英賽德 (INCY) 2007 Q2 法說會逐字稿

Greeting, ladies and gentlemen, and welcome to the Incyte Corporation second quarter 2007 financial results. (OPERATOR INSTRUCTIONS). As a reminder this conference is being recorded. It is now my pleasure to introduce your host, Ms. Pamela Murphy, VP, Investor Relations and Corporate Communications for Incyte Corporation. Thank you, Ms. Murphy. You may now begin.

Thank you. Good morning. Thanks for joining us. With me today are Paul Friedman, Incyte's President and Chief Executive Officer; Dave Hastings, Executive Vice President and Chief Financial Officer. Steve Friedman, Senior Vice President of Discovery, is also with us. We'll begin with an update from Paul on progress in our clinical program, and Dave will provide a brief overview of Incyte's second-quarter financial results. Then we'll open up the call for Q&A.

Before beginning, I'd like to remind you that some of the statements made during the call today, including statements regarding our plans and expectations for our drug discovery and development programs, and our financial results and guidance, as well as the potential efficacy of our compounds, are forward-looking statements. These statements are subject to a number of risks and uncertainties that may cause our actual results to differ materially, including those described in our most recent 10-Q for the period ended March 31, 2007, and from time to time in our SEC documents.

Paul?

Thanks, Pam. Good morning, everybody. The last four months have been really incredibly productive for us. I'm going to take you through my reason for saying that. We've filed multiple INDs, we've reported positive Phase IIa data for our CCR5 antagonist, and we've initiated a number of proof-of-concept trials. We're already (technical difficulty) very encouraging clinical data, albeit it in a limited number of patients, in several of these proof-of-concept trials, including trials with our sheddase inhibitor for HER-2 positive breast cancer and other solid tumors, and our newly announced JAK inhibitor program for inflammation in certain cancers.

I'm going to begin with the CCR5 program. As most of you saw, we presented the results from a Phase IIa 14-day placebo-controlled trial for Incyte 9471 at the IAS meeting in Sydney earlier this week. The study included 23 patients; 19 on the drug, four on placebo. Of the treated patients, 10 were treatment-naive; nine were treatment experienced but not currently on antiviral therapies. All the treated patients in this cohort received 200 mg of 9471 once a day. The results showed impressive reductions in the viral load that were sustained well beyond the 14-day dosing interval. Patients receiving 9471 achieved a 1.7 log mean viral load drop at day 14, and a nadir in mean viral load decline of 1.8 was seen at day 60. Importantly, at day 28, two weeks after their last dose, we continued to observe about a log drop in viral load, a result consistent with the compound's long half-life.

Importantly, in addition to being highly efficacious, the drug was extremely well-tolerated. This week's presentation at IAS of the maraviroc results in naive patients reinforces my belief that 9471 could well be the best-in-class CCR5 antagonist. 9471, unlike maraviroc, can be dosed once daily with or without ritonivir, making it an attractive option for patients in whom R5 tropic virus predominates, and offering the potential for incorporation into fixed-dose (inaudible) combinations.

Additionally, because of the better PK profile and longer half-life of 60 hours, 9471 therapy should be significantly more forgiving during lapses in compliance, such as missing a dose of 9471, or of ritonivir in ritonivir-containing regimens.

We also believe, based on emerging clinical data, that 9471 may well be more effective in maintaining target therapeutic concentrations throughout the dosing interval. Specifically, the data that has been presented on maraviroc suggests that somewhat less than three-quarters of patients on ritonivir-boosted regimens achieve target trough levels, while less than a quarter of patients on non-boosted regimens do. Based on the clinical data we've generated to date, it appears that 9471 has real potential to significantly improve on the percentage of patients who achieve target trough concentrations.

This, coupled with the impressive and sustained viral load reductions and excellent tolerability we've seen with 9471, plus the maraviroc results, which overall look pretty good, provides us with a terrific opportunity. And I'm sure we'll get into this more in the Q&A session.

To support the Phase IIb program, we're currently evaluating two additional doses, 100 and 300 mg once daily, and conducting required drug interaction studies. As we said in the press release, we just met with the FDA regarding the Phase IIb program. While we believe we have reached agreement with them on the overall design of the trial, the Agency has asked that we provide them with the results of our dose ranging and drug interaction trials before they give a final endorsement of our planned IIb studies. And that's to be expected; no surprise there. They've asked that we focus our IIb program on treatment-experienced HIV patients. The six-month safety and efficacy data that we would generate in the treatment-experienced Phase IIb population would then, if positive, support initiation of pivotal Phase III studies in both treatment-experienced and treatment-naive patients.

We plan to begin our first Phase IIb trial in the first quarter of 2008. It will involve about 150 patients and would include two ritonivir-boosted doses of 9471 and an optimized background regimen placebo arm. The second trial should begin about three months later and will involve about 75 patients with two unboosted doses of 9471 and an optimized background regimen placebo arm. If we enroll the studies as we expect and the trials are positive, the overall timeline of the program will not change, as we remain on track to begin our Phase III program in the second half of 2009.

Moving to our lead HSD1 inhibitor, Incyte 13739, we are currently enrolling type 2 diabetic patients in a one-month Phase IIa trial assessing the effect of the compound on insulin sensitivity. The trial, which includes three investigative sites and is expected to enroll 24 evaluable patients, 16 on drug and eight on placebo, involves a two-step insulin clamp that is conducted under tightly controlled metabolic conditions, and, when done properly, is expected to provide a direct measure of the drug's effect, both on hepatic insulin sensitivity as measured by glucose production, as well as on peripheral insulin sensitivity as measured by glucose uptake.

For two of the three investigative sites, the technical execution of the clamp turned out to be very challenging, and retraining was required. And that has set us back about eight weeks. These sites now appear to be executing the procedure properly. Despite these technical issues, which made the first set of clamps uninterpretable, we still expect to have topline data sometime in the third quarter, and we remain on track to initiate the three-month dose ranging Phase IIb study early in 2008.

The planned three-month study will have a primary endpoint of hemoglobin A1c lowering, and will also allow us to characterize the effect of 13739 on other cardiometabolic risk parameters, such as blood lipids and blood pressure. As we do in all of our programs, we've advanced a follow-on compound, Incyte 20817, for which we filed the IND earlier this month.

Now, for Incyte 7839, our sheddase inhibitor, currently in a Phase Ib/IIa dose escalation trial in patients with refractory solid tumors, we have seen highly significant reductions in clinically relevant biomarkers; in particular, circulating levels of both HER-2 extracellular domain, the ECD, and in the EGFR, or HER, ligands.

ECD is produced by sheddase activity, and elevated plasma ECD levels, which are seen in about half of HER-2 positive metastatic breast cancer patients, are associated with a distinctly poor clinical prognosis. Sheddase activity is also responsible for the release of the active forms of the EGFR ligands, that is HER ligands, which induce the dimerization of HER monomers, and that's required for signaling in these pathways. We've seen significant reductions in the circulating levels of the EGFR ligands TGFalpha, heregulin and amphiregulin, suggesting that 7839 is indeed, as was predicted, inhibiting signaling through all four EGF family receptors.

As we hope to present data at the San Antonio Breast Cancer Conference in December, I'm not going to provide you with all of the clinical response data, but I can say that we're beginning to see positive clinical responses. Among the evaluable patients studied, results in HER-2 positive breast patients have been particularly encouraging, with the majority achieving stable disease.

Additionally, we've now reached the maximum tolerated dose, or MTD, and the dose-limiting toxicity which we've seen in a few patients is deep vein thrombosis, DVT, which we believe can be managed by employing an interrupted dosing schedule and/or unmonitored use of a low-dose anticoagulant. We have amended the current Phase Ib/IIa protocol to provide for enrollment of additional patients to test this strategy. In addition to reducing the risk of DVTs, this approach will allow us to explore higher doses while we move forward in parallel with initiation of two Phase II studies in breast cancer patients beginning in the fourth quarter. In these studies, which will be conducted outside of the U.S., we plan to evaluate 7839 in HER-2 positive metastatic breast cancer patients in combination with Herceptin, and also as monotherapy.

For our janus-associated kinase, or JAK inhibitor program, we've initiated a series of clinical trials for the lead compound Incyte 18424, including, but not necessarily limited to, a Phase I trial in patients with a myeloproliferative disorder, or MPD, and a Phase I trial in healthy volunteers to support the development of 18424 as a potential treatment for rheumatoid arthritis and psoriasis. We've obtained very encouraging pharmacodynamic data in blood samples taken from healthy volunteers, as well as MPD patients, after dosing with 18424, showing marked inhibition of the JAK pathway.

Given the growing body of impressive positive clinical data suggesting that the JAKs are critical components of signaling mechanisms in a number of chronic inflammatory diseases and myeloproliferative diseases in certain cancers, the PD data should be predictive of positive clinical outcomes.

As I've said before, we're on track to report proof-of-concept results for two or more of these indications before the end of the year. And I say this with some confidence, as we've already seen early and impressive improvements in clinical signs and symptoms in the limited number of MPD patients treated so far. There's a very real possibility that we will have positive proof-of-concept studies for several other indications, either before year-end or very early in 2008.

While the JAK inhibitor space is competitive, for us it's an important area in which we have a very strong program with potent, selective orally bioavailable compounds for multiple distinct chemical scaffolds. We believe we're the first to study a JAK inhibitor in MPD patients, we already have a number of Phase I and Phase Ib/IIa trials underway, and we have a second compound progressing through IND-enabling studies. As this compound -- as this program is one we can develop and commercialize on our own, and is one which aligns quite well with our understanding of the underlying biology of oncology and inflammation, we're very unlikely to partner.

I'll end with a brief summary of our CCR2 program. As you saw in the press release, Pfizer has advanced the compound into a Phase I trial. This compound is one of many from a broad chemical platform, giving Pfizer the ability to choose the optimal compound for pursuit in a particular indication. This compound's advancement into the clinic exemplifies that ability. Pfizer is actively exploring CCR2 antagonism in a wide range of indications, and is evaluating a number of different compounds, including Incyte 3284, the most advanced compound at the time of the deal. That said, at this time we would expect that our next milestone from Pfizer would probably be for another IND filing. For our own compound, Incyte 8696, we expect to begin the Phase I trial in healthy volunteers early in the fourth quarter.

Now I will turn the call over to Dave Hastings.

Thank you, Paul. Good morning, everybody. I'll start my brief overview this morning by discussing our cash position.

We ended the second quarter with 290 million in cash and investments. Our cash used so far this year has been 43 million. This excludes the $3 million milestone payment received from Pfizer during the second quarter. Cash guidance for the year remains a use of 88 to 95 million, which excludes the in-license or purchase of products and any milestones received from the Pfizer collaboration.

Another area I'd like to discuss this morning is revenue. As I mentioned, we received a $3 million milestone payment from Pfizer. This amount has been recorded in the second quarter, and therefore, we are increasing our 2007 guidance from a range of 22 to 25 million to a range of 29 to 31 million. In addition to revenue, we are also increasing our 2007 interest income guidance from a range of 11 to 12 million to a range of 12 to 14 million. This is a result of slightly higher yields from our investment portfolio.

In terms of our operating expenses, our performance remains on plan. We remain focused on investing in our pipeline. And despite our aggressive clinical development plan, we are still on track to incur between 88 and 95 million of R&D expense for 2007. This is unchanged from our previous guidance.

And with that, I will now turn the call back over to Paul.

Operator, can we please open the call up now for Q&A?

(OPERATOR INSTRUCTIONS). Richard Smith, JPMorgan Chase.

Just to clarify, on the JAK 2 program, you mentioned proof-of-concept data; two indications later this year. Of those two indications, you mean, in MPD or --?

No. Different indications.

So could one of them be MPD and another one in inflammatory, another -- are you suggesting rheumatoid or psoriasis or something?

That would be a logical conclusion. We haven't identified those yet, and -- those would be logical places to go. But the ones -- the additional ones I'm talking about would be in addition to what we're currently doing, and would be in addition to the myeloproliferative disease studies that we have embarked on.

And your understanding of the JAK 2 target, do you think you need a different degree of inhibition in the oncology setting versus in the inflammatory setting?

That's a difficult question to answer at this point, but it certainly is one that should be asked. We don't know. What I would say is that -- myeloproliferative diseases, in my opinion, kind of bridge the gap from a benign disease to a malignant disease. There's kind of a continuum there. And some end up with acute leukemia. Some end up with CML. So what we've seen so far is that at doses that we would predict are in the same range as doses that would work in inflammatory disease, we're seeing some very impressive clinical response, improvements in signs and symptoms. Whether some of the other frank malignancies that we have considered looking at with JAK inhibitors -- whether or not those would require higher doses, we just don't know, because we haven't gone there yet.

Are you seeing any dose -- any toxicity side effects, etcetera?

The drug has been actually extremely well-tolerated so far.

John Watkins, Banc of America Securities.

A couple of quick questions around the sheddase inhibitor program and the toxicity. Have you seen activity at lower doses, or was it (inaudible) higher doses where you also saw the deep vein thrombosis as well?

We've seen activity at the maximum tolerated dose, which is one dose below where we define dose-limiting toxicity.

And you mentioned that you've amended the protocol; should we assume that you have been in discussions with the FDA and they're okay with that, and okay with the trial continuing and looking at higher doses as well?

Yes.

Thomas Wei, Piper Jaffray.

I had a question on the sheddase inhibitor. First, just to clarify, can you remind us -- I think you had said before that you had seen actual objective responses in the sheddase inhibitor. Or did you just -- did you mean the stable disease in HER-2 positive breast cancer patients?

We've seen stable disease in between 20 and 25% of the patients we've treated, and in 80% of people with HER-2 positive breast cancer. We had one individual who had definite tumor regression who was HER-2 positive, and a shedder, but who did not quite make the criteria of a partial response.

That's very helpful.

The data look quite similar to the early Ib/IIa results that we're seeing with Tykerb.

On 9471, I know you had talked before about the problems, or the conundrum, of potentially having a lot of new agents approved and available for use in the optimized background regimen, and the potential hurdles in showing activity on top of a strong background. How did you end up working that out with the FDA?

We haven't -- until we actually have final approval from the FDA for the IIb studies, I would rather not get into that. But we have what we believe are good Phase IIb studies with primary endpoints being the usual primary endpoints that you would look for in such studies -- that is less than 50 copies, and multiple secondary endpoints like less than 400, greater than a log, greater than 1.5, greater than 2 logs -- those kinds of parameters. The primary objective of the Phase IIb program is going to be to select boosted and unboosted doses for Phase III, and to get longer-term safety data. And until we actually have a lock from the FDA on the program, I'd rather not publicly be talking about the trials.

Sapna Srivastava, Morgan Stanley.

It's actually Dave calling in for Sapna. I just have a question regarding the toxicity that you've seen with the sheddase. What was the -- sort of some color around what the DVTs were like? Were they sort of just one-off, or was it extensive? Did anyone have pulmonary embolism? And exactly what type of anticoagulation do you think you'll use? Is it going to be coumadin or aspirin, or (inaudible)?

The DVTs are in the location you would expect. They're down in the calf veins. Because we had seen one earlier in the program, we do Dopplers, and they were asymptomatic. And we also follow -- have been following a marker of fibrin being laid down, which is called D-dimer, which you can measure. They're not extensive, because you can -- if you follow them by Doppler, you pick them up before they get extensive, or before they end up going into the pulmonary circulation.

What we found is that either by a few days off the drug, or in those patients who entered the study -- because, as you know, people with extensive tumor burden are at increased risk of venous clotting anyway -- a number of these patients come in either on low-dose aspirin or low-dose coumadin. We found that in the patients who are on low-dose aspirin, their D-dimer levels don't go up; in patients who are on low-dose coumadin, they are significantly blunted, and none of those people had DVTs. .

In addition, we have developed an animal model. We don't completely understand the mechanism yet for the DVT increase. But the rodent model does show increased laying down of [clod] when you give a sheddase inhibitor. We haven't used 7839, but we've used a related compound. And you can effectively treat in the rodent with aspirin, and I'm quite sure you'll be able to treat with coumadin as well. We haven't completed that experiment yet. So we have multiple relatively simple ways to deal with a reversible side effect. So to me, as opposed to seeing something like pulmonary fibrosis, cardiac failure, or other -- or hepatic failure that you can see as maximum tolerated doses with other antitumor agents, this is a pretty good

Sure. Just one quick follow-up. Are people in studies going forward with this program going to need to get regular lower-extremity Dopplers, and labs, and INRs and whatnot?

You won't need INRs. With either low-dose aspirin or low-dose coumadin, you don't have to monitor.

How low-dose? 1 mg a day or something?

2, maybe. But that's the range. 1 to 2, or 80 mg of aspirin. I would think -- for the foreseeable future, in the programs that we would do Dopplers long-term, if we demonstrate that we can control the incidence of DVT, I think we would dispense with it. But for the next studies, I think, it makes good sense to have that incorporated into the study design.

One quick question on the CCR5 program. Do you see any read-through to your program with what the Pfizer program has been sort of stalled at the FDA a little bit? Do you think this is a class thing, or do you think there is something specific to their application and program that is causing their delay?

I think one of the read-throughs is it probably has had a depressive effect on our stock price. And I think, hopefully, when they get approval, which they will, we'll get a bounce back from that. My understanding from the rather bullish press releases that they have put out, saying they're going to get approval, they don't need to do anymore studies, is that they're in negotiations with the FDA about labeling and about Phase IV commitments. And I don't know how long it will take them to get approval for their treatment-experienced indication.

So I actually -- also, although we have to study the IAS presentation in a little more detail, I think that the results that they have seen gives us a really huge opportunity to be best-in-class, and to demonstrate that the compound can actually unseat efavirenz as the preferred non-nuke used in naive populations. If you look at the results that they got, and you start to tease away what happened, they barely missed non-inferiority at the 50 viral copies per ml.

If you look at the data from the big markets, Europe, North America and Canada, it actually looked a little better than efavirenz. If you look at the "failures", the preponderance of them appear to be tropism issues. And if, as we fully expect by the time we do our pivotal that we anticipate doing in naives, there is available a more sensitive assay to weed that group out to start with, we would expect that we would be able to unseat efavirenz. And I'm partial to efavirenz. But I think there's a huge opportunity here in the naive patient population, in addition to the treatment-experienced population.

Annabel Samimy, UBS.

(inaudible) quick questions. The activity you're seeing on the JAK 2 in myeloproliferative disorders is interesting. Can you give us a little bit more color on some of the symptoms (inaudible)?

What we've seen is a dramatic decrease, beginning within one week, in spleen size. These people have very large spleens. We've seen within one week improvements in the smears, the hematologic smears, in several of these people. These people -- we've had one individual who had orthopnea, which means he couldn't lie flat. He within one week was able to lie flat. We've seen a reported -- a sense of well-being that they haven't enjoyed in quite a while. And they have what appears to be dramatically increased energy level. One person who was pretty much incapacitated, the guy with the orthopnea, has gone back to playing golf on a regular basis, etcetera, etcetera. So what we're seeing are some very dramatic changes in these people. And it's early days. It's not like we've seen dozens of them, but we've seen enough that these are real, and are almost certainly related to administration of 18424.

The inhibition of JAK 2, have you -- I mean, the big concern was that this (inaudible) inhibition of JAK 2 would cause some hematological toxicity. Have you seen any of that? And do you have a sense of how much inhibition you're actually causing?

We do know how much inhibition we're causing, and I think we'd like to get a bit more data on that before I give you a hard number. But we are seeing these dramatic effects without taking out JAK 2 during the dosing interval, through the entire time. And therefore, there's a reasonable amount of off time from JAK inhibition. I guess one of these people is into their second month; the other is about to go into second month. So it's a little early to be able to predict what you would see vis-a-vis the hematocrit or hemoglobin. I will say that looking at the retic counts, they seem to be pretty steady. It's not like we've seen dramatic decreases in reticulocyte counts, which would be predictive of shutting down red blood cell production.

A question about the CCR5. Can you help us understand a little bit more about the metabolism of CCR5 versus ritonivir? And I guess I'm curious. You have stated several times that CCR5 does not need ritonivir boosting, and you are moving into some ritonivir boosting studies, obviously, for background therapy. But can you sort of talk about that a little bit?

Because many people have ritonivir in their regimens, as you alluded to, if you're going to go into the treatment-experienced population, you're going to have to study your compound on that background. But there is movement even in treatment-experienced patients to try to find regimens that don't have ritonivir, and you certainly don't use ritonivir in the overwhelming majority of first regimens. So the point we are making is that in regimens that don't have ritonivir, we are still easily once a day because of the long half-life, which distinguishes us from where maraviroc currently is and is likely to be able to go with their half-life and the issue that they have with blood level-related orthostatic hypotension.

In terms of the metabolism (inaudible) metabolized through the same pathway, could that affect CCR5 in any way?

Yes. They're all -- vicriviroc, maraviroc and our drug are CYP3A4 metabolized. So you will have to dosage us down in a background regimen that has ritonivir. So the dose will not be the same in a ritonivir-boosted regimen as in a ritonivir-free regimen, as is the case with maraviroc.

Eun Yang, Jefferies & Company.

This is Molly in for Eun today. Two quick questions on 9471. Have you guys seen any additional side effects coming out of the 300 mg dose that you're testing?

Well, we haven't finished that cohort yet. But up to this point, the answer is no.

Good. You were discussing later on, potentially, 9471 being preferred non-nuke. Do you guys think there might be a need to compare maybe non-inferiority to (inaudible) maraviroc, or do you go straight after efavirenz?

We've talked about that. I think, based on just reading between the lines and thinking about the maraviroc presentation, I think we can beat efavirenz. I'm saying I think that. I think it would be -- I think a -- I think you would want a once-a-day regimen -- to compare to a once-a-day regimen -- I think it would be a little bit difficult to blind such a study with maraviroc, which is going to be [BID]. I also think the likely nukes that you would use would be Truvada and not Combivir. So, most likely, if the FDA concurs, when we get to that point, we would probably design a study to compare ourselves to an efavirenz-containing regimen.

Thomas Wei, Piper Jaffray.

Just one follow-up here. A more general question on looking at the various pieces of biomarker data that you've now got for the HSD1 and the sheddase and the JAK 2 program, in the context of the various indications that you're exploring, how should we think about how you feel if you had to rank order probability of success, or just your excitement about the various programs now?

I think that's a fair question. I very much like the JAK program. I've said that publicly a number of times. I think for us, what Pfizer has already reported on their modestly JAK 3 selective compound, showing that they get good activity in RA and in psoriasis, with what we know about the JAK 2 activating mutations in MPD, and what we are finding is interesting possibilities in some other tumors that we haven't discussed, we think that it's a size program that we can handle. It's in our area of expertise, and there's a very high probability that we're going to get positive proof-of-concept in most of the things we look at.

I think the MPD stuff that we already have is exciting. It's analogous, in my opinion, to -- from what we've seen so far, to CML Gleevec. And for that reason, I really, really like that program. It's hard not to like (inaudible), but I like that one a lot. Sheddase has been kind of a sleeper for us. But the data that we have, and the fact that the dose-limiting toxicity is one that is imminently controllable, and the really compelling hypothesis and preclinical data that we have indicating marked synergy with Herceptin, gives me a high degree of confidence when we get into that study that we're going to see something quite interesting. And we should see it relatively early in 2008. So I like that program as well.

And then CCR5, we have proof-of-concept, and we have a path now to best-in-class. And you can't not like that program. HSD, if we get positive results, would be fascinating and huge. It would probably be a program that we would have to find a partner for. And CCR2, it's too soon to know. So that's kind of the way I would stack them up.

That's helpful. Just on the JAK 2 program, what is the official endpoint that you're looking at here in the MPD setting that you're going to make the proof-of-concept go/no-go decision on?

We're looking at a variety of things. The organomegaly is a pretty objective endpoint. And we would also like to see -- see, what happens in this population is they -- especially the ones who are farther along in the progression, people who have significant myelofibrosis, they begin to have extramedullary hematopoiesis, which means that their marrow has fibrosis, and so they start to make their peripheral blood cells in liver and spleen. And so, if you can get rid of the bad cells, the ones that have the V617F activating mutation, and tamp that down and allow normal marrow cells to begin to repopulate the bone marrow, and you reverse what you should be able to do, the fibrosis -- all of that goes hand in hand -- but the most objective response you would see early on would be a reversal of the organomegaly. We are seeing that quite dramatically.

I'll give you an example. We had the -- the first patient we treated had a spleen that was measurable 22 sonometers below the bottom of the rib cage. That's a big spleen. That's almost down to the pelvic bone. Within one week, his spleen was 15 sonometers in size, and has now -- I think the last time it was measured, was in the 10 to 11 sonometer size range. We are seeing a very dramatic decrease in organomegaly, and it's progressively getting better. So that is -- that's a major endpoint. There are lots of things we're looking at in the patients. We have to -- but that one is quite objective, and not as subjective, for example, as somebody feeling more energy.

(OPERATOR INSTRUCTIONS). Richard Smith, JPMorgan Chase.

Just briefly on the HSD1 program, can you just outline what you mean by positive trends, what you're looking for there?

What we would like to see -- so let me -- so, when people do clamp studies, what they usually do is a very high level infusion of insulin. And they're looking for evidence of peripheral uptake, which is mainly into muscle. That's done in many centers quite commonly. The twist in the study that we're going is that we're doing an initial insulin clamp with a low amount of insulin being infused to try to isolate first the liver's -- the amount of glucose the liver makes. So if you have improved insulin sensitivity with your drug, you should see, for a given amount of insulin, more of a shutdown of the liver's production of glucose. That's a trickier thing to do, and it makes this technique dicier than if you were just doing a routine single-insulin clamp.

So what we're trying to see are trends in either what's called the EGP, which is a measure of liver glucose production at zero time versus 28 days after taking the drug, so the patient can serve as their own control. And additionally, we're doing some placebo patients. And then we go to the high insulin clamp, and we're looking for evidence of an increase in peripheral uptake. So with this small [N], it's unlikely we're going to get statistically rigorous P value from the study. So we're looking for compelling trends either in a decrease in liver glucose production with insulin stimulation, which would be a measure of increased insulin sensitivity at the hepatic level, and an increase in the uptake of glucose at the high insulin infusion, which would be a measure of insulin sensitization in the periphery.

So you think you'd probably see a more significant effect on the liver effect than the (inaudible) peripheral effect?

That's what was seen with Actos in the one published study that [Meredith Hawkins] did. You saw more effect and a higher percentage change in the hepatic insulin sensitization than she was able to show for peripheral uptake after a three-week, four-week type study, which is what we're doing. This is not an easy study. And if you recall, we debated whether to do a one-month study as we awaited our three-month animal tox data, which then allows us to more do the more formal Phase IIb hemoglobin A1c reduction. And we decided we'd take a crack at it, and (technical difficulty) had a rocky first two months, trying to get the sites to do the total clamp properly. And right now we -- they seem to be doing it properly, and we have our fingers crossed that we'll see something. I think, by the UBS meeting, end of September, we should have some relevant data from this study.

I actually should also add one more thing. The question hasn't been asked, but I debated whether to present data that we have, early data, in a separate program with JAK, where -- in another disease entity. We're actually seeing something quite interesting there as well. I'm going to -- we elected to hold that off, again, until probably the UBS conference, where I can introduce it in a more complete way. This is not an ideal forum to introduce something that you haven't heard about before. But there is another patient study that we're doing, in a different indication than MPD, where the early results look very promising.

Eun Yang, Jefferies & Company.

This is Molly again. Just a quick question on the sheddase inhibitor. Would you expect any impact on your trial design for potential approval of Tykerb in the EU, or, in the future, your trials in the U.S.?

It will not have any effect on the studies that we're planning to do because we're not doing them in the EU. We actually have done combination studies with Tykerb in collaboration with GSK people. We've seen -- this is preclinical. We've seen some very interesting synergy there. And I don't -- I can't speak for them at this point. But in future studies, combinations of Tykerb with the sheddase inhibitor actually would be pretty compelling types of studies to consider.

There are no further questions at this time. I would like to turn the floor back over to Dr. Paul Friedman for closing comments.

Thank you. Thanks, everybody, for those interesting questions and for listening in today. As we have said to you, this is shaping up to be an especially productive year for us. The CCR5 antagonist program is moving ahead as planned, and 9471 certainly has the potential to be the best of that class. In addition, we're already seeing encouraging early clinical data with our sheddase inhibitor and in our JAK programs. So, to use the usual analogy, we have multiple shots on goal here. And so far, the shots appear to be on target. I'm looking forward to sharing full results from these trials, as well as the ongoing two-step insulin clamp study, with you over the next several months. With that, I would like to end the call and thank you all again for tuning in.

This does conclude today's teleconference. You may disconnect your lines at this time. Thank you for your participation.