finally, before handing over to Dave, a look at our objectives and our goals for the remainder of the year. As I have said, the first HuCAL antibodies should go into the clinic during the course of the year - that's the GPC antibodies. We project 20% revenue growth this year. That revenue growth will come from new deals, from milestones and existing partnerships and also from our Antibodies by Design initiative.

We aim to partner the ICAM-1 program during the course of the year. We will generate animal data in the MOR202 program. As I mentioned, we will initiate at least five new therapeutic antibody programs, either with new partners or with existing partners. We will be - and you will see details of this when Dave gives you our financial guidance for the year - we will be EBITDA profitable in 2004. In fact, in 2004 we expect a record year. Revenues higher than we have ever achieved before, based on the strength of our business model and the opportunities in the marketplace for partnering deals.

That concludes my part of the presentation. I would like to hand you over now to Dave for an analysis of the financials.

Many thanks, Simon. In opening, I am happy to say that, financially speaking, 2003 was an especially strong year for MorphoSys. As Simon mentioned before, for the first time in our history, we achieved cash flow from operating activities in a positive way and this spread actually to an increase of our cash position at year-end. Another way to view our cash base performance is to take a look at EBITDA and using that measure we were also cash positive during the year.

There are a lot of reasons why the results are much better than last year, but probably the biggest involved the way we have brought down our cash burner. In the end, we managed to reduce expenses by 56%, which led to a bettering of the net result by more than 80%.

Let's move on to operating revenues. Revenues for the full-year 2003 decreased by 9% to €15.3m compared to €16.8m in 2002. However, if we had used constant exchange rates from 2002, our revenues would have been about €600,000 higher, giving a total of €15.9m. That comes pretty close to the guidance that we gave at the beginning of the year, of €16.0m. Other than foreign exchange developments, some deal and milestone delays also adversely affected revenues.

Looking at how our 2003 revenues are made up by segment, 82% of the revenues relate to therapeutic antibody collaborations, and 17% to target research collaborations. Less than 1% of sales arose from our new Antibody by Design initiative. On a percentage basis, the split between therapeutic antibody revenues and target research collaborations is similar to what it was last year. The percentage of revenues coming from Antibodies by Design, which is the newly created arm of MorphoSys designed to exploit non-therapeutic applications of HuCAL, was relatively small. Remember, the unit was created in 2003 and, looking ahead, we expect more from that unit.

Another way to view our revenues is to see what activities generated those revenues. The biggest part of revenues in 2003 arose from annual licensing fees from our existing partners. Milestones revenues amounted to 0.5m for the year or 3% of total revenues. That's less than 10% than they contributed in 2002. Grant revenues amounted to about €100,000 in 2003, and was basically unchanged from the prior year.

If we take a look at where sales geographically rose, 81% of MorphoSys's revenues came from the United States, and 19% from Europe. That’s similar to last year's result of 76% and 24% respectively.

Moving on to the operating expenses for the year 2003, total operating expenses including R&D expense, SG&A expense and staff-based compensation expense decreased by 56% to €18.8m. That's a reduction of about 23.5m and the result is better than expected.

Costs for research and development fell by €10.6m compared to the prior year. The decrease resulted chiefly from lower licensing costs as a result of the licensing and settlement agreements with CAT and XOMA. Also impacting costs was the company's decision to refocus efforts in proprietary product development. As you know, the company now out-licenses the compounds at the pre-clinical stage instead of developing them in the clinic and hence costs are accordingly lower. Both external lab funding and personnel costs were lower as a result of this change in focus.

Sales, General and Administrative expenses amounted to €7.6m, which compares to €18.7m in 2002. The almost 60% decrease in SG&A expenses was due to the fact that there was no cap litigation in 2003. Also significantly contributing to savings was the closure of MorphoSys USA Inc. Closing our US sub resulted in cost savings for MorphoSys of about €1.8m per year.

Stock based compensation in the amount of €2.2m for the year was recorded as a non-cash charge. That amount is less than it was last year; the main reasons for the decline include fewer employees, more forfeitures, the exploration of amortization periods and a lower MorphoSys stock price.

On that note, I want to mention that MorphoSys is one of the very few companies that actually expensed their stock options under FAS123 in US GAAP and we have been doing that since the year 2000.

Let's take a look at the expenses by cost type. Personnel costs amounted to €7.5m or about 40% of total costs. As such, they are the largest cost block of costs that we have. Contrast that to last year, where intangible costs was the largest cost block. Intangibles cost includes patent litigation cost and patent and licensing amortization, and those amounted to 36% of our total costs in the year 2002. Fortunately for the agreements with CAT and XOMA, we have been able to significantly reduce those expenditures. You can see that actually in the percentage of total expenses that we have in 2003, where it only contributes 5% of our total expenses.

External services, which includes external lab funding and various outsourced administrative services, comprised about 20% in 2003, and was essentially unchanged from the prior year. The same goes for infrastructure costs, which were about 12% of total costs and remained largely unchanged.

Let's take a look at the savings that we had in 2003 compared to 2002. As you can see from the financials, a lot of the financial improvement came from savings in the operating area. Total operating expenses decreased by 56% - almost €21m. That chart there breaks down how those cost savings arose. As you can see, in order, the biggest cost savings arose out of intangibles costs, external services and personnel costs. As I mentioned in the previous slide, intangible costs were sharply lower in 2003, and contributed two-thirds of total savings. Costs for external services, which were primarily lower levels of cost related to external lab funding and M&A related expenses, contributed to about 20% of the total.

Finally, during the year MorphoSys personnel costs also fell from the leaner staff structures which we have when we implemented the company's restructuring plan. Personnel costs fell by 2.6m, of which MorphoSys US contributed €1.2m.

As has been mentioned earlier in this presentation, EBITDA is one way to take a look at operating cash flow. In our case, EBITDA, as defined by earnings before interest, tax, depreciation and amortization, excluding stock-based compensation, was positive in the amount of €1.2m for the year. As you can see on that slide this compares to a negative €18.8m number last year, reaching an EBITDA positive number of course for us, a MorphoSys first, and we are very happy to be able to announce that today.

Bearing the positive development of EBITDA at year-end 2003, the company had approximately €23m in cash at the end of the year. That compares to €19.1m at the end of 2002. That's an increase of more than 20%. That's another MorphoSys first - the first time we have ever increased our cash from operating activities as opposed to from financing activities.

Underscoring that point, you can also take a look at the cash provided by operating activities in the cash flow statement, which amounted to €5.8m positive, which strongly contrasts to the cash used in operations of €15.2m in 2002.

Let's move over to the balance sheet. During the year 2003 the company's current assets decreased by €3.3m to €26.2m, primarily as a result of lower receivables costs at year-end 2003. Total assets increased by 3.4m to 45.8m in 2003, the difference is mainly due to an increase in intangible assets, which we had one on the back of the acquisition and related capitalization of the CAT settlement.

On the liabilities side, during 2003 total current liabilities fell by 0.8m principally due to the settlement of licenses payable of €4.9m which was in large part settled to the equity issuances to CAT and XOMA. Deferred revenue increased by 2.3m to 10.4m, largely as a result of the collaborations entered into in the last quarter of 2003.

In 2003, MorphoSys issued equity twice. The first equity issuance was part of the MorphoSys XOMA licensing agreement signed in 2002, under which MorphoSys elected to issue about 363,000 shares to XOMA. The shares were issued to XOMA in the first-half of the year, and in coordination with MorphoSys, XOMA successfully cleared all of their position by Q3 of 2003.

The second issuance related to our settlement with CAT, and as part of the agreement MorphoSys issued 588,000 shares to CAT. Those shares were issued in August of 2003, and CAT still retains 100% of those shares. That being said, some of those shares are issued subject to a lock-up.

On that note, if you take a look at our shareholding, our biggest shareholder is currently Cambridge Antibody Technology, who owns about 12% of our shares, following by Schering, who owns 7%. The rest, the freefloat, which, according to the definition defined by Deutsche Borse, is about 80%, which includes about 3.5% of shares owned by the company's management and supervisory boards.

Let's take a look at the number of employees. At the end of the year 2003, MorphoSys employed 95 people, compared to 110 at the year-end 2002. Of the 95 employees at the end of 2003, 71 worked in research and 24 in SG&A, so roughly our historical average.

Before I start with guidance, I want to make a brief mention of international accounting standards. In fact, we intend to switch over our accounting principles over to IFRS from US GAAP sometime during the course of this year. As I am sure you are all aware, all European publicly traded companies need to switch their accounting over to international accounting standards by 2005. MorphoSys has been doing its accounts according to US GAAP since 1998. However, the time of change is upon us and sometime this year we intend to convert our accounts over to IFRS.

When we switch, of course we will be in a position to give you full reconciliations to US GAAP and explanations of the main reconciling items. A little later this year, we will give guidance exactly which reporting event we intend to switch over our accounts. That being said, for comparative purposes, we will continue throughout this year to give guidance and reconciliations to US GAAP.

Let's move on to guidance. I'd like to conclude the financial guidance, which I said is based on US GAAP for 2004. As I previously stated, and what you have heard from Simon today, is we expect double-digit revenue growth in 2004. We estimate year-end 2004 revenues at €19m. We estimate total operating expenses at roughly €23m. That amount includes €1m worth of stock-based compensation expense. Although this financial picture gives a financial loss for the year, we actually expect an EBITDA positive result for 2004. In terms of cash, we expect the cash position at the end of the year of at least €20m.

In closing, I would like to say we had, I think, a great year financially. I think what we have seen is that we have done a lot to reduce our burn rate in 2003 and as you can see from the projections we are back on track to achieving double-digit revenue growth.

Looking ahead, we expect strong revenue growth. We continue to exploit new initiatives such as the Antibodies by Design initiative and as well we are investing in our future by investing in proprietary products. All of these things I might add against the background of financial stability.

That concludes the financial analysis and the guidance for 2004 and I would now like to hand back to Claudia.

Thank you very much Dave. Thank you very much for your attention. I would like to open the floor now for questions. I would suggest that we start with questions coming from the conference call. May I ask the operator whether there are questions from the conference call?

Thank you. The Question and Answer Session will be conducted electronically. If you would like to ask a question, please do so by pressing the star key followed by the digit one on your touchtone telephone. If you are using a phone with a mute function, please make sure your mute function is turned off to allow your signal to reach our equipment. We will proceed in the order that you signal us and we will take as many questions as time permits. Once again, please press star one on your touchtone telephone to ask a question. If you find that your question has been answered, you may remove yourself by pressing star, two. We will pause for just one moment to give everyone an opportunity to signal. Our first question comes from Nick Turner with Jefferies International Limited. Please go ahead.

Hello Dave. You talk about revenues next year growing at a double-digit rate, €19m. How much of that is already on the books? I am making the assumption that revenues from Pfizer don't appear in your 2003 numbers and therefore maybe a significant portion of the projected growth for 2004 is already on your books and therefore implicit in your guidance. How many new deals do you think you might need to strike to meet that target?

Secondly, from the point of view of antibodies in development, I mean you talked and showed some very nice data on your collaboration with Roche. However all of the milestones in that project have been met, certainly 12 months ago I guess now. Could you give me some indication if you can as to what is ongoing at Roche and what is perhaps holding Roche back from committing to clinical development of that particular antibody?

I'll take the first-half of that question. Nick, you are correct. The revenues that we saw in 2003 related to Pfizer were very small. We expect to see that revenue over the next couple of years being amortized over the lifetime of the agreement, which is five years. You asked the question, how much revenue do we have on the books that is currently secure. We estimate that amount to be slightly more than half, so slightly more than 50% today of those projections are already secured through committed contracts and committed revenues.

I'll take the second part of that question with respect to Roche. Unfortunately we are not at liberty to talk about Roche's plans for that compound, which we would like to be able to do, but subject to our agreement with Roche, we simply can't comment on the next step and subsequent developments.

Thanks.

Is there another question?

Thank you. Once again, if you would like to ask a question please press star one now. It appears there are no further questions from the audio conference at this time. I would therefore like to turn the call back.

Okay. Well I think we'll start now with questions from the audience. Are there any questions here?

Thomas Hoger, DZ Bank. Two questions concerning your pre-clinical programs. First of all when do you expect the comparable data with Amevive compared to your antibodies?

Secondly, you mentioned with the burn model that there is an increased flow of capillary blood. Could this be a surrogate market for clinical development?

To the first part of your question, when would we expect comparative data, that day should come in the summer. Again, just as a reminder that is a comparison of MOR102 versus Amevive versus Enbrel side by side in the SCID mouse model that I described. So mid-year we expect that data.

The second part of the question, whether the in vitro burn is a surrogate market for clinical development, we believe so, we believe so. Of course that doesn't obviate having to do the regular clinical development with that compound, but we chose to do that study because it happens to be a good and predictive model of what you can expect in vivo in humans.

Roger Baker, Live Cycon(ph) I have a question to your new business unit, Antibodies by Design. Is it correct to assume, you mention - if it is correct to assume that your major customer will be academic institutions, then the question arises whether the prices for new designed antibodies can compete with those produced by traditional methods? Can you comment on this?

Yes I can. I can tell you that of the over 100 orders that we have had so far, it represents a real mixture of users. So not only academic institutions as you would expect, but also companies of all sorts and shapes and sizes - pharma's, biotech, research companies and so on and so forth. In terms of the pricing, for competitive reasons, we don't talk in too much detail about the pricing. Suffice it to say that we feel comfortable about our ability to compete and to offer a product that is a very competitive product versus conventional custom monoclonal antibodies for example.

Remember that what we are offering customers here is a huge advantage in terms of speed. Typically, if you order an antibody from a custom-producer, you wait six, maybe nine, months and if you are unlucky you don't get anything. In our case what we are talking about is a turnaround in something like eight to ten weeks, and that for many researchers is a major advantage, which allows us to command a price that we feel is very competitive.

Thank you. If I may I have a second question. Perhaps I missed it, I was a little bit late this morning. Concerning your antibody MOR202 which is targeted against CD38, you mentioned and that is new to me, that you not only are able to patent or have already patented this antibody, but also you patented or you are planning to patent the epitope or the interaction between the antibody and the epitope. Did I get this correctly?

Yes you did. What is surprising and novel here and therefore inventive is the fact that this antibody exerts its effect by binding to a particular part of the CD38 antigen. Because that is a novel and new discovery that is useful potentially, then that’s something that you can protect. So the mechanism of action of this antibody through binding to that particular epitope.

[inaudible] in your balance sheet we see license rights of some €10m. They increased by a factor of 2 from last year. Could you describe what is behind that?

Sure. I think you are talking about the license fees which went to 10.8m from 2.3m. The difference is the CAT license. We capitalized license when we raised the shares in August and that difference is essentially that. When we did the settlement with CAT, there were basically several components to the contract. One component was paying for past use of certain of their technologies and another component of the contract was to take a license from CAT as it relates to certain activities that we do. That's why we raised the shares during the year, as we were paying for that license by raising the shares. We have a license from CAT to perform certain activities with certain of our technologies. That difference in licensing fee that you see there is exactly that.

Maybe this is a good point to just add a supplementary point to that. As Dave mentioned, there are various components to that settlement. One component is a license agreement for a certain part of our technologies. The most important component is the so-called covenant not to sue, and that covers our HuCAL Gold activities. What that means is we had complete freedom to apply HuCAL Gold to any target, even if that target has already been reserved by CAT or licensed to some partner by CAT. That's a key, key distinction between the license and the covenant not to sue.

The other consequence of that is that we don't pay any royalties on HuCAL Gold products to CAT, again because the activity is covered by the covenant and not by the license.

Coming back to the new business unit, do you plan to report on the new business unit separately? The second question is what is the estimate of the save of this new business unit to the guidance you gave for 2004?

Okay. In response to your first question, that will depend on how quickly we move over to IFRS. I believe under IFRS the guidelines for what you define as a segment are much more strict than they are under US GAAP and certainly IFRS we will immediately report them as a segment. That being said, by Q1 we will certainly report at least their sales separately, most likely their costs.

In answer to your question what do they compose of the total revenue projections we have made of the year? We estimate to be slightly under 10%

LBW. In addition to the former question on this new section on Antibodies by Design, what is your plan to market and to bring this new field to the academics?

We see a number of segments, sub-segments of this research market and the way we consider them is there is a catalogue segment - these are companies that have a lot of antibodies that they sell via catalogue. There is a custom segment, where people want antibodies generated de novo, and we also think about the array segment, where people are increasingly interested in making antibodies on arrays for various research and diagnostic purposes.

At the moment the bulk of our activities are custom activities - academics, companies, you name it, researchers - who contact us, ship their antigen to us and we make the antibody against their antigen. The way we target that market is through conventional advertising in trade journals and trade magazines. We are going to conferences and trade fairs, through email shots to life science researchers. With time we will build a pool of antibodies ourselves which we will then be able to sell either ourselves, through a catalogue-type approach, or in collaboration with an existing catalogue supplier. So the custom segment is the one we are tackling first, then that will lead naturally into a catalogue opportunity and the ship opportunity we see as being interesting but really too early an stage at the moment.

Maybe one follow-up? On slide 10, you found that GPC, your collaboration with GPC, is not yet in the pre-clinics, are there any other collaborations ongoing except the one which will be in clinic in summer?

Sorry I missed that, the --?

Just in slide 10 that you have collaboration with GPC, which is in the research stage?

Sorry, you shouldn't interpret the arrows as belonging specifically to the companies on the left hand side. There's no attempt to line up the companies with the arrows there. Unfortunately, in many cases, we can't disclose how many programs are actually ongoing with particular partners. In the case of GPC, I can tell you that there is one active program, which is the antibody, which we expect to go in the clinic this year.

Rachel Graham[ph] from Dow Jones. I have a couple of questions. On the new indication for ICAM-1 in eye disease, are there any other drugs on the market for that particular - for Diabetic Retinopathy, to save me researching that, if you can tell me that? With regard to partnering, the psoriasis projects and the burn projects, is that going to be the one partner or is it possible that that will be two partners, and any other information on that?

With respect to the first question, what's interesting here is the recognition that Diabetic Retinopathy is indeed an inflammatory disorder is a relatively new discovery or recognition if you like. There is actually old data from the 60s which show that diabetes patients with rheumatoid arthritis who are heavily, heavily treated with Aspirin, actually showed a very low level of Diabetic Retinopathy, which now, in retrospect, seems to underscore this point that it is an inflammatory disorder.

There are tests ongoing with other anti-inflammatories to see whether they also work, but this is essentially taking existing anti-inflammatory drugs and testing them in Diabetic Retinopathy. It is essentially a recognition of this new, if you like, discovery that Diabetic Retinopathy is indeed inflammation.

The second part of your question with regard to partnering of MOR101 and MOR102, potentially we see the opportunity to do three different deals here: a deal in the burn field around the Fab fragments; a deal in chronic inflammation which could be psoriasis, it could be rheumatoid, it could be Crohn’s, it could be something else; and the third one would be a deal in eye disease because that’s compartment, if you like, that you can nicely separate from the other applications.

It's almost impossible to imagine doing different deals in psoriasis and rheumatoid for example because of the close relationship between those diseases, but we could imagine doing three different deals. What it will actually transpire and look like in the end, we can't say yet, but it depends really on the requirements and the demands of the partner that we finally settle on. We certainly see that the eye disease is something that could be carved out into a separate deal compared to the other components.

[inaudible] combination question really. Simon in your part you mentioned that the pipeline, where you are expecting the pipeline to be at the end of 2004. On the slide there isn't any introduction of new programs from MorphoSys, although a year ago you mentioned that you were going to introduce about two and a half programs a year. In connection with that, your expenses are estimated to increase in 2004 by about 5m. Can you just give me an idea where this is going to come from? Where do you expect this hike in expenses to be coming from?

Yes, let me speak a little bit to the point about number of own programs. Of course we base our internal investment decisions on own programs on the quality of the data we generate and the potential for the programs that we are pursuing. It may turn out that, if we are particularly excited about a particular program, that we may choose to invest more in that at the expense of introducing a new and therefore early stage program. In the case of MOR202, we are very excited about the data there, but we want to be able to invest more in improving those antibodies and are looking at various ways of doing that.

Secondly, this eye disease opportunity in ICAM has now come up as being something relatively new. Therefore, of course we make a decision. Does it make more sense to invest perhaps more money in ICAM versus starting a completely new program? The information we gave regarding the number of new programs we would start each year, you should take very much as a rough guide and not as a definitive commitment to start that many programs per year. It depends on the quality of the data and our desire to invest in more promising and perhaps more advanced programs, rather than starting completely new programs.

[inaudible]

No, frankly. Targets we see coming from either in licensing deals like the ICAM-1 or public domain, CD38 is a public domain target where we believed, and we were proved right, that we could generate a better antibody that could have interesting effects, and maybe even the opportunity to create some intellectual property as we are doing with this new epitope. So we don't really see that access to targets as limiting. What we are looking to do is we are looking to channel our investment into those programs where it makes the most sense rather than to religiously commit to initiating a certain amount of new programs per year.

Martin Possienke, Equinet. I have got three questions. The first one regarding the Antibodies by Design and I guess it's a follow-up again. I just wonder, as it seems to be such a logical step to enter this market, why you have done it that late or why other library companies did not enter this market yet. Secondly in this context as well, if I got it correctly, you guide €2m for 2004? An EBIT margin guidance would be quite helpful in this context or maybe you can say if it is a profitable business unit even in 2004? Thirdly, a more strategic question. Maybe you can share with us your long-term view of the company or MorphoSys's future, where do you expect MorphoSys to be in five years time? Fourthly, a financial question, after the change to IFRS do you still have to expense for stock-based compensation?

I'll take parts one and three of that and Dave you'll take parts two and four. Part one, why did we only just start with Antibodies by Design now? Frankly the intellectual property cloud, if you like, was an issue that we didn't want to have to explain to several thousand potential academic or other research customers. Now that that cloud has cleared completely, it makes it much easier and much more direct to do business which involves selling on a weekly basis.

Why hasn't anyone else done it? Frankly we don't know, and I think one reason could be that people who have new technologies don't see this as being the most lucrative space to work in, and therefore have chosen, like Cambridge Antibody Technology for example, have chosen to focus on therapeutic antibodies because there is a perception that that is the most lucrative space. So that's our explanation of why no one has chosen to do it. As we have said, we see this as a golden opportunity to have our technology have a major impact on a new market that is currently unserved.

Part four was where do we expect to see the company in five year's time? As I said during the presentation, we think we have a very compelling business model. It is, we believe, the ideal mixture between partnering activities which are revenue-generating activities and own activities, where we will retain a greater share of the value of those programs. When you look at us, you should look at the breadth and the depth of our product pipeline, and if we are right in saying that our technology will result in a higher probability that therapeutic antibodies will come to market, then that very broad pipeline we have established will result in multiple products coming to market. That's the real true long-term value in MorphoSys - our participation in those multiple product programs.

Frankly where we are going, is we are aiming at becoming the preeminent source of therapeutic and research antibodies for consumers of those, which means pharmaceutical companies that are developing therapeutic drugs and researchers who are buying those for research purposes. We think we can do that based on the strength of the technology.

Are you thinking about switching your business model again to clinical development, for example?

No, we have no plans to switch our business model. Of course should we want to, that's always an option. We can always decide should we want to at some stage in the future, that we have got such a promising compound, it looks so good pre-clinic, that we may decide to go further with it, but we have no plans to do it at the moment. The business strategy remains as communicated over the last couple of years now, which is pre-clinical development of compounds, out-licensing of those compounds, and partnering based on the existing model. Sorry there were parts one and three to your question which Dave will answer.

Regarding the break-even of the Antibodies by Design unit, the unit will be roughly break-even. It could be slightly positive, slightly negative by small amounts, but it should be roughly break-even. Regarding the question on stock-based compensation, one of the reasons we switched to stock-based compensation back in 2000 was we believed that both US GAAP and IFRS were headed in that direction. I still believe that is going to be the case and most likely we will continue to expense stock options under IFRS.

Thomas Hoger, DZ Bank again.Two questions - first of all what is the current status with regarding the patent litigation with Applied Molecular Evolution? Secondly, during the tenth anniversary of MorphoSys, your collaborator Bayer AG was rather bullish about the clinical development, or the start of clinical development, with one of the antibodies Morphosys produced with Bayer. Ever since, I have heard nothing about it. Could you give us an update?

First, with respect to AME, we have no new information. As a reminder, one year ago actually in January of last year, we got a ruling by the Magistrate in our favor, in summary judgment, that we did not infringe the AME estate, and the next step was supposed to be decision by the judge. We haven't heard. Unfortunately we have no means of knowing when the judge will rule. We simply have to wait and he will rule at his leisure. Of course what has happened is in the meantime AME has been acquired by Lilly, and that acquisition has been improved recently by the shareholders. We are none the wiser as to the plans of Lilly with respect to that process and that litigation.

The second question, Bayer and their plans for clinical development. Whatever statement Bayer makes is up to Bayer, and we are not at liberty from our side of this collaboration to comment on their plans. I'm afraid you will have to ask them with respect to those programs.

[Inaudible] Bank. I have got a question on your antibody MOR202 targeting CD38 cells. You mentioned that there is a clinical trial going on in the UK on multiple myeloma. Could you mention which company is carrying on these trials and what status they are in?

It is actually the Cancer Research Campaign. It's an academic driven trial sponsored by the Cancer Research Campaign that, to the best of our knowledge, is currently in Phase I for multiple myeloma, and more information than that, unfortunately we don't have.

Are there any further questions?

You mentioned the lock-up period for CAT shares. Can you tell us when that ends?

I'd love to, but due to the confidentiality between us and CAT, I think it was agreed not to disclose that.

Thank you.

Perhaps we go shortly back to the conference call, if there are any more questions from the conference call?

Thank you. As a reminder, if you would like to ask a question, please press star one now. We now have a follow-up question from Nick Turner.

Just a question for Dave. R&D expenses in Q4 were only €500,000 compared to 2.7m in the preceding three quarters. I wonder if you could talk me through that? Did I understand from you that in fact you are delaying booking R&D costs, I guess you are capitalizing R&D, are you delaying booking the costs until projects move to the clinic? Did I hear that correctly?

Let me handle the first part of the question. The reason our R&D expenses are so low in Q4 is, if you take a look at the cash flow statement, you will see that we had a change in accounting estimates of 2.2m which was a reversal of R&D expense in that amount, and R&D expenses without that would have been 2.2m higher, plus the 500,000 you mentioned. The second part of your question, I'm sorry, I didn't quite understand it. Could you repeat it?

I just thought that I'd heard you mention in the presentation, perhaps I misheard, that perhaps you were capitalizing R&D expenses and that therefore R&D expenses would be booked through the P&L when projects move into clinical studies rather than pre-clinical, or did I completely mishear that?

Probably the latter. We don't capitalize any R&D expenses. We capitalize only the licenses that we acquire.

Okay, is there another question from the conference call?

It appears there are no further questions from the audio conference at this time.

Okay. Are there any more questions here? Did I miss somebody?

Christian Orquera, Bankgesellschaft Berlin. I have a question regarding your proprietary pipeline. How far can you influence in such kind of deals when such a [indiscernible] would go into the clinic? Is that an important criteria? The next question is also regarding your pipeline. You are retaining a co-development opt-in. Does it mean that at some point in the future you may, if trials come out good, you may jump in co-development in Phase II or Phase III [indiscernible] that way? Maybe one last question. We can also see, regarding targets for 2004, we see five development programs. Does it mean that you are expecting the earning activity to increase or you will put less focus in licensing targets for your proprietary pipeline? Thank you.

To question one, how much can we influence the partners program as to when it goes into the clinic, for example? The answer is not at all. The decisions about the development after we have delivered the antibody to the partner is purely up to the partner. We have no influence over their decision to take it forward or not to take it forward. All of our agreements have diligence language in them, which commits the partner to develop the compound diligently, and failure to do that, in most cases, would result in us having the right to get the compound back. As to their decision whether to take it forward or not, that's up to them and we have no control over that.

The second question was whether, under this opt-in, we have the possibly to join development programs. That is the intention, the intention is when we out-license our own programs to build into the agreement such an option clause under which we could look at the data at a certain stage, say at the end of Phase II, and then if we exercise the option we could join back in a co-development situation where we would co-finance and then co-own whatever comes out at the end. This is a relatively common structure that is used in the industry.

The third question related to new program starts, that spoke partnered programs, and what that meant was, just so that we are clear, that amongst our existing or new partners, at least five new programs would be initiated. It doesn't speak to the number of new deals that we anticipate and we indeed give no guidance on that because it's very difficult to predict. We simply say that with a new or existing deal, at least five new programs will start.

Are there any further questions? I will ask for the last time back in the conference call, on the audio conference whether there is another question.

It appears there are no questions at this time.

Okay thank you. I think we are at the end. I would like to thank all of you for coming to our conference and I would like to invite you to a little lunch outside. There will be the possibility for some more discussions with us. We are all over there. Thanks for coming and thank you.