Exelixis Inc (EXEL) 2005 Q1 法說會逐字稿

Good afternoon, ladies and gentlemen, and welcome to today's Exelixis first quarter 2005 financial results conference call. At this time, all lines have been placed on a listen-only mode and the floor will be open for questions following today's presentation. It is now my pleasure to turn the call over to your host, Mr. Charles Butler. Sir, you may begin.

Thank you. Good afternoon and thank you for joining the Exelixis management team on our first quarter 2005 financial results and business update conference call. Participating in today's call are George Scangos, President and CEO, and Frank Karbe, Senior Vice President and CFO.

Following this introduction, George will give a brief recap of the quarter, and some perspective on 2005. Frank will then review the Company's financial performance for the first quarter, and following the financials, George will provide additional comments and outlook on the year ahead, after which we will open the call for questions. But first, allow me to make our disclosure statement regarding forward-looking statements in today's call.

Please note the following discussion contains certain statements that are forward looking, including answers to questions at the end of the formal remarks. These statements are only predictions and are based upon our current expectations. Forward-looking statements involve risks and uncertainties.

Our actual results and the timing of events could differ materially from those anticipated in our forward-looking statements as a result of these risks and uncertainties, which include, our ability to enter into new collaborations, continue existing collaborations, and receive milestones and royalties derived from future products developed from research efforts under collaborative agreements; our estimated future license and contract revenues, as well as expenses; our estimated future balances of cash, cash equivalents, short-term investments and restricted cash; the potential failure of our product candidates -- to demonstrate -- to demonstrate safety and efficacy in clinical testing; our ability to file IND applications and initiate clinical trials at the referenced times; our ability to conduct the Phase III clinical trial for XL119, sufficient to achieve FDA approval; plans to conduct Phase I trial for XL784 in 2005; our ability to conduct the Phase I trials for XL647, 999, 880 sufficient to achieve positive completion; our ability to successfully initiate Phase I trials for XL820 and 844; our ability to successfully advance and develop additional compounds including XL184; and our ability to develop drug candidates and/or INDs as part of the metabolism program.

These and other risk factors are discussed under "risk factors" in our annual report for the year-ended December 31, 2004 and other SEC reports. We expressly disclaim any obligation or undertaking to release publicly any updates or revisions to any forward-looking statements contained herein.

And with that, I will now turn it over to George.

Okay. Thanks, Charles. I'm glad you had to read that instead of me. I will save the bulk of my comments for the end, but I just wanted to say I think we've had a remarkable year so far. We've conducted one Phase III trial for XL119 and two Phase I trials for 647 and 999 and kept them all on schedule. We began a Phase I trial for XL880, we filed two INDs for 820 and 844. We brought an additional two compounds into development, those are XL281, which inhibits RAF and XL418 which inhibits Akt and S6k.

We favorably renegotiated the terms of our amendment with GSK such that the payments are accelerated in GSKs, and therefore, Exelixis rights are defined. We achieved two milestone payments totaling 35 million from GSK, which we announced this morning. So I think further demonstrate the excellent progress we made in that collaboration.

We've made excellent progress as well in a number of other business discussions. We have continued to build out the infrastructure of our Company to support our growth. We have taken concrete steps to finance our growing clinical -- our activities without a large dilutive financing. At this point I think the organization is really firing on all cylinders. We are moving full-speed ahead. We are looking forward to the clinical data that will emerge from our compounds.

I'll have more to say about these at the end, but first, I'll turn the call over to Frank, who will give the financial report.

Thanks, George. As George just mentioned, one of our goals is to ensure that we have the organizational structure and financial resources we need to continue our productivity and advance compounds in our pipeline. As I've explained in the past, we think about financing in multiple dimensions. Using multiple sources of funding to finance our activities.

This includes executing on our existing partnerships, which could result in substantial milestone payments; seeking new partnerships for some of our yet-unpartnered assets; and exploring the suitability of clinical development financing vehicles. The achievement of the 35 million in milestones from GSK that we announced just this morning is one example of how we turn our financing strategy into a reality. And we're working hard on putting other pieces of the plan in place as well.

As a further example, the amendment to our GSK collaboration provided an important step in laying the groundwork for exploring the suitability of clinical development financing vehicles for a subset of our clinical compounds. Furthermore, the amendment to our Genoptera collaboration with Bayer Crop Sciences also supports our organizational and financial needs.

The amendment resulted in the termination of our research efforts effective March 31st, '05, and was structured in a way that we believe fully accommodates us for any financial disadvantages resulting from the termination through a $10.9 million termination fee. More importantly, the termination allows us to shift resources to our clinical and preclinical pipeline of compounds.

Let me now turn to the financial results for the first quarter of 2005. And as usual we are reporting results both on a GAAP and non-GAAP basis. Non-GAAP results exclude restructuring expense, non-cash charges for stock compensation, as well as amortization of intangibles. A reconciliation of GAAP results to non-GAAP is contained in the press release announcing our 2005 first quarter results, which is on our website at www.exelixis.com.

Let me begin with net loss. Net loss on a Generally Accepted Accounting Principles was approximately 27.4 million, or $0.36 per share. Compared to a GAAP net loss of 28.8 million or $0.40 per share for the first quarter of '04. Non-GAAP net loss was approximately 27.2 million, or $0.36 per share. Compared to approximately 28.1 million, or $0.39 per share for the first quarter of 2004. Cash, cash equivalents, short-term investments and restricted cash totalled approximately 139.1 million at March 31st, '05, as compared to 171.2 at the end of 2004.

Let me turn to revenues. Revenues were approximately 12.9 million compared to 11.9 million for the same period in '04. The increase from 2004 to 2005 was primarily a result of the increase in research and development funding from our collaboration of GlaxoSmithKline and additional research and development funding from our collaboration with Sankyo, which was acquired through our acquisition of X-Ceptor last year, but which expired in March 2005.

These increases were partially offset by a decrease in revenues related to the expiration of one of our Bristol-Myers Squibb collaborations during July 2004, and a decrease in revenues related to the termination of our combinatorial chemistry collaboration.

Research and development expenses were approximately 33.3 million, compared to 34.2 million for the comparable period in '04. The decrease from '04 to '05 was primarily a result of a decrease in lab supplies as a result of our June '04 restructuring and the termination of most of our combinatorial chemistry collaboration. The decrease was partially offset by increases related to the expansion of our development activities associated with advancing our clinical and preclinical development activities, which we expect to further increase in future periods. The total FTE headcount for the Company was 516 as of March 31st, '05, and our increase in headcount is primarily driven by the expansion of our development group.

I'm going to finally turn to general and administrative expenses. G&A expenses were approximately 6.2 million compared to 5.6 million for the comparable period in '04. The increase from '04 to '05 was primarily a result of increases in legal, insurance and facility expenses to support our clinical and preclinical development activities.

Now let me turn the call back to George, who will give you further background on our accomplishments thus far for 2005, and a bit of an outlook for what lies ahead.

Okay, thanks, Frank. I will spend a few minutes reviewing the progress of our compounds and then I'll come back at the end to some of the business and financial issues that are in front of us. I think to start with the most-advanced compound XL119, the trial is on track. Currently there are 49 active centers and enrollment is in line with our projections.

The Phase I trial for 647, which you may remember inhibits an EGF receptor, HER2, and VEGF is progressing well at Stanford and at the Mayo Clinic. Similarly, the Phase I trial for 999, which inhibits FGF, PDGF and VEGF receptors continues to progress rapidly at the University of Texas, San Antonio and Case Western. We anticipate that both of those Phase I trials will be concluded this year.

We are far enough along in both trials to have reasonable confidence that both of the compounds will proceed into Phase II. We intend to conduct broad Phase II programs for each compound, and of course the final Phase II trial programs are still being worked out. But basically we're taking note of the results we're getting in Phase I. We know what compounds that are more advanced in clinical development have done, and we know a lot about our compounds and we'll take all three of those factors into consideration as we design the Phase II program. And, again, we anticipate that both of those Phase II programs will begin in the second half of the year.

The Phase I trial for XL880, which inhibits Met and VEGF is underway at Wayne State University and the Dana Farber Cancer Institute. Patients are now being enrolled. XL784, which inhibits ADAM-10 is moving rapidly towards the Phase I/II trial in renal failure, which we anticipate also will begin in the second half of the year.

The INDs for 820, which inhibits VEGF, KIT, and PDGF. And 844, excuse me, which inhibits Chk1 and Chk2 were submitted in April, and those Phase I trials will commence once the IMEs are active. We've also made good progress on XL184 and anticipate finally that IND in the first half of this year. And we've also brought two additional compounds into preclinical development as the first candidates for INDs next year. These are XL281, which inhibits RAF, and XL418, which inhibits Akt and S6k. Both of those compounds are now moving through preclinical development and have shown encouraging preclinical activity.

So if you look out later this year, we expect to have one Phase III trial for 119, three Phase II programs for 647, 999, and 784. Four Phase I trials for 880, 820, 844, and 184, and additional preclinical programs. This is quite amazing productivity and has really surprised even us. And again I want to emphasize that we've done our best to make sure that each of these compounds is high quality, and we believe that each of them has the potential to be either first-in-class, best-in-class, or both.

And as we move forward, certainly we'll have some attrition among these compounds. But one good aspect of our strategy is that as we move forward, with so many compounds, we actually are incented to kill less promising compounds so that we can adequately fund others. That of course would be different if we only had one or two compounds, our incentive would be good reasons to keep them going. The compounds that we -- that continue to move forward do so because, based on the data that we have to date, we believe that the compounds have substantial potential.

So the good news here, is that we generated a large robust pipeline of high quality potentially important compounds in a very short period of time. I think that productivity really hasn't been surpassed anywhere, to my knowledge. The question that many of you have, of course, is how we are going to pay for this remarkable pipeline as more and more of the compounds progress to the clinic, and I think you've already seen the first -- some of the first steps in that direction.

You note that our cash balance, as Frank reported at the end of first quarter, was 139 million, since that time we received almost 11 million from Genoptera, and we just announced an additional 35 million in milestones from GSK. And, of course, we received the annual GSK R&D payment in the fourth quarter. So even without that R&D payment, the other two events bring the effective Q1 cash balance to the equivalent of 185 million.

So we're working on a number of other transactions that have the potential to increase our cash balance even further, so we always like to have more money, of course, but we're comfortable that with the resources we have, we have sufficient resources to continue to aggressively develop our compounds.

Now, all of this progress hasn't come about through magic. It's a result of the talent and I think incredibly hard work of the employees of Exelixis. So as I conclude here, I think many of the employees listen to these calls. And I'd like to take this opportunity to really publicly thank them for the extra effort they've put in and for all of their accomplishments. We've come a long way. We've made great progress. We are aware we have a long way to go, but we can see the light at the end of the tunnel at this point and we're all anxiously and eagerly looking forward to the benefits that our compounds may bring to patients.

So at this point, thank you all for your attention and I'll stop and open up the call for questions.

Thank you. The floor is now open for questions. [OPERATOR INSTRUCTIONS] We have our first question coming from Hong Li of SunTrust Robinson.

Oh, thanks. Just a quick question. Does GSK have option right to the compounds that you acquired from X-Ceptor?

No. GSK does -- those are not part of our collaboration with GSK.

So the two preclinical compound you mentioned, 281 and 418, are they X-Ceptor compounds?

No, those are not X-Ceptor compounds. Those are cancer compounds. Those are -- 281 is targeted at RAF, and 418 is targeted to Akt and S6k. And both of those compounds are a part of our collaboration with GSK. Remember, there are 12 programs that are part of that collaboration with GSK from which GSK gets to pick two or three, the remainder of them belong to Exelixis.

Okay. I think the -- your last quarterly call you mentioned that we're going to see some data at ASCO for 647 -- 647.

Yes. There is an abstract in the abstract book at ASCO on 647.

Okay. One last question. This is for Frank. Regarding the cash [inaudible] GSK milestone payment just announced today. So in the second quarter -- this quarter you received 21 million cash payment from termination of a Bayer collaboration of the insecticide collaboration, and also on top of this there is 35 million milestone payment from GSK, and understand that you're going to recognize 50% of GSK, 35 million milestone payment right away. So that if this is based on if the R&D expense is holding constant, you may have a positive bottom line EPS for second quarter. Am I right or --?

Well, we may get very close. We don't know exactly yet where we may come out, but we may get very close to that, that's right. But let me comment on one other remark you have made on how we might account for the $35 million milestones from GSK. Turns out that there are potentially two different ways how we could do that. As you know, historically we've always said we amortize milestones over the term of the underlying agreement. The question here is, is the underlying agreement the original GSK agreement that was signed in October 2002? Or is it the amended agreement that was signed in January '05? If it's the later, we recognize actually less revenue this year and more north of subsequent years. And that is still under review and the final determination has not yet been made.

With regards to the Genoptera revenue, we will recognize $21.1 million of revenue from the termination of the Genoptera JV in Q2 that's made up of two components, the $10.9 million termination fee, which we have already received and is currently booked in Q1 on a deferred revenue on our balance sheet , and the remainder is the acceleration of deferred revenue.

I see. Okay. Thank you.

Thank you. We have our next question coming from -- we have our next question coming from Ted Tenthoff of Piper Jaffray.

Thank you. Can you hear me okay?

Yes. We can hear you fine, Ted.

Hi. How's everything? Looking forward to seeing you next week. Real quickly on XL647, I know that -- I believe that it's just going to be an abstract in the book, you had mentioned that it continues to progress. Can you give us any more update on that? And then maybe even a kind of more 30,000-foot question. And I'm sure that this will depend on kind of how each drug plays out and the benefits of each, but how do you plan to develop this multitude of cancer drugs in an increasingly competitive regulatory environment?

Yes. That's a good question. As far as the specific question about 647 or -- and 999 for that matter. What we can tell you is they are well-behaved pharmaceuticals. They are -- there's good bio availability, they have good half lives and so they look good. We're obviously not in a position to specifically comment on any of the data in the trials until those trials are concluded.

But your 30,000-foot question, we have taken each of these compounds, of course, and looked at them preclinically. We have compared them to other compounds. We have not moved forward compounds that we don't believe stack up -- unless we believe they stack up extremely favorably in terms of potency, selectivity and other aspects. So we anticipate that our compounds in the end will provide or may provide a substantial therapeutic benefit. The cancer space is getting very crowded, there are many companies developing kinase inhibitors, we're aware of all of that. We believe ours stack up extremely well.

In terms of the clinical design program, we are still working that out and what I can tell you at this point is that we anticipate that we will conduct a mixture of trial. Some of them rather standard in their design and some of them more innovative in their design, in an effort both to ascertain quickly the expense to which our compound of new provided a benefit for patients and then obviously to demonstrate that they provide benefits to patients that would make them have substantial market penetration.

Okay. Good.

And beyond that, I really don't want to get too specific at this point.

Understood. Thank you.

Thank you. We have our next question coming from Jeffrey Zekauskas of J.P. Morgan.

Hi, good afternoon. Just a couple of questions. How many -- all things being equal, for how many compounds would you expect to have Phase II results in 2006?

Well, that's a hard question to answer, Jeff. Our goal --

Or what would be the goal?

Yes. I mean what I -- I don't want to state it even as an official Exelixis goal. I would say personally I would like to have Phase II data for 647 for 999 and for 784, potentially for 880. Now, whether we can achieve that or not depends on a lot of factors. It depends on when the Phase I trials actually come to an end, we don't know that yet. It depends -- and -- on the design of the -- some of the Phase II trials, which we're just finalizing now.

And, frankly, it's also partly a budgetary consideration. Because if you decide how many patients you need and you know how many patients per center you can enroll, you can extrapolate and do the rather simple math to decide how many centers you need, but of course the more centers, the more expense that you have. So we're trying to balance all of those factors right now and -- but the goal is to have -- if at all possible, to generate Phase II data for multiple of those compounds in 2006.

Let me try it a different way. If you had to guess which order you might get the results, how would you do that?

I think it's completely data driven. 999 and 647 are both fairly far along in Phase I trials.

Right.

Those two will each go into Phase II programs. Which of those trials generates interesting data first is very hard to project. 784 is on a very different trajectory. It -- we finished the Phase I trial for 784. This was a single-dose trial with a liquid formulation, as we said we now have a -- capsules of 784 which are bioequivalent preclinically, so we have to show that they also are bioequivalent clinically in humans. We have single-dose Phase I studies, we will need to do repeated dosing and we can rapidly move into a Phase II program. So I -- all of those are moving kind of in parallel at the moment. But they won't -- I'm sure they won't all end in parallel, but it's very difficult at this point to say which of them will come first.

Okay. And just a couple of questions for Frank, if I may. With the termination of the chemistry collaborations, all things being equal, this is maybe 5.5 million in revenue and -- are the costs greater than 5.5 million or the normalized costs lower than 5.5 million?

We've -- we've never really disclosed that, but there is a substantial amount of costs associated with that.

In other words, is this accretive or dilutive to you, by terminating the transactions?

Again, I think we've never really given results by division for the Company and I don't want to do this year now.

Yes. I think, Jeff, what we did say is that at the beginning of those collaborations, they were profitable. As time went on, the types of structures that we were being asked to make were more complicated and, therefore, more costly. And so it didn't make sense from a business perspective to continue to conduct them. And they were also kind of a diverting effort from our own internal projects and were not generating a lot of free cash so that's why we terminated.

And in the 13 million that you reported for this quarter, the 12.9, are those all cash revenues? In other words, will you get 12.9 million in cash, or will you get some other number?

No. That's not all cash. That is what we recognize as revenue, some of that includes the recognition of upfront payments from some of our collaborations, that includes the amortization of the annual R&D funding that we get from GSK, etc. The cash number's actually different from that.

Okay. Is there -- and there's no wacky trinopterin [ph] number in there?

No. There's no wacky trinopterin [ph] number in there.

I would say, Jeff, that we don't have any wacky numbers anywhere.

We can be -- on Genoptera.

It wasn't a good choice of words.

Well, the Genoptera revenue portion is pretty much unchanged in Q1 from what it has in the past. It was actually the last quarter when we still received R&D funding from Genoptera.

Okay. Thank you very much.

Thank you. Our next question is coming from Eric Schmidt of SG Cowen.

Good afternoon. George, I was hoping for just a little bit more information on the Go decision on XL647 and 999 into Phase II studies. I assume that you wouldn't be basing that decision on just good bio availability and good half lives, but that there also might be an efficacy component to that decision. Is that a fair conclusion?

Eric, I really can't answer that question one way or the other. I really can't comment on data the we've seen. I can tell you that these are decisions where it has to do with a number of things. If you see, hypothetically speaking, some responses, that gives you some encouragement. If you see that your compounds are in plasma and they have a good half life and that the levels in plasma correlate with therapeutic levels that you saw in animals, that gives you encouragement to go forward. So it's based on a number of factors.

Can you tell us about when we might see more complete Phase I data from these trials than is presented in the ASCO abstract for 647, would that be a second half event also?

Yes, yes.

And then last question, you mentioned these alternative financing vehicles a couple of times now over the last several calls. Could you talk maybe a little bit more about when we might see such a deal and potentially the likely form such a deal structure might take?

Yes. I think we -- we make it a policy not to comment on the timing of any business development transactions. But you're right, we have been working on these for a while. We've been talking about them for a while. We've continued to make progress.

I think the form is relatively straightforward, that a financial investor would put up money for the clinical development of one or more of our compounds. We would do it in a way that it was on our balance sheet. We're not looking for any wacky off-balance sheet financing or funny financing, we want to keep it on our balance sheet. We want to keep control of the trials ourselves. We're not giving up control of the trials. But we would have the resources from a financial investor to conduct a broad Phase II program for those compounds. So we're not budget constrained and having to focus on one Phase II trial when we might want to do three, four, five Phase II trials in parallel and we would be able to do that.

Should the compound succeed in those trials and merit further development, then we have the unequivocal, unquestioned right to buy the compounds back. We would do that, of course, at a price that provides the investors who put up the money with a fair return. If those compounds fail and they don't merit further development, then we have no obligation to buy them back. So we will -- we would -- somebody else would have funded that development.

Now, there have been a number of these things in the past, and the terms can be attractive or the terms can be rather unattractive, to say the least. And I think the -- the terms you can get on this type of transaction really depends on the perceived quality of the compounds and the number of the compounds because both the quality and the number are -- relate directly to the risk that the investors are taking, so the higher quality, the lower the risk, the more better in the vehicle -- the lower the risk. And so we believe that we'll be able to get very attractive terms for these vehicles and that they will be really a minimally dilutive way to provide a substantial amount of financing, to go ahead and aggressively develop the compounds.

I guess it's fair to say by pursuing these discussions that you're finding the terms for such a privately structured deal more attractive than the terms the public markets are offering now in terms of just raising straight equity; is that fair to say?

Yes, absolutely.

Yes, that's fair to say.

Yes, that's quite fair to say.

Thanks a lot.

Yes.

Thank you. Our next question is coming from George Farmer of Wachovia.

Hi. Thanks for taking my question. George, regarding 647 and 999, you guys are committed to the Go decision, is that what I'm hearing correctly?

Yes, I mean, unless we get some horrible news tomorrow. If not -- you can always learn something that you don't know, but we have a lot of data. Where the compounds are looking good so far and we are full-speed ahead.

Is that right? So you've hit an MTD with both of those compounds?

We're really not commenting on that.

And regarding 784, are you going to Phase I/II trial in diabetic nephropathy patients.

Right.

Are you going into healthy volunteers first, and then moving into a disease setting or are you going to go right into a diseased patient.

That's a good question, George. And our current thing is we're going to healthy volunteers first because to do the bioequivalent study to do the multi-dose study you can do that really quickly if you do it in healthy volunteers. It's very short, it's a month or two to do that in healthy volunteers. If you do it in patients, then of course, enrollment of the patients is slower and you stretch out that part of the trial. So you can get to the end point faster if you do those initial parts in healthy volunteers, which is why it's called a Phase I/II trial, because if you do this in volunteers, it technically is a Phase I trial.

So you'll be able to get -- figure out bioequivalency in the Phase I portion and then just move that whatever that dose of it might be into the Phase II portion.

Exactly.

And how big of a trial would this be in the Phase II?

It's not -- it's not that big. I don't actually remember the exact number.

Is it a control trial?

Yes. We'd have to do a control --.

And what are the end points you're looking for in that trial?

Well, we'd be looking for proteinuria [ph] and a reduction in the rate of increase in the rate of increase of proteinuria [ph]. And we've had a lot of discussions as to what types of patients you want to do. Do you want to do it very early in the disease where you are looking at microalbuminuria or later on and I think the current thinking is to take kind of mid-stage patients. Early patients don't progress rapidly enough, you need too many of them and follow them for too long. Late-stage patients might be too difficult to treat, so it seems there's a slice of patients in the middle in terms of how much protein they're excreting, which would be the basis for the trial.

Okay. And, Frank, could you comment on what sort of payment you're expecting in the fourth quarter from GSK?

Yes. Our annual R&D payment that has actually increased as of the third anniversary of the agreement and that is as of last October. And we never really set what it is on an annual rate, but we've set what the aggregate number was for the remainder of the term when we announced the amendment of the collaboration in January and that number was 47.5 million.

Okay. And you talked about different ways you can amortize the -- this $35 million payment that you received today. And you talked about the different -- depending upon the structure of which deal you base it on. Just to kind of help us in our modeling, can you give us any more flavor for how -- what direction you guys are thinking about -- about how that will be amortized over time?

I think my current estimate it's probably going to be the later of the two methods that outlined, ie, starting the amortization as of the signing date of the amended collaboration, ie, January '05. And we would then amortize it until -- over a period stretching over 2010.

Okay. Great. Thank you very much.

Thank you. Our next question is coming from Charles Duncan of JMP Securities.

Hey, guys. Congratulations on the IND filings and the milestone payments. I had a quick question on 119. You didn't give much color on what you meant by that compound being on track. Can you -- can you provide us a little bit more detail?

Yes. What we have said is that we anticipate that that trial will take three years, if we have to take it all to completion. Remember, there are two inner moniliasis built in, but if we take it to completion, it's a three-year trial and that of course is based on assumptions about enrollment rates. And we can tell you that our enrollment rates are in line with the projections that would allow that trial to be completed in that three-year time span. So they're about what we had predicted.

Could you remind me of the total number of patients that you had anticipated enrolling in that?

600.

And then some of the hurdles with regard to what you expected for clinical activity?

Yes. The -- what we -- the reason we have 600 patients in there is 300 in the treated group.

Sure.

The 119 group and 300 in a control group, is that we have powered the study in order to show a two-month increase in survival. And in order to do that, you need the 600 patients. And we have an SPA with the FDA where a two-month increase in survival is an agreed-upon end point.

It's two months off of a base of six?.

Six. Exactly.

Okay. So enrollment's on track in terms of meeting your expectations?

Yes.

It hasn't -- has it been slower than expectations or anything or --

Well, if it were slower than expectations then it wouldn't be in line with our expectations.

It wouldn't be meeting, that's it. Okay. Next question and apologize for the additional 30,000-foot question. But how do you -- how do you value such a large pipeline? Are you -- are you starting to see synergies not only in the discovery of clinically active compounds, but potentially synergies in the clinical development of these compounds, George?

Well, I don't quite -- I'm not sure I understand the question, Charles. The -- I think maybe I can answer it in a couple of different ways. We have -- I think there is a value to having a large pipeline that goes way beyond just multiple shots on goal. Five compounds instead of one and the only thing that meant was we had five shots on goal rather than one, I'm not sure that's a very good investment thesis, actually.

But what I believe that it means is that each of those five is better than it would be if we only had one because we -- we have an organization of critical mass and excellence, we can hire better people, we have a better organization. You can't hire the same quality of people to head up an organization of 10 people as you can to hire an organization of a couple of hundred people. And so I think you get better people, you get more experienced people, you can -- you can build up an infrastructure that allows you to characterize those compounds more carefully, which you couldn't afford if you were a smaller organization.

So each of them gets characterized, I think, more thoroughly than they would. So I believe that the average quality of those goes up as well. And then you get experience in developing them, that's true. And what you learn from the first one informs the second and so on down the line. So you get better over time. And we have seen that, frankly, as well. We have a lot of very experienced people in here.

But the first time an organization goes through a process, filing an IND, conducting the Phase I trial, whatever it is. Even if everybody's done it before, the first time the organization does it, there's some organizational learning that has to occur, and so each subsequent time you do that you get better -- the organization gets better at it. So I think there are a lot of efficiencies and synergies that occur as a result of having multiple compounds. I -- was that the gist of your question?

It also seems -- yes, for the most part. But it also seems to me that you're able to kill compounds quicker earlier if you have others that you can fall back on.

Yes. And that's absolutely true. Part of our philosophy here is to kill things quickly and early. And we have killed a lot of things quickly and early and those -- of course, we don't talk about those. We talk about the ones that are alive and moving forward. But at each stage, we have had multiple compounds that we have worked on for a while and have not moved them forward because we didn't feel that we had the data that justified moving them forward. And I'm sure that will continue to happen.

We have all of these compounds in our pipeline now. I don't think anybody listening to the call will believe that all of them are going to reach the market. We don't believe they will all reach the market. We know we will have some attrition and I think it -- the -- having a large pipeline like that allows us to make the tough calls, kill compounds, if the data are equivocal or the data aren't encouraging, then put those resources behind more promising compounds, and we won't hesitate to do that.

But from a statistical perspective, it looks like you have a critical mass whereas many of your mind [ph] share competitors don't, if you have normal attrition rates, right?

No argument from me.

But what I was specifically asking is with regard to clinical development, are you able to better focus your development efforts because you have such a bevy of products in terms of targeting patients and working with specific clinical investigators?

Yes, we are. I think we -- we have -- part of the philosophy, though, that each of these compounds are -- and frankly a lot of the competitors' compounds really have been designed to target specific receptors or specific combinations of receptors based on our current understanding of what types of tumors will respond to that combination of therapy. And so to that extent, yes. And we have designed -- each of the compounds has a unique spectrum and for a different reason.

Having said that, in the end there's -- even though we are -- we and everybody else is incredibly specific, that's how you design and make these compounds, you still cannot predict completely how they'll behave in the clinic, and you can make some educated guesses as to what kind of tumors will respond, but you can't really predict that with complete certainty. So you've just got to go ahead and do a broad range in clinical program, sensible clinical program based on a series of informed guesses, I guess, as to how do you best utilize each compound.

In your experience as an ex-pharma guy, what would an organization of similar size and productivity, if it resided in pharma cost, spend, whatever for the pharma company relative to what Exelixis is burning through?

You mean how efficient are we relative to a big pharma company?

I mean, what are they kind of overhead costs, etc., what's the fully-burdened costs of having an organization like yours, if it existed within a Pfizer.

I -- actually, I don't know the answer to that question. I I think it's -- I'm not here to be critical of pharma, I worked in pharma for a decade and there are a lot of issues, there are a lot of problems with big pharma companies. They know them, a lot of smart people in the pharma companies sorting -- trying to sort through them. So I don't know.

All I can say is I think for an organization of our size and our maturity and for how long we've been around, we've generated a pipeline that I think we're really genuinely proud of and it's not hype or trying to generate enthusiasm. I think we feel really good about this pipeline in terms of the number of compounds and the quality of those compounds and I think that -- that speaks for itself. We're confident that over time we'll be able to generate the data in humans so that everybody can see that these compounds really are doing what we hope they're doing.

Okay. Thank you for the clarity.

Thank you. Our next question is coming from Bret Holley of CIBC.

Hi. I just had a follow-up question on the attrition rate and what -- standards? You said there's a lot of different standards. And I'm wondering if those standards are developed in collaboration with GlaxoSmithKline and do they have input into the Go decision for 647 and 999? And is there a line in the sand that either you or GlaxoSmithKline is looking for as far as efficacy or safety?

Well, there are a number of criteria which have been developed in collaboration with GSK for the compounds at multiple stages. And yes, they do have input into it and it's been a very collaborative relationship, actually. And there are some very smart and experienced people at GSK and we value their opinion. So, yes, to answer that part of your question, they do have input into it.

The second half. Like most projects, there is no bright line in the sand. There are certainly bright lines that are killing factors for the compounds, but in terms of deciding to move compounds forward -- in a Phase I trial, which is with all comers -- and these are patients for whom no effective therapy exists, which means they failed all available therapies, is a tough population. So you have to look at the data you get with a grain of salt and make the best decision you can. And we make that decision with the people we have here, with our clinical advisory board, and with our colleagues at GSK.

So would it be fair to say based on that comment and based on your experience in major pharma, George, that it would be along the expectations that you would expect for major pharma criteria and for moving compounds forward?

Yes.

Okay. Thanks.

Thank you. [OPERATOR INSTRUCTIONS] Our next question is coming from Jeff Zekauskas of J.P. Morgan.

Hi. Just a few more questions. In terms of the Genoptera revenue recognition, does it all get recognized in the second quarter now of 2005?

Yes, Jeff. It gets recognized in the second quarter of 2005. In Q1, however, you will see that we have booked $10.9 million as deferred revenue in our balance sheet.

Right.

That was for the termination fee. But the revenue recognition doesn't appear -- doesn't occur until Q2.

Okay. And for the GSK collaboration in terms of the milestones. The timing of the cash that Exelixis receives is irrelevant to the accounting treatment? That is, that hitting the milestones is now going to be amortized over the remainder of the contract depending on what contract length you choose; is that correct?

Yes. I mean the most important determining factor is whether you start amortizing as of October '02, ie, the signing date of the original GSK agreement, or as of January '05, the signing date of the amended agreement?

Okay. So it's --

The timing of when we receive the cash really only determines when -- when you start booking your revenue.

Right. So in the second quarter of '05, all things -- all things being equal, there's not this tremendous bump up in revenues relative to say the third quarter of '05 from GSK, you just have to decide what the rate of amortization is according to the period.

That is correct. However, if you were to start calculating as of October '02, the catch up here it is longer and you would have a bigger bump than you would have if you start amortizing as of January '05.

Okay. The third question is for George. George has been talking about the possibilities of other sorts of ventures or other financing vehicles, whatever the term of art is. Can you predict in general the compounds that GSK won't choose, or do you have a sense of what they don't want?

No. And they don't have to make a choice until we get proof of concept Phase II A data. So I think if I were GSK I certainly wouldn't choose now, I would wait until we get the data. So -- but what we do have is an agreement with GSK that we can take the three most advanced compounds, which are the ones of most immediate concern to us, which are 647, 999, and 784, and put them into such a vehicle. And so they can go into that vehicle without knowing beforehand whether or not GSK will choose those particular compounds. And GSK has also agreed to increase the selection milestone for any of those compounds by 25% if -- if they select them.

So say you have good Phase II A data on one of these compounds, when does GSK have to select it if they do want to choose the data -- if they do want to choose the compound that you have good data for?

Well, when we -- the protocol's like this, Jeff, if we agree with GSK on what constitutes a proof of concept program.

Yes.

So what -- what the trials are that would constitute that. We do those trials, when they're done GSK has a period of time -- short -- relatively short period of time to decide whether or not to take the compound. If they take it, there's a milestone and they take it. If they don't, then the rights come back to Exelixis.

So relatively short, means some number of months. Relatively short period to choose?

Yes. Yes. I don't know. I forget what it is. I don't know if we said it publicly, but it's relatively short. Yes. Some small number of months.

Okay. Thank you very much.

Thank you. Our next question is coming from David Witzke of Banc of America.

Thank you for taking my question. Just a quick follow-up to an earlier question. Is it reasonable to conclude based on your confidence in moving 647 and 999 forward, that you have now escalated to dose levels where you're getting blood concentration where you might see a signal and so far so good regarding safety?

I think it's -- want to be careful here because I don't -- I know that we don't want to be commenting on the results of ongoing trials. I think it's safe to say that if we didn't think we were achieving blood levels that were consistent with the therapeutic effect before we saw toxicity, that we would not advance those compounds.

Okay. That's clear. And, as I recall, the IC50 is on 647 for EGFR for something like 10 to 100 fold more potent than darsiva [ph] or RISA risks, are you looking for acne form rations in these patients and can you comment on your expectations there?

Well, the -- the rash is a known side-effect of EGF inhibition. So certainly we would be looking for that rash, yes.

Can you comment on whether you have seen that at this level?

No, I cannot comment.

Great. That's helpful. And just for Frank, sorry if I missed it. Update on cash balance at year-end guidance?

Yes. I'll -- I'll estimate for year-end cash balance is to end the year with at least 100 million.

Okay.

And I would say that's a conservative number.

Thank you.

Thank you. Our next question is coming from Hong Li of SunTrust Robinson.

Just a quick follow-up on the termination payment from Bayer. Have you recognized any part of it in the first quarter?

No. We booked it as deferred revenue on our balance sheet in Q1. We will recognize the revenue in Q2.

Got it. Thank you.

Thank you. Our next question is coming from Jeff Zekauskas of J.P. Morgan.

Hi, this is Karen Buchkovich with Jeff. Can you tell me about the half life of 880, 820, or 844? And also is 844 early bio available?

I cannot tell you anything about the half lives of those compounds in humans yet. And, yes, in -- certainly in the preclinical models, 844 isn't readily bio available.

What about the half life in animal models?

You want to know the half life of each of those compounds in the various species where we've tested? I don't think I can tell you that off the top of my head, Karen, but we certainly have the data, we can get it for you if you want.

Does the Chk1 and Chk2 inhibitor inhibit other kinases at low animal levels?

No. It seems to be relatively specific for those two.

For the Chk inhibitors, do you envision combination trials with DNA damaging agents?

Yes. Inhibit both single agent and combination trials.

Which will you start with?

Too early to tell yet. We're just -- we are just finalizing those plans.

Okay. And are you testing a large number of -- of second agents, whether it's radiation or chemotherapy?

You mean for 844?

Yes.

I think initially you want to know that 844 can -- is orally available in humans, it has good bio availability, it has good kinetics. And so there's a lot we need to learn about the compound by itself. And, of course, then we'll test it in combination with a number of different DNA damaging agents.

Okay. With 784, you've spoken about how you've moved to the capsule formulation, how you developed a new formulation. Are there -- are there any new toxicities? How often will you be dosing per day to maintain desired levels of the drug?

Well, I think that's -- that's -- let me answer the first part of your question. Are there any new toxicities as a result of the new formulation? No. Are we -- how will we dose this once a day or twice a day? Don't know yet. 784 is about an eight-hour half life in human plasma. So we will do both once a day and twice a day studies initially and we'll see. Once we choose one.

Okay. Thank you.

Thank you. There are no further questions at this time. I now turn the floor back over to Charles Butler for any closing remarks.

Okay. Well, I look -- really appreciate everybody's attention. I think there's some interest by the number of questions. I hope we've answered them and we'll all get back to work now. So thanks again, everybody, for your attention and your interest .

Thank you. This does conclude today's teleconference. Please disconnect all lines and have a great day.