ImmunoGen Inc (IMGN) 2008 Q3 法說會逐字稿

Good day, and welcome, everyone, to this ImmunoGen Third Quarter Fiscal Year 2008 Conference Call. Today's call is being recorded.

At this time, for opening remarks and introductions, I'd like to turn the call over to the Executive Director, Investor Relations and Corporate Communications, Carol Hausner. Please go ahead.

Thank you. Good afternoon, everyone, and thank you for joining us on our quarterly conference call. At four o'clock this afternoon, we issued a press release that summarizes our financial results for the third quarter of our 2008 fiscal year, the quarter ending March 31, 2008. I hope you've all had a chance to review it. If not, it's available on our website at ImmunoGen.com.

During today's call, we will make forward-looking statements. Our actual results may differ materially from the projections made. Descriptions of the risks and uncertainties associated with an investment in ImmunoGen are included in our SEC filings, which also can be accessed through our website.

With me today are Mitch Sayare, our Chairman and CEO, and Dan Junius, our Executive Vice President and CFO. Mitch will provide an update on ImmunoGen, and Dan will discuss our financial results. We'll then open the call to questions.

I'll now turn the call over to Mitch.

Thanks, Carol. In the past three months, we've made solid progress with our IMGN901 and IMGN242 clinical stage compounds, and our Genentech collaborators have reported that they plan to make a Phase III go/no go decision with trastuzumab-DM1 this year. Our partner, sanofi-aventis, is aggressively developing the three compounds that are already in the clinic, through their collaboration with us, and additional ones behind these have passed internal milestones. And we remain on track for three more TAP compounds to enter the clinic this summer, through our own programs and through those of our partners, Biogen/IDEC and Biotest.

Let me take you through all of these now, starting with trastuzumab-DM1, or T-DM1. As you probably know, this compound consists of Genentech's trastuzumab, or Herceptin, antibody, our DM-1 cell-killing agent, and our thioether linker. In their recent quarterly conference call, Genentech disclosed that they plan to make the Phase III go/no go decision for T-DM1 in 2008. They also took care to point out they don't need to wait until data from the ongoing Phase II study have been reported for them to make this decision.

We're thrilled that Genentech's planning to make this decision this year. As you recall, they only started the Phase II last July. Then Roche opted in December 2007, and both companies prominently discussed the compound in this year's annual reports.

One question we've been asked is whether we get royalties on the T-DM1 sales made by Roche. Just to be sure there's no confusion on this, we do, and on the same terms as if sales were made by Genentech. Of course, Roche brings to the table their global strength in product development and sales. To put in this context, 2007 worldwide sales of Herceptin were CHF 4.85 billion, more than $4.5 billion, inclusive of the $1.3 billion sales made by Genentech.

In their quarterly conference call last month, Genentech also reported that they'll be presenting Phase I data with T-DM1 at the ASCO meeting that's coming up later this month, so lots going on with this compound.

Turning now to our wholly-owned IMGN901, this compound is currently being evaluated in three clinical trials, two in solid tumors and one in multiple myeloma. Our focus has been on the development of the product for multiple myeloma, since this has the potential to be the fastest route to regulatory approval. The myeloma trial we have underway, our 003 study, evaluates IMGN901 when used as monotherapy in patients whose cancer has failed to respond to multiple prior therapies, a very tough patient population. We've reported objective responses in this study and are still dose-escalating. I know that I've promised to provide you with our development plans for this compound, and want to update you on where we stand.

A few weeks ago, we convened a panel of experts in the treatment of multiple myeloma, leading clinicians in the field. We presented data from our 003 study to date, as well as information from our pre-clinical studies, assessing the compound when used in combination with approved therapies--Revlimid, Velcade, and so on. They were impressed with the monotherapy activity that's been seen in this heavily pre-treated patients and encouraged us to continue to assess IMGN901 in this population. There's a real need for new multiple myeloma therapies for patients who failed the available ones, and consequently, seeing convincing activity with IMGN901 in such patients could enable us to proceed directly to a registration study.

The experts were also very, very impressed with the tolerability profile of IMGN901 and that its mechanism of action is different than that of approved agents. As a result, they encouraged us to evaluate IMGN901 when used in combination with Revlimid or Velcade. We know from our pre-clinical studies, data we reported at the recent AACR meeting, that IMGN901 shows impressive synergy when used in combination with Revlimid, and to a somewhat lesser extent when combined with Velcade. The combination regulatory pathway would be somewhat slower for us to pursue than a monotherapy route, since we'd need to evaluate the combination in a brief Phase I study before we could proceed to Phase II and beyond.

Thus, to advance IMGN901 quickly, as quickly as possible, we're pursuing two parallel tracks. First, we're continuing our 003 monotherapy study to establish the maximum tolerated dose in this patient population. Once that's established, we'll treat additional patients at MTD, and see whether the compound produces compelling activity as monotherapy in heavily pre-treated patients, and will enable us to assess whether a monotherapy registration trial with IMGN901 in this patient population makes sense. We should know this later this year, probably around the time of ASH. We also expect to report study results at ASH.

For the second track, we plan to start a Phase I trial that evaluates IMGN901 when used in combination with an approved agent--Velcade, or more likely, Revlimid--so we can move quickly into a Phase II trial if the combination route makes sense. The Phase I study will be in an earlier stage patient population, and we expect to have it underway by the fourth quarter of the calendar year.

We'll also continue our current solid tumor studies, looking for additional pathways to approval. For example, a number of patients with Merkel cell carcinoma are lining up for enrollment in our 002 study. You may recall that we obtain a durable CR in a Merkel cell patient in this study. Depending on the findings in these patients, we may pursue Merkel cell carcinoma as another route to approval for IMGN901, since it's a disease with very limited treatment options.

I also want to update you on the clinical evaluation of our IMGN242 compound, presently in Phase II testing for the treatment of gastric cancer. We're taking a highly disciplined approach to its evaluation. That is, we'll need to see an objective activity with IMGN242, a solid PR or better, by resist criteria, in at least one of the first 23 patients, for us to expand this study and continue our development of the compound. If we see activity, then we'll expand to 40 evaluable patients. This will provide sufficient statistical power for us to determine whether to continue development of IMGN242 as monotherapy or not.

Of course, clinical studies are just that--they're studies. And thus, they often yield information that helps in the development of the drug and suggests where adjustments should be made along the way. After we began enrolling patients in the study, we observed that the dose being tested--168 mgs per meter squared--can be too high for patients who have low levels of soluble target antigen circulating in their blood. In a few of these patients, we encountered the same eye toxicity, a loss of visual acuity, that we reported in the Phase I trial with the compound. Consequently, we amended the study protocol so that patients with low levels of soluble antigen be dosed one step lower than those with high soluble antigen levels--that is, at a 126 mgs per meter squared rather than 168. The reduced dose is about 3.4 mgs per kg, in the same ballpark as the 3.6 mgs per kg that Genentech uses in their Phase II study with T-DM1. Patients with high levels of soluble antigen continue to receive the 168 mgs per meter squared dose, or about 4.5 mgs per kg. Because of this change, we're not quite as far along in patient enrollment as we would have liked, but feel comfortable that we'll complete the enrollment of the 23 patients this year.

As we've noted before, if we see an objective response in these 23 patients, we'll announce it when it's confirmed. If the compound isn't as active as we think it is, we'll only know that after we've dosed all 23 patients. So if we see no response, we'll let you know that, too, since we'll stop the study and discontinue developments of the compound.

We'll report clinical data on IMGN242 at the upcoming ASCO meeting. This poster will principally focus on pharmacokinetic findings, and may include some information from our Phase II study.

Sanofi-aventis is continuing to make solid progress with the three compounds already in the clinic through our collaboration--the TAP compounds, AVE9633 and SAR3419, and the naked antibody, AVE1642.

Two more compounds have recently passed pre-clinical milestones--the TAP compound SAR566658 for the treatment of solid tumors, particularly breast and ovarian cancers, and the naked antibody SAR650984 for hemotologic malignancies. In fact, about half of sanofi-aventis's entire pre-clinical and Phase I oncology portfolio are compounds they developed in collaboration with us. 566658 and 650984 join a growing body of compounds that are advancing toward the clinic through our own programs and through those of our partners. In fact, we expect three TAP compounds to enter clinical testing just in a few months. Our partners Biogen/IDEC and Biotest both have submitted INDs for their TAP compounds, and we remain on track to submit the IND for our own IMGN388 compound this quarter.

Dan will now discuss our financial results, and I'll be back to you soon.

Thank you, Mitch. Good afternoon, everyone. Let me go through the third quarter, which is the March 31 quarter, for ImmunoGen.

We recorded revenues of $14.6 million in the quarter, compared to $9.8 million in the same period a year ago. Giving you the individual components, our research and development support fees in the quarter were $3.5 million, down from $6.6 million in the same period last year. Remember that this is primarily funding earned through our research collaboration with sanofi-aventis and, to a lesser extent, with other partners. And also recall, we are in the final contract year with sanofi-aventis. That year began last September, which provides for reduced R&D support fees versus last year, as development activity is being assumed by sanofi.

Our license and milestone fees were $5.2 million in the quarter just ended, compared to $1.5 million a year ago. This includes $1.5 million earned with the filing of the BIIB015 IND by Biogen/IDEC, and it also includes $2 million we received from Genentech in the first quarter of this fiscal year with the initiation of the T-DM1 Phase II testing, but that $2 million was not earned until this quarter, with the achievement of an event.

Our clinical material reimbursement was $5.8 million in the last quarter. That compares with $1.8 million in the same period a year ago. Again, remember that we make clinical materials on behalf of our collaborators, and earn clinical material reimbursement revenue with a supply of these and other materials to our collaborators. The greater revenue in the third quarter of fiscal '08 versus the third quarter of fiscal '07 is primarily due to our supplying $4 million of one of our cytotoxic agents to a collaborator.

Our operating expenses in the third quarter of '08 were $28 million, compared to $15.8 million in the same period in the prior year.

Our research and development expenses were $14.3 million, up from $12 million in the same period last year, primarily driven by a $2.9 million increase in antibody development and supply costs incurred in the period just ended. Our cost of clinical material reimbursed was $9 million in the third quarter of 2008 versus $1 million in the same period a year ago.

As noted above, or just discussed, we shipped a substantial quantity of one of our cytotoxic agents to a collaborator. This partly contributed to the significant increase in the cost of clinical material reimbursed in our Q3 of fiscal '08 compared to the same quarter last year. Also in Q3 of this year, we obtained substantial quantities of our DM1/DM4 cell-killing agents from a contract manufacturing organization. With this, we recorded $3.7 million to write this material down to its net realizable value, and to write off material in excess of currently forecasted demand, which is in accordance with our inventory policy. This material was produced for us by the supplier as part of scaling up the production process for these agents, resulting in more material being produced than is anticipated to be needed by our collaborators in the next 12 months. And just as a note here, our inventory policy is driven by material to be used by our collaborators, not by us, as ours would be classified as research and therefore written off.

While we have booked an excess reserve for this excess material, we expect it to be used by us or by our partners within the next several years.

Our general and administrative expenses were $4.7 million in the quarter just ended versus $2.8 million in the same period a year ago. During the quarter just ended, we recognized $700,000 of expense related to the rental of new laboratory and office space that we occupied in late March of this year, and also $600,000 in move-related expenses, classifying both as G&A expense.

We also saw patent expenses up $400,000 versus last year due to the timing of patent activity. This led to a loss from operations of $13.4 million, compared to a loss from operations of $6 million in the same period last year.

Our other income net was $500,000 in the quarter just ended, down from $800,000 in the same period a year ago. This is principally interest income and gains recognized on forward contracts. Also in the quarter just ended, this included $300,000 of impairment charges on investments and about $200,000 of losses due to foreign currency translation, these two partially offsetting the interest income and gains on forward contracts in the quarter.

The result, then, is a net loss of $12.8 million in the third quarter of fiscal 2008, or $0.30 a share, compared to a loss of $5.2 million, or $0.12 a share, in the comparable period last year.

Turning to the balance sheet and our cash flow, our cash and marketable securities as of March 31, 2008, were $41.2 million. That compares with $59.7 million as of June 30, the end of our last fiscal year. The capital structure continues to consist entirely of common equity; there was no debt, no converts, no preferred, no warrants on our books. Our cash used in operations for the first nine months of fiscal 2008 was $2.5 million. This compares with $11.5 million used in the first nine months of fiscal 2007.

Our capital expenditures during that period were $17.6 million. This compares to $1.4 million for the same period last year. Included in capital expenditures for the current period is $3.6 million of improvement of the Company's capabilities in our manufacturing facility in Norwood, Massachusetts, as well as $10.9 million to build out the laboratory and office space at the Waltham facility we occupied in late March.

These $10.9 million in leasehold improvements are being paid by our Waltham landlord and benefited our cash used in operations in the period. There's no net cash cost to ImmunoGen.

Finally, let me note that our guidance for the full fiscal year of 2008 remains unchanged. We expect our net loss to be between $28 million and $31 million, which is consistent with the guidance that we provided at the end of the second quarter. Similarly, our expected cash used in operations is expected to be $14 million to $17 million, and capital expenditures are expected to be between $20 million and $21 million, both of those figures consistent with our previous guidance.

Let me now turn the call back over to Mitch.

Thanks, Dan. As you can tell, we're very excited by our progress and that of our partners and expect a number of achievements in the next few months alone. As ASCO next month, or the end of this month, clinical data will be presented for both T-DM1 and IMGN242. We expect to submit an IND for our own IMGN388 TAP compound within the next few weeks and to start dosing patients this summer. We expect Biogen/IDEC's BIIB015 TAP compound to be in clinical testing by this summer, and we expect Biotest BTO62 TAP compound to be in the clinic by summer.

So by our next quarterly call, there should be eight TAP compounds and one naked antibody in clinical testing through our own programs and those of our partners--IMGN901, IMGN242, IMGN388, T-DM1, 889633, SAR3419, BIIB015, and BTO62 are all TAP compounds, and AVE1642 is a naked antibody.

During 2008, we also expect Genentech to disclose their Phase III go/no go decision for T-DM1, we expect to provide an update on our IMGN242 gastric cancer trial, we expect to expand our study 003 in a Phase II leg to evaluate IMGN901 as monotherapy in additional patients who failed multiple prior agents, and we'll initiate a trial evaluating the compound and combination with an approved agent, like Revlimid or Velcade. And we expect a wealth of clinical data to be reported at conferences in the second half of the year, in particular, EORTC and ASH. I look forward to keeping you posted on our progress. Carol?

Thank you, Mitch. Operator, we are now ready to open the lines for questions.

(Operator Instructions) We'll go first to Brian Rye with Janney Montgomery Scott.

Well, good afternoon, guys, and thanks for taking my question. Mitch, I guess one thing right now, on the relationship with Genentech, if I recall, I think it was three years ago that you all formally extended what was initially a five-year term for the broad-based, I guess, technology access agreement. I know that doesn't really impact the separate, target-specific agreements that you all have entered into over the course of that agreement, but I'm curious if that deadline has either passed or is about to pass, and if there have been any discussions about either extending that or somehow amending that agreement to continue to give Genentech that access?

Well, the deadline is about to pass, and there's been lots of discussion internally here at ImmunoGen. You know, the arrangement was that any time they exercised a license, the terms of those licenses would be the same, and those terms were negotiated eight years ago, Brian, as you point out. So we think we can do much better today, given the acceleration of the deal terms that we've had in our more recent deals, so we're pleased that when--if and when Genentech wants to take another license, that we'll be able to negotiate from a starting position de novo.

OK. And then I guess secondly, on the 242 data, the amendment to the ongoing trial, I'm curious to see, you know, learn anything more about this, the visual toxicity and how a same dose can essentially result in the observation of these toxicities for people with low levels of KNA expression but not high levels?

Right. So, it turns out that, you know, the half-life of the product in patients with low levels of circulating antigen is about 6.5 days, but is somewhat shorter with patients who have high levels, because the resulting immune complex is removed from the serum that much faster. So as a result, the total exposure of the patient to the agent is higher in patients who have lower circulating antigen concentrations than those that have higher. And as a result, we think we're seeing the toxicity. So that's the reason we stepped down half--stepped down one step in our dose escalation scheme. We went back one level. And we think we have this problem licked.

Great. And then lastly, on the naked antibody that sanofi-aventis is developing in pre-clinical development that you mentioned here, is that something that they received from you all, or how is that related to ImmunoGen?

They did. They received that from us in, when we first organized this deal five years ago, although it was not yet a clinical product. And with them, in our collaboration, we put it into the clinic.

So you still stand to get milestones, royalties, as you would for a TAP product candidate?

It's exactly the same deal, $20 million to $30 million in milestones and royalties that go into the double digits.

Great. Congratulations on the progress.

Thank you.

Thank you. We'll now go to Joel Sendek with Lazard Capital Markets.

Hi. Thanks a lot. Let's see, a couple of questions. First of all, on DM1, I just want to walk through a couple of scenarios. So I'm sorry, the Herceptin or trastuzumab and DM1.

Right, we got it.

I think you have more than one DM1 compound.

We do, but we get a lot of questions about this one.

Right. So the first is, do you have any sense of how far along the current study is? And then the second question is, is it possible that a Phase III could start before they have the Phase II data?

Well, to answer your first question, the objective was to accrue, as we understand it--this is what's publicly disclosed--the objective was to accrue 100 evaluable patients, and they have over 40 centers open to do that. And they started last July. And there's an awful lot of excitement out there about this product, so I have no idea how far along they are, but I imagine they're pretty far along in terms of accruing patients. So that's the most information that we have.

On the second topic, at their quarterly call, their most recent quarter, Sue made the point that they would not need to disclose the results of the Phase II before making the decision on Phase III. I think what could be inferred there is that they're going to see the data. They don't need to disclose the data in order to make that decision.

And maybe this is wishful thinking, but could the decision be to file if the data are very strong? I mean, is that out of the question, do you think, or is that--?

Well, I think that to file, they would need to be doing a pivotal study. And probably with pivotal material. And although they engaged us some time ago to develop a commercial manufacturing process to yield pivotal material, and we've done that and delivered it to their CMO, we're not in the position to say whether or not they've been using pivotal material in the current Phase II. They've certainly not given any public indication of that, Joel.

And along those lines, you delivered this $4 million worth of cytotoxic agent to a collaborator. Can you tell us who that collaborator is or what the non-financial significance of that is?

No, we really can't, Joel. We don't identify specifically material that we provide to a third party.

Okay. And then I'll end with just a financial question. So with regard to this write-down and how you're taking the hit this quarter, but you can actually use this stuff in your own R&D in the future, should that effectively depress the R&D expense for a couple of future quarters lower than it would otherwise be? Or how should I interpret that?

Yes, I think that's a fair statement. That would be true, I think, both with that material, the clinical material reimbursed, as well as to some extent the R&D, because I referenced some antibody costs as well. We've been talking for a while that we had some significant costs this year associated with antibody development as well as with DM1 and DM4 development. And some of that's, a good portion of that's reflected in some of the unusual numbers this quarter.

And to create an analogy, you actually have seen a precedent event, because the Company, because of how it has to source its cytotoxic agent, it comes in quantities in excess of--they're sort of incompatible, if you will, with our inventory policy. So our inventory policy, as I noted, is geared towards what we provide to collaborators. So we went through a similar exercise in fiscal 2004 to fiscal 2005, where we ended up creating reserves for excess material. And actually, that excess material was used either by us or our partners over the succeeding three years to the tune of roughly $3 million.

So I think that you'll see the same thing, if--looking at, for example, our fiscal 2007 results. If you look at the spread between clinical material reimbursed as well as the cost of the material, you'll see some margin in there. A good portion of that margin is the recovery of the material that we had written off in earlier periods.

Got it. Okay. That's it. Thanks a lot.

Sure.

Of course.

We'll go next to Pamela Bassett with Cantor Fitzgerald.

Hi. Thanks for taking my call on. So much is, there's so much going on, and I apologize. I missed the very beginning of the call and would like to hear a time line for 901 development plans at this point. Just very briefly.

Well, what we said is that we're continuing the monotherapy Phase I study.

Right.

Along the lines of what our expert panel that we convened last month suggested that we do, and then we'll expand the top cohort once we reach MTD. We haven't reached MTD yet. So once we do, we'll expand that and get a real good sense of whether or not this drug is suitable as a monotherapy agent to treat end-stage multiple myeloma. Because that's what all these patients are. These patients have failed everything. If that's the case, then we would consider at that point--and we're encouraged by this group--to go to a pivotal study. We expect that news to be public by the end of the calendar year--before the end of the calendar year.

Now, simultaneously with that, we will initiate a Phase I study examining--and this is again on the recommendation of the expert panel--a combination study with our agent and either Velcade or Revlimid, probably Revlimid, since our animal studies show really profound synergy with Revlimid. Still synergy with Velcade, but not quite as much. And see what those results are, and if the monotherapy looks like it's a starter, then we'll go that route. If it appears to be a non-starter, that is, that we couldn't get enough activity in monotherapy in this patient population, then the combination study would be a more reasonable approach to go to. And so those--.

And when do you think you'll have the combination study results? At the same, by the end of the year as well?

No. We'll have started the combination Phase I before the end of the year, but I don't think the results will be in by that time, Pamela.

Okay. So that's the going to flow into '09.

Yes. I mean, it's a simple, straightforward study, and it should go quite quickly in that the dose of Velcade and Revlimid is well known and established, and we have some idea of the safety of our product, although we haven't reached the top dose yet. So we probably won't start all the way at the bottom. So the escalations will be faster. But nevertheless, we need to dose escalate until we see the maximum tolerated dose in combination with Revlimid and say, high dose dex, which is probably what we would do.

All right. So you'll have the monotherapy data before the combination therapy data. And assuming that the monotherapy is good and it's positive and you want to move forward, are you considering--and then, if really best case is combination is good as well, combination data is good--will you proceed with both?

Well, that's a good question and it depends on the state of our balance sheet, amongst other things. But it also depends on something else, and this is something that was one of the principal reasons we brought this expert panel together. If we were to do the Revlimid combination study in a Phase II setting, it would be second-line. And as you know, there's a lot of products out there that are competing for patients for second-line, plus or minus end product with Revlimid. And so if we think--and, more importantly, FDA agrees--that we can come to an agreeable and agreed-to threshold above which activity will yield registration in monotherapy, then I think that would be the preferred route. It would certainly be faster.

Okay. So--?

Did I answer the question?

Yes, yes, very much so. And thank you for going through it again.

Of course. Is that it?

That's it.

Okay. Nice talking to you, Pamela.

You, too. Thanks.

Thank you. We'll go next to Jason Kantor with RBC Capital Markets.

Right. Thanks for taking my question. I want to understand this shed antigen issue. Do you understand why some patients are shedding more antigen? Are those the patients with higher tumor burden? Let's start there.

That's a good question. No. It turns out that it varies by tumor. Colorectal, almost everybody has high levels. By high and low, we're making it qualitative. We draw a threshold at a certain level, say, "If you're above that threshold, you're high, and if you're below, you're low." And you find that colorectal, the majority of patients--overwhelming majority of patients--have levels of circulating antigen above the threshold. Pancreatic is split 50/50, in our experience. And gastric is 70/30, 70-low, 30-high. And so we just have to screen and determine it. Jason, I don't think it's related to tumor burden. But I couldn't, I'm not 100% sure of that.

Now, is this, is this just a first-dose issue? Or once you clear the circulating antigen, is it right back at you the next time you're, the next dose you could reach for?

That's right. The dosing schedule's once every three weeks, and three weeks has gone, and the circulating antigen level is back. It's not perhaps as high, but it's still back.

Or are you modifying the dose on a every-dose basis on this, for this?

We haven't done that, and we're--.

So it's just baseline.

We think we're okay with not doing that based on the pharmacokinetics of both the product and the circulating antigen. You raise a good point, but we think we're okay here.

And do you have any good sense of what is this eye toxicity that you're seeing? That's on a huge dose differential between the two doses. I mean, how concerned are you that you're going to be hitting those same toxicities at the doses you're using?

Yes. Well, that's also a good question. The information comes from our Phase I study, obviously, and for the limited number of patients we've dosed in Phase II. The eye toxicity expresses itself as a reduction of visual acuity. And put it in lay terms, the patient, the couple patients who have had it describe it as "blurry vision." On examination, and you can take our word for it, these patients are scrupulously examined, one sees a ring of particles in the cornea. And that apparently contributes to the, or causes the blurry vision. The particles diminish and are gone within a month or two once treatment stops. So everybody who has had this loss of visual acuity has recovered to exactly the state they were before they were dosed. But that's what's seen.

We don't know what it is. It has the appearance, it resembles an inflammatory process in terms of the outcome, that is, this ring of particulates. But we don't know. But we can reproduce it, sort of, in rabbits. It takes a somewhat different form, but we have seen a similar corneal deposit--not exactly the same--in rabbits.

But when you looked back at your Phase I data, the people who got this toxicity, were they people who had low circulating shed antigen? The body released--?

No, they were almost all high circulating antigen, and they got the two, I think, that had this--or one had the expression of it and one had a mild version--were at much higher, well above 168. In other words, 168 was established as the NTD because at higher doses we saw it. But colorectal, you can, I can't give you the exact number, but nearly everybody has high levels of circulating antigen.

So for NTD for high level, it's 168. What was normally established as the NTD, we backed off the high dose to 168. I've never seen, never saw colorectal with this problem at 168, at high levels of circulating antigen. Gastric, generally low levels of circulating antigen, 168 appears to be too high. At least it was in one or two patients.

Okay. Just real quick. How many patients have you enrolled in the study, and is there going to be any, a delay? Do you have to get new IRB approvals for this new protocol? And lastly, what's your next milestone for TDM-1? Do you get paid if they make a Phase III go decision?

All right. So on the first question, we haven't disclosed the number of patients, but clearly it's enough patients to have seen this issue and make it meaningful. Number two, yes, it requires a new IRB approval, but the good news is, that's happened. And so we're back recruiting patients. And number three, we haven't disclosed milestones beyond that which we receive. We received $5 million for Phase II. You can make your own inferences as to--let me say this. We're due $44 million in milestones. We've gotten $7 million, and none of them are associated with sales. So that's all I can say.

Thanks.

You're welcome. Nice talking to you.

We'll go next to Robert Manning, private investor.

With all the talk coming from so many places about the importance of antibody drug conjugates, I've been trying to find a way to quantify just how important ImmunoGen is in the overall picture. So I went looking to see how many drugs I could find that had actually made their way into the clinic, and counting Biotest and Biogen/IDEC, which I guess are in the process of getting started, I found a total of 10, seven of them from ImmunoGen and your partners. And the only other ones I could find is a Wyeth drug that uses calicheamycin and their linker and a couple from Seattle Genetics and Curagen using Auristatin. But everything else is ImmunoGen's linker and (inaudible) payloads. Now is this a fairly complete picture, or is there a lot of other stuff going out there that I haven't found?

Well, interesting. You found the same ones that--well, let me say this. We're aware of, of course, the one marketed product, Mylotard, that uses--.

I left that one out, yes.

Yes. I think Wyeth actually has two--no? I thought they had two. Okay. Well, one. And yes, and that's it. We're not aware of any others.

Okay. Thanks.

You're welcome.

Thank you. It appears that that concludes the question-and-answer session today. At this time I'd like to turn the call back over to Carol Hausner for any additional or closing remarks.

Lovely. And I wanted to just invite, thank you, audience, and also let you know, too, that we--as you know, we have moved to our Waltham facility, so if you have any additional questions, give me a call at 781-895-0600. And I wish you a good evening. Take care.

Thank you.

This concludes today's teleconference. You may now disconnect, and have a nice day.