Exelixis Inc (EXEL) 2003 Q4 法說會逐字稿

At this time I would like to welcome everyone to the Exelixis fourth quarter and year end 2003 financial results conference call and outlook. (OPERATOR INSTRUCTIONS) I would now like to turn the call over to Ms. Green. Ma'am, you may begin.

Good afternoon. And thanks for joining the Exelixis management team on our fourth quarter and year end 2003 financial results conference call. Present during this call are George Scangos, President and Chief Executive Officer; Frank Karbe, Senior Vice President and Chief Financial Officer; and Kristine Ball, President of Finance.

Following this introduction George will provide prospective on this quarter and the year. Frank will review the Company's financial performance for the quarter and the year and provide guidance for the first quarter and full year 2004. And then George will provide additional comments and we will then open the call for questions.

Please note that the following discussion contains certain statements that are forward-looking, including without limitation answers to questions at the end of the formal remarks. These statements are only predictions and they are based upon our current expectations. Forward-looking statements involve risks and uncertainties. And actual results from the timing of events could differ materially from those anticipated in our forward-looking statements as a result of (indiscernible) and uncertainties which include without limitation risks related to our ability to enter into new collaborations, to continue existing collaborations, and receive milestones and royalties through licensed future products developed from the search effort under collaborative agreements, the rate of growth depending on licensing and contract revenue, the timing (indiscernible) expenses associated with the growth of proprietary programs, and the GSK collaboration, Exelixis' ability to enter into new collaborations, continue its system collaborations, and receive milestones in royalties through future products developed from research efforts under collaborative agreements, the rate of growth if any in licensing contract revenues, the timing and level of expenses associated with the growth of proprietary programs, and the GlaxoSmithKline collaboration, potential failure of clinical testing of Exelixis' product candidates to demonstrate safety and efficacy, the ability of Exelixis to file IND applications at the (indiscernible) time, the ability of Exelixis to conduct a Phase 3 clinical trial of XL119 sufficient to achieve FDA approval or to initiate the planned Phase 3 clinical trial in the second quarter of 2004, the ability of Exelixis to successfully advance and develop additional (indiscernible) compounds, including XL647, XL999, XL844 and others. These and other risk factors and discussions of risk factors on our quarterly report on Form 10-K for the quarter ended September 30, 2003, and other SEC reports.

We especially disclaim any obligation or undertaking to release publicly any updates or revisions to any forward-looking statements contained herein, or to expect any change in our expectations with regards to, or any change in events, conditions or circumstances in which (indiscernible). Now I will turn the call over to George.

Well, I think Exelixis has had a great year in 2003. We achieved our clinical. We achieved our strategical, reached a high level of fiscal responsibility and operational efficiency. We generated a rich and a growing development pipeline. XL119 is about to enter a Phase 3 clinical trial with the potential treatment for (indiscernible) cancer.

We have multiple compounds, including XL647, 999 and 844 advancing through the pipeline heading toward IND filing. Then we have exciting compounds in earlier stages of development that have the potential to become IND applications in 2005 and beyond. Our current partners are pleased with our performance. And based on our assessment of the partnering landscape this year, we believe that we will have additional partnering opportunities that can further leverage our capabilities and strengthen our financial standing.

Today Exelixis is a dramatically different and more advanced Company from what we were just a year ago. We are product focused and market-driven. And we have been relentless in our commitment to achieve our goals. We're proud of what we have accomplished and optimistic about our prospects for 2004.

I will expand on these points in just a few minutes. But first I would like to officially welcome Frank Karbe to the Exelixis team as Senior Vice President and Chief Financial Officer, and ask him to review the financial results for the fourth quarter and the full year of 2003. I would like to just say that Frank has been here for two weeks. He is already making a great contribution to the Company, and we're really delighted to have him.

It's a great pleasure to be here and to participate in my first financial results conference call at Exelixis. Since I have not had a chance to meet personally with most of you on the call, let me make a few opening comments.

I'm very excited to be part of the Epstein team. Going forward I will focus on maintaining strong financial discipline. And I consider it part of by mission to help Exelixis develop to its full potential.

Let's turn to the numbers. I will start with an overview of our financial performance for the fourth quarter and the full year of 2003, and then provide guidance for the first quarter and the full year 2004. Exelixis' financial performance in the fourth quarter and for the full year '03 was strong. We were able to exceed our cash and expense goals and continue to invest aggressively in advancing our development programs. We also exceeded the quarterly guidance we provided at the end of the third quarter for achieving revenue growth of over 10 percent, and keeping our operating expenses flat as compared to the third quarter of '03.

For the full year '03 as compared to '02 our revenue increased 16.3 percent while our operating expenses, excluding non-cash and restructuring expenses, increased only 13.2 percent. In addition, we ended 2003 with approximately 242 million in the bank, which is 20 million more than we had at the end of '02.

Let me now provide additional detail on our performance. As in the past, we're reporting results both on a GAAP and a non-GAAP basis. Non-GAAP results exclude restructuring expenses, gain from insurance settlement, loss from discontinued operations, and non-cash charges relating to stock compensation expense and amortization of intangibles.

Let's first talk -- let's first look at net loss. And I will start with the quarterly results, and then talk about the full year numbers. For the quarter ended December 31, '03 we reported a debt net loss of approximately 23.3 million or 33 cents per share, compared to a debt net loss of 20.9 million or 36 cents per share for the quarter ended December 31, '02.

On a non-GAAP basis, we reported a net loss of approximately 22.6 million or 32 cents per share for the fourth quarter '03. And the comparable non-GAAP net loss for the fourth quarter '02 was approximately 19.8 million or 34 cents per share.

For the full year ended December 31, '03 the Company reported a GAAP net loss of approximately 94.8 million or $1.45 per share, compared to a GAAP net loss of 86.1 million or $1.52 per share for the year ended December 31, '02. On a non-GAAP basis we reported a net loss of approximately 93 million or $1.42 per share for the full year '03. And the comparable non-GAAP net loss for the full year '02 was approximately 81 million or $1.43 per share.

Let me now turn to cash. Our cash burn for 2003 was 86.4 million. As in the past, we define cash burn as the sum of the net cash used in operating activities, flat (ph) purchases of property and equipment, net of proceeds from bank obligations, flat principal payments from capital lease obligations, notes payable and bank obligations as derived from our consolidated statement of cash flows prepared in accordance with GAAP.

As I mentioned earlier at December 31, '03 cash, cash equivalents, short-term investments and restricted cash totaled 241.9 million compared to 222 million at the end of '02.

Let's now look at revenues. For the quarter and year ended December 31, '03 total revenues were approximately 13.8 million and 51.5 million respectively, compared to 12.5 million and 44.3 million respectively for the same periods of '02. The increases in '03 compared to '02 were driven primarily by revenue from our corporate collaboration with GlaxoSmithKline and from compounded deliveries under our chemistry collaborations with Cytokinetics, Elan, Merck, Scios and Schering-Plough, partially offset by the May '03 conclusion of our collaboration with Protein Design Labs.

Next I will cover R&D costs. Research and development expenses for the quarter and year ended December 31, '03 was 32.6 million and 127.6 million respectively, compared to 27.7 million and 112 million respectively for the equivalent period of '02. The increase in '03 from the '02 levels were driven primarily by the expansion of our drug discovery and development operations and activities associated with advancing our clinical and preclinical development programs.

Let's turn to G&A costs. General and administrative expenses for the quarter and for the full year '03 were 4.2 million and 18.6 million respectively, compared to 4.8 million and 18.8 million respectively for the comparable period in '02.

Let me finally share with you our outlook for '04. And please keep in mind that as I talk about expectations for '04 we have not included the potential impact of any product end licensing, equity offerings or business combinations that may be closed or entered into during the course of the year.

We anticipate that revenues as compared to '03 will increase in the range of 15 to 25 percent. And that operating expenses, excluding non-cash and restructuring charges, will increase in the range of 23 to 28 percent. The increase in operating expenses is primarily related to the continued expansion of our development and drug discovery operations and to advance and expand our clinical and preclinical research programs.

We expect our cash, cash equivalents, investments and restricted cash balance to exceed 175 million at the end of '04, including estimated proceeds of 30 million in 2004 from our loan facility with GSK.

Let me also announce a little adjustment to our current practice of reporting cash burn. Beginning with today's press release and our communications in this conference call, we plan to report our quarterly and end of year cash balances, and will discontinue the practice of reporting cash burn. Cash burn, as you know, is a non-GAAP calculation. And we believe that cash usage based on our beginning and our end of year cash balances is an equally meaningful metric.

With regard to our financial expectations for the first quarter of 2004, we expect our revenues to decline in the range of 15 to 20 percent from the fourth quarter '03 levels due to anticipated timing of compound shipments under our covenantory chemistry collaborations. Due to the nature of these collaborations, compound shipments and gross revenue recognition generally do not occur evenly quarter to quarter. Our compound shipments in the fourth quarter '03 were quite strong. And we anticipate a lower number of shipments in the first quarter of '04.

We expect our operating expenses, excluding non-cash and restructuring charges, to increase by 10 to 15 percent over the fourth quarter '03, primarily due to the continued expansion of our development and drug discovery operations.

Now I would like to turn the call back over to George.

Let me ask -- first update everyone on our progress for 2003. And then I will provide some comments on our prospects for this year and the future. And let's focus first on the development pipeline starting as usual with XL119, which is our most advanced compound. I'm happy to report that we're on track to initiate the Phase 3 registration trial for XL119 in the second quarter of this year. We have selected our CRO to manage the trial. And we should be able to announce that selection shortly.

The processes of identifying and setting up centers, getting R&D approval and other activities necessary to conduct the trial are underway. We have applied for orphan drug status. And we continue to work with the NCI to plan additional Phase 2 exploratory studies designed to study 119 in upper GI tumors and leukemias in a combination with other same cytotoxifacations.

XL119 is only one of what we believe is a rich and growing pipeline of valuable assets. All of our other compounds in development derive from our own internal R&D efforts. And we're excited about their therapeutic and commercial potential. The most advanced of those is XL784. And it is a compound that originates from our anti-androgen Genesis (ph) research program. It is a potent inhibitor of a cell circuit comp (ph) protease called ADAM-10. This molecule is a metallic protease that cleves (ph) growth factors involved in cell growth, and has been shown to play an important role in blood vessel formation and self abration (ph).

As you know, a number of compounds designed to inhibit matrix metallic proteins have entered the clinic in the past and had been limited by their lack of efficacy and/or toxicity. And it is thought that the toxicity of these compounds is mediated at least in part through their inhibition of the target matrix metallic protease MMP1.

XL784 was specifically optimized to be MMP1 sparing, thus potentially significantly enhancing the safety profile and enabling higher dosing in comparison to the MMP inhibitors. In preclinical studies XL784 dosed orally demonstrated excellent pharmacokinetics properties and was remarkably free of toxic side effects. Additionally, the Phase 1 of orally administered XL784 in healthy human volunteers showed that single doses of the compound were free of side effects and to have an attractive pharmacokinetic profile in humans as well.

Preclinically, XL784 demonstrated some (indiscernible) inhibition, the growth of human tumors xenografts derived from a variety of carcinoma cell lines. And additionally the compound shows good activities in rat models of renal and cardiac failure. We decided to pursue development of XL784 as a potential treatment for renal and cardiovascular disease and may pursue its utility in cancer as well.

Given the rich assembly of promising cancer compounds in our pipeline, and given the significant market potential of what could be a new treatment for renal disease, the prioritization of a renal development path for this compound makes a lot of sense to us. We're developing a new formulation of the compound suitable for chronic administration in patients with renal and cardiac failure, and are conducting additional pharmacological testing with the goal of pursuing further clinical development in renal and cardiovascular disease.

The next two compounds in development, XL647 and XL999, are both spectrum selected kinase inhibitors. They are members of the protein family called receptor tyrocine kinases that are involved in the growth and proliferation of many kinds of celss, including cancer cells. The biology of these receptor tyrocine kinases is just beginning to be understood. And we are exploring what combination of these target molecules would be optimal for inhibiting certain types of tumors.

We have a broad-based program designed to identify compounds that selectively inhibit multiple or single kinases, and a number of promising, exciting and potent compounds have emerged from this program. The two most advanced are XL647 and 999.

647 is a potent inhibitor of the EGF receptor, a target of Exelixis, HER2, a target of (indiscernible) plus EphB4, a receptor targeting kinase. XL647 has good oral bioavailability and showed sustained inhibition of target kinases in vivo following a single world dose. In preclinical models in major tumor types including human breast, lung, colon and prostate cancer, 647 demonstrated potent inhibition of tumor growth.

Consistent with its spectrum of activity, analysis of tumors from XL647 treated animals showed significant decreases in both tumor vascularity and tumor cell proliferation and an increase in tumor cell death. 647 is currently in late preclinical developments. And we anticipate filing an IND application very shortly in this quarter.

XL999 is a potent inhibitor of key receptor tyrocine kinases that are implicated in the development and maintenance of the tumor vasculature and the proliferation of tumor cells themselves. XL999 simultaneously inhibits the FGH recepter, VEGF receptor, PDGF recepter and Flt3 kinases with high levels of potency.

Preclinical models of major tumor types including breast, lung, colon, and prostate cancer 999 demonstrated potent inhibition of tumor growth and induced tumor regression. 999 showed rapid onset of action in vivo with significant tumor apoptosis and necrosis and vascular disruption observed after a single oral dose in two different cancer models.

XL999 is suitable for both oral and intravenous dosing, and shows sustained inhibition of target kinases in vivo following a single oral dose. In addition, 999 is a potent inhibitor of Flt3, which is an important driver of cell proliferation in many patients with acute myelogenous leukemia and demonstrated remarkable potentcy in a Flt3 driven leukemia model. We're on track to file an IND for XL999 in the second quarter of 2004.

The next compound in this series is XL844, which is a potent selective inhibitor of Chk1 & 2, protein kinases that induce cell cycle arrest in response to a variety of DNA damaging agents. The Company believes that XL844 is the first potent selective Chk inhibitor to advance toward the clinic.

In preclinical studies 844 demonstrated significant potency in biochemical and cellular assays, good oral and bioavailability and an attractive pharmacokinetics profile. 844 potentiates the efficacy of chemotherapeutic agents in preclinical tumor models without a concomitant increase in systemic toxicity by exploiting genetic liabilities that arise during tumor cell expansion.

Exelixis will continue to evaluate the synergistic effects of XL844 in combination with a variety of DNA damaging agents in different cell arms, both in vitro and in vivo, and to explore the compound's potential as a radiation synthesizer. The Company anticipates filing an IND application for 844 in early 2005.

We also have a broad portfolio of compounds in rediscovery and optimization that can form the basis of potential IND applications in 2005 and beyond. These include XL9820 that targets VEGF receptor, Ty2 (ph), KIT and Flt3; XL 2880 that targets VEGF receptor, PDGF receptor, MET and Flt3; XL6309 that targets ALK; and XL2942 that targets serine-threonine kinase called P70S6K. All of these compounds are moving forward rapidly and all derive from our proprietary discovery efforts.

We're really proud of this productivity. We have generated a rich pipeline of promising cancer compounds with the potential for many more to come. We're confident that by continuing to generate multiple high-quality product opportunities we increase our odds of success over the long term.

And I will just comment now a little bit now on our partnerships, how we continue to enjoy a high level of support and collaboration from all of our partners. And we delivered on all of our partner goals in 2003. We extended our cancer collaboration with BMS, which is a relationship that we believe is one of the best and most collegial in the industry.

Similarly we believe that GSK is extremely pleased with the rapid progress we have made, and maintains a high level of interest in our expanded pipeline, and that relationship continues to grow very well. Our agriculture relationships with Bayer, Dow AgroSciences and Renessen are also proceeding well, and we believe are on track relative to deliverables and milestones for all of those collaborations.

And right now we believe that 2004 is shaping up to be a promising year for additional collaborations. We're already engaged in several substantive discussions, and the interest in us within the pharmaceutical industry is high. We tend to be very selective in structuring new alliances. We want to cultivate proprietary assets that we cannot at this time afford to fully develop on our own, and we need to get fair value.

Some of these relationships, if consummated, could have the beneficial effect of expanding our efforts into therapeutic areas outside of cancer. We are also looking at options to in license as well as out license, and we're engaged in several discussions along those lines. In contrast to year ago when a lot of our focus was on assimilating our collaboration with GSK, we are now aggressively pursuing several significant partnering opportunities.

We're focused on making good drugs and building a sustainable business. That takes a strong commitment to excellence and to execution to achieve the ambitious goals we have set and consistently met each year. We believe that 2004 is going to be another great year. We intend to get the XL119 trial started, file INDs and initiate Phase won trials for 647 and 999, and then Phase 4 (indiscernible) and several other compounds through preclinical testing and toward INDs in 2005, support our current partnerships and build new ones, and manage our cash and maintain very high levels and standards of fiscal responsibility.

Morale within the Company is strong. Personally I have never felt better about this Company, about its prospects for the future and about the people. And I know that that sentiment is shared by the rest of the management team and by the Board. And I thank you all for your attention. And we will be happy now to take any questions.

(OPERATOR INSTRUCTIONS). Matt Geller with CIBC.

Can you go through when we will expect the first, or in the case of XL119, the next clinical data from each of to these compounds? Anything coming up at any of the conferences this year? If not, when next year? Or if not, presentations on some of the underlying technology behind some of these compounds in preclinicals?

Let me answer that 119. The trial is going to start in the second quarter of this year. There are two interim analyses. One of them about a third of the way through the trial, and the other one about half the way through the trial. And so all we will learn at those interim analyses is whether to keep -- stop the trial or keep the trial going. And we can stop for futility, or we can stop for already having achieved success, or we can keep the trial going.

So the first data that that we will have relative to the ongoing Phase 3 trial is at the first year end analysis, which is approximately a third of the way through. And since you know how these trials go where they've ramp up and then they eventually reach their full enrollment speed, but you don't enroll a third patients in the first part of the time. It takes longer than that. So we anticipate nothing at this year's meeting based on our clinical trials. Whether the NCI investigators (technical difficulty) this year I just don't know.

On the other compounds, 647 and 999, we will file INDs this quarter and next quarter. We will start those in patients as soon as those INDs our active, and we will keep you posted. And we designed those trials, I believe, in a way that will obviously test the safety of the compounds, which is the main purpose of the Phase won trial, but also have some chance of actually showing us whether those compounds have any biological activity. So we're hoping that we can get that data relatively quickly.

One quick follow-up question. Maybe it is not quick. Can you talk about -- in terms of -- if you look at the compounds that are actually getting in clinic, what aspects of your underlying technology do you think are keying most into actual clinical products?

I don't quite know how to answer that question. I think it is a little bit of everything. These are receptor -- these first ones we talked about are receptor tyrosine kinases, 647 and 999 are receptor tyrosine kinases that inhibit a combination of kinases. Some of the kinases at these target have been in the literature and are subject to commercial drugs ERISA, or are the target of (indiscernible) clinical development. So we know that those targets yield a clinical benefit into compound. So it rapidly reflects our discovery capability. So I don't think we would be able to do this without either (technical difficulty).

It is not trivial to get these things developed quickly. The development is a regulated process. You can't do ninety-day comps in less than 90 days. So there are certain parts of it that you can't shorten. You can lengthen them if you don't do them right, and you have to redo them, of course. There other parts that you can. And the ability to turn these over quickly and take the data and file the INDs and meet our timelines also reflects, I think, the excellence and the growing size actually of our clinical and regulatory groups. I don't think you can single out any one part of the Company. I truly believe it is the integrated capabilities that we have.

Thanks. And congratulations on all the progress.

David Witzke with SunTrust.

I guess a follow-up on the prior question for George. If we look at the number of targets you screened last year, I guess maybe 35 to 40 range, if that is accurate, how many were on novel targets versus known targets where you looking for novel chemistry?

Is probably half and half. I don't know that that is an exact number, but it is some around there. We tried to have a mix. We have a lot of very interesting proprietary targets that have come out of our own internal R&D program. With the exception of ADAM-10 in 784 we haven't talked about any of those. 647 hits, among other things, the Efren (ph). That has not been previously targeted. That is a target that came out of our internal program. 647 also hits three kinases that are widely known.

So it is a combination. We have some things like 784 that are entirely derived from our internal programs. Some things like 647, which incorporate both novel biological insights and well-known insights. And then we have other molecules which target well-known validated targets, and that is in effort to spread the risk, and have a nice mix of compounds.

I should say that even on the ones that target multiple kinases a lot of the kinases that we have chosen to target and build in and out -- I say built in and out inhibition of certain kinases is based on biological insights that we have generated here. So even in those cases where the individual targets are known, particular combinations and the effectiveness of inhibiting that combination sometimes are based on proprietary insights that we have made.

And, George, was the 35 to 40 range correct? And is that correct going forward?

Yes. I think the exact number last year was 38. And this year we may -- we have the capability to run that number. We actually may run a few less. I suspect we will run about the same number.

On the RTK, how many can actually assay for out of the number of known kinases?

You know, there are about 500 and some odd. I forget the exact number -- 500 and some odd kinases. I think between what we can do internally and what we can contract on the outside we can assay almost 200 of those. I think that is the right number.

A final question on the ADAM-10 inhibitor. Did you give timing on next steps there? Do you have a formulation, an oral formulation set?

We did not give timing, and the reason is because of the formulation. We are developing a formulation of that which is not suitable for chronic dosing in renal failure. We need a better formulation that can be used for cancer. And so it is really dependent on the timing of that formulation. If it is trivial, we will ramp back into the clinic very quickly. And if it takes longer, it takes longer. And we just don't know.

Edward Tenthoff with Piper Jaffray.

Maybe it is a quick follow-up question from Matt. What should we be looking for at the end of March at the AACR? Do we have any preclinical data there or anything of interest?

Yes, we will have a presentation at AACR. Greg Plowman, our Head of Research, will be at AACR, and is on the program. And I don't remember exactly what session he is in, but he will be talking there.

Bret Holley with CIBC World Markets.

I just had a question on what drove the decision -- or it sounds like you're pursuing both cancer and renal failure for XL784. And I was just wondering what data you saw that drove the decision to pursue -- highlight renal failure versus cancer?

Sure. In a rat model of renal failure these are salt sensitive volec (ph) rats. And when you give them a high salt diet they develop hypertension. And in response to hypertension they develop renal and cardiac failure. We did a pharmacological test of 784 of capafril (ph), and another combination of 784 and capafril. Capafril ace inhibitors being the standard treatment for congestive heart failure.

So we saw some very encouraging data that Capafril, as expected, had an impact on both cardiac and renal failure in that model, slowed the progression of the organ degeneration. In the renal failure 784 had a larger impact than Capafril did. And the combination of the two had a bigger effect then either one alone. And based on those data we convened some pathologists, clinical people, to look at these data. They were quite captivated actually by the data, as were we. And so we believe that this compound has substantial potential in renal disease. It is a poorly treated and very large indication with a significant medical need.

So if we have the potential to develop a drug for that indication, it is not only mildly a highly ethical thing to do for those patients who are in need of it, but it is obviously a very big drug to launch. And so it is inconceivable that we wouldn't move down that development path with this kind of drug.

And at the same time we are continuing to do pharmacological experiments to carefully assess the types of renal failure in which 784 might be applicable, hypertension induced renal failure, diagnosed atrophy and other indications as well. So it is very encouraging data, a very large poorly met indication. That is really the two factors that caused us to prioritize that as a development route.

Just a quick follow-up. My understanding is since the development of 784 has begun in cancer that this probably is subject to the GlaxoSmithKline agreement? You can correct me if I'm wrong.

The question I have is that under that agreement you need to show Phase 2 proof of concept data. Would that include in this case just data in renal failure or cardiac failure potentially as well? And what kind of trial and what size of trial would you need to have to generate that data?

I think that is an issue that we probably will keep proprietary with GSK. I can tell you that in the contract -- you seem to know the contract pretty well -- there is a cap on the size of the Phase trial that we can do in that contract. So we will reach agreement with GSK on what the design of that trial would be like. We haven't reached that agreement yet, and I'm sure we will.

Charles Duncan with JMP Securities.

Congratulations on a good year of progress. I have a couple of related questions. You mentioned the collaboration environment, and you also talked about revenues increasing over the course of the next year. Could you give us some color specifically as to where you think there is demand for your current pipeline, or what the demand for potential collaborations is, and run through your assumptions and the increasing revenue year on year?

Let me answer it first and in a general sense. On a qualitative sense -- I have been doing this for a long time, both from within a biotech company and on the opposite side of the fence from a pharma company, and you can tell when you enter into negotiations or discussions whether these are serious and substantive, whether there is really legitimate interest on the part of the other parties, or whether they are just information gathering or curious, or whatever. And I can tell you that we're having a number of discussions right now that we believe are substantive. They are high-level. And where there is real interest on the part of the partners, and in several different types of collaborations with us. And those collaborations could involve different therapeutic areas.

And I can tell you that we are interested in the collaborations which will cover our costs. We obviously can't increase our burn rate. We would like to decrease our burn rate. And where we retain substantial equity in the products that come out of that. In order to do that, we're going to have to deliver compounds at some stage of development. So we are interested in collaborations that have those characteristics.

In fact, that seems to be what the pharma industry is interested in that too. I can't speak for the pharma industry certainly, and it is not a homogeneous group. But there seems to be a shifting of the emphasis in companies looking for technologies that will help their internal R&D, which is what companies were looking for a few years ago to mount an alternative source of compounds to supplement the pipeline that they make from their internal R&D.

So Exelixis and other companies that have capabilities, requisite capabilities -- I think, are in very strong position right now. And of course we did have some money from new collaborations, including in our 2004 revenue projections. But we're not prepared to break that out right now.

So you do expect new collaborations this time this year?

We have put that into our financial numbers, yes.

(OPERATOR INSTRUCTIONS) Karen Vokavitch (ph) with J.P. Morgan.

Congratulations on your progress this year, particularly on XL784 with the IND and the completion of Phase 1. I have a couple of questions on the therapeutic index of some of your leading compounds. For XL784 when you speak of it being free of side effects, is that your target dose, therapeutic dose, or is it several times several times above?

Yes. Certainly at a dosage level that exceeded anything that we envisioned as a therapeutic level, it was free of side effects. So as high as we took it, we didn't see significant side effects, which was above the projected therapeutic range.

And with XL647, when you see regression in tumor cell models, is this at a dose free of side effects, or are you really maxing out there? I'm thinking of 8431 data?

Yes. What we do in the (indiscernible) model, and as you know it differs from species to species, it is determined a maximum tolerated dose. And then we test the compounds at that maximum tolerated dose, as well as a variety of doses below the maximum tolerated dose. And obviously the higher dose you give the more effect you see. Now I can tell you even at dosage levels well below the maximum tolerated dose there is slowing or stopping of the tumor cell growth. And at the maximum tolerated dose there is quite a pronounced regression of the tumor.

And lastly with XL999 when you compare it to Gleavitt (ph) in the mouse models looking at survival, it fares better than Gleavitt. You have more survival -- the mice survive longer -- more mice survive? Is this it at an optimum dose of Gleavitt?

Yes. Bit you have to remember that Gleavitt is really part of abos (ph) in chronic myelogenous leukemia. And that the model and the type of leukemia for which 999 seems to the most effective is acute myelogenous leukemia, which is driven by Flt3 and not by abos. So, yes, that is a therapeutically relevant dose of Gleavitt, but it is not the type of leukemia for which Gleavitt is commercially available.

And lastly the Renessen collaboration started a little bit over a year ago. Can you give us a little more insight into how that is progressing and --?

Sure. I guess I will answer that for those people on the call who are unfamiliar. Renessen is a joint venture between Cargo (ph) and Monsanto. We entered into a collaboration with Renessen to identify genes that confer on soy and canola high oil content. And we have identified a number of such genes in a model plant called arabenocous (ph). These are interesting genes because while they raise oil content, they do not concomitantly decrease protein content, which has been the case with a number of genes before.

We tended a number of those genes over to Renessen. They have tested a number of them. Testing means identifying the equivalent genes from the commercially important species. And the data that we generated seems to be confirmed. The real proof of that pudding here is whether the genes will confer high oil content on the soy and canola. And we don't just have those beta yet. We will have some data this summer and some data the following summer. So far this collaboration is going extraordinarily well. All the data that we have so far are extremely confident and extremely encouraging, but we haven't really reached the critical milestones yet.

Aaron Schwimmer with Goldman Sachs.

This is May-Kin Ho actually. On your RTK inhibitors, since you're also targeting some new molecules here, what do you predict to be the potential side effects of these because they have multiple actions?

Yes, they do. I think in practice all of the kinase inhibitors hit multiple targets. So if you take Gleavitt, which was thought to be specific for abos kinase, it hits a number of other RTKs. If you take (indiscernible) compounds 906, which was developed for graph (ph) kinase inhibition. It also inhibits a number of the (technical difficulty).

So the fact that bars (ph) inhibit multiple kinases is not unique. I think what is unique is we have -- we appreciated the fact early that it had hit multiple kinases, and we were able to optimize the combination. I think with some of the other kinases the side effects you see are related to scan (ph) in the GI tract. We would expect bars to be similar in that regard. And of course it is in the end of the combination of therapeutic levels to side effects. And we just won't know that until we get human data.

How does your 999 compare to let's say Millenium's fit (ph) free inhibitor, which they are targeting AML for?

Right. I don't know the answer. I can tell you biochemically we don't know of any compounds that are more potent prtotease inhibitors than ours. I think in the end it will depend on which compounds have the highest effective level at the sight of action. And the biochemical potency, if you have more biochemical potency, it certainly gives you a leg up. You would take higher than lower. But that is only one of a number of factors that contribute to the effectiveness at the sight. And the others are things that have to do with (technical difficulty). And then there is some tissue distribution and things that you can predict somewhat from mammals, but vary a lot from species to species. We will tell you a lot more once they get into humans.

And then --.

I'm not trying to be glib, but I think we're just trying to be cautious about what statements we make. In other words, these look as good as they can look right now. They look really interesting. And so in animal models we are very confident that these compounds are as good or better than any compounds we have tested. But we obviously haven't had them in humans.

I know that somebody asked this a little bit before in terms of 844. What exactly are your rights here? Do you actually feel that it is better to have a partner earlier on to move into areas like cardiac failure?

Yes, absolutely. It is an interesting decision for us because what I said is true. There is a huge market and is not very well treated right now. And we have a compound that at least based on the data that we have to date seems to have a major impact on that, on renal failure.

On the other hand, we are not prepared as Exelixis to undergo -- undertake a large Phase 3 trial with many thousands of patients all over a long period of time. That is not a trial that is within our financial capability of undertaking, or within our area of expertise. The 784 does fall under the GSK arrangement. They don't have to decide whether they are going to take it or not until Phase 2a. And we are -- of course Phase 2a in theory means some human safety data to show that this is the most obvious toxicity risks are taken care of, and some signs of biological activity.

And so the good thing about the good thing about renal failure is there are lots of survey marker around, or markets for higher kidney dysfunction. And so it should be relatively straightforward to design a reasonably sized trial do just demonstrate some data to show whether 784 is having an impact or not. And we haven't yet had the discussion with GSK about what's they would consider to be the right body of data. And we will have that discussion soon.

If they decide in a sense to opt in, what happens?

It is just like any other compound, that collaboration. If they decide to opt in they pay us a substantial milestone payment at the time of opt in. They then take over the clinical development. They would do all subsequent clinical trials, Phase 2 and 3, and market introductions and the marketing. We would get additional milestones. We have very nice royalty range in there, as a range, but all the ranges are in double digits. And we have copromotion rights in North America. So we will participate meaningfully in the economics of those products should GSK take them. If GSK decides not to take them then basically we own them. So that is the situation.

And most of these -- as you know, drug candidates for let's say heart failure, etc., generally one would use a library of compounds and there are backups. Do you have that with 874?

Yes, we have a whole series of compounds in back of 784. And yes, we have them there, but so far we have no reason to think that 784 isn't a good compound. Usually the reason you have a backup compound is because your first compounds shows some toxicity. Your first compound has great bioavailability in the animal models but zero in humans or very low viability in humans. We know that is not the case with 784. So we know it has good bioavailability. We know that it has a good half-life in humans. We know it is well-tolerated, at least in the Phase 1 trial. So it looks like this is a well behaved compound.

What is the half-life and how is it metabolized?

Yes, the half-life is eight hours, which is about what you want. And metabolism, so far we really don't have a good picture of that.

David Witzke with SunTrust.

A couple of quick financial questions. First the cash balance ended up significantly higher than our estimate. Did you tap into a financial loan facility from GSK, or was there something in the cash balance?

Yes, we have indeed. There are two significant events in the last year that we have to take into account. One is the equity follow on (ph) offering which raised about $75 million. And then secondly we drew down $30 million under the GSK loan facility.

Was the draw down fourth quarter?

Yes.

A question on the increase in R&D costs for next year. How much is out-of-pocket clinical costs versus adding development personnel? I guess the breakout where the incremental R&D dollar is going?

We typically don't break this out. And I would prefer not to talk about it. Actually I don't think we have those numbers available.

Will the increases be more back-end loaded in the year? Could you comment on that?

We're trying to be careful here because we have not broken those out in the past for some good reasons. We prefer to stick to that.

Karen Vokavitch with J.P. Morgan.

Hi, Frank. You explained the increase in 2002 -- 2003 revenues relative to 2002. What about for the quarter, for the fourth quarter? Is it pretty much the same reasons or does something else stand out?

No, is pretty much the same reason.

There are no further questions. Are their any closing remarks?

Let me just thank everybody for your attention. And I appreciate your interest in the Company. We think we really did have a great year and going to have another one this year. So keep tuned.

This concludes today's Exelixis fourth quarter and year end 2003 financial results conference call and outlook. You may now disconnect.