ImmunoGen Inc (IMGN) 2016 Q1 法說會逐字稿

Good day, and welcome everyone to the ImmunoGen first quarter fiscal year 2016 financial results conference call. Today's call is being recorded. At this time for opening remarks and introductions, I would like to turn the call over to Executive Director, Investor Relations and Corporate Communications, Carol Hausner. Please go ahead.

Thank you, and good morning. At 6.30 this morning we issued a press release that summarizes our financial results for the quarter ended September 30th, 2015, the first quarter of our 2016 fiscal year. I hope you've all had a chance to review it. If not, it's available on our website. During today's call, we will make forward-looking statements. Our actual results may differ materially from such statements. 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. In our call today, our Chief Executive Officer, Dan Junius, will provide an update on ImmunoGen, our Chief Development Officer, Dr. Charlie Morris, will discuss our wholly-owned compounds in greater depth, and our Chief Financial Officer, Dave Johnston, will discuss our financial results and guidance. We'll then open the call to questions. Dan.

Thanks Carol, and good morning everybody. In previous calls, I've referenced this being a transformational year for ImmunoGen. I think that characterization continues to be accurate, given the progress we're making with our lead product, mirvetuximab soravtansine, and our other wholly-owned programs, as well as the activity we're seeing from our partners. We're extremely excited about mirvetuximab soravtansine and it's potential to make an important difference for patients with cancers that its express its target folate receptor alpha, and we're successfully executing on our strategy of aggressively advancing mirvetuximab as a single agent for pretreated ovarian cancer, while also expanding its potential uses. Data in support of mirvetuximab soravtansine continued to expand, as evidenced by the AACR-NCI-EORTC, or what I'll be referring to as triple meeting abstracts that became public last night.

Of particular note as the association we're finding between level of folate receptor alpha expression on patient tumor cells, and response to treatment with mirvetuximab monotherapy. As you recall, at ASCO we reported an impressive 53% response rate with mirvetuximab in patients was platinum resistant ovarian cancer. When we look at responses stratified by the level of target expression on the patient's cancer cells, the findings become even more interesting. Notably, with the data available at the time the abstract was submitted, there was a remarkable 80% response rate among the 10 patients with the highest level of folate receptor alpha on their tumor cells. Charlie will take you through this in greater detail. Clearly this is an exciting finding, and we'll be providing updated data on these patients at the triple meeting.

I'm pleased to note that we completed enrollment for the full 40-patient cohort in August, well ahead of plan, and that patient enrollment is going faster than expected in our ovarian cancer cohort which require biopsies. We look at this progress as being a strong indicator of investigator support. We're planning to meet with the FDA about an accelerated pathway for mirvetuximab in ovarian cancer in the first half of 2016, and to report findings for the full 40-patient data set at ASCO next year, and of course, we'll be preparing to start our FORWARD I and FORWARD II trials by the end of 2015.

As we're successfully executing on mirvetuximab soravtansine, we're also advancing our promising earlier stage programs, we're on track to begin assessment of our IMGN529 and coltuximab ravtansine ADCs, in combination regimens by year end, and in 2016 respectively. These trials will focus on treating patients with diffuse large B cell lymphoma. Also of significance we've submitted the IND for our IMGN779 product candidate. This is the fourth IND in four years, ahead of our goal of submitting an IND for new product candidates every 18 months. This ADC is a potential treatment for acute myeloid leukemia and myelodysplastic syndrome, where new and better therapies are needed, and it uses the first of our new IGN payload agents. We see great potential with our IGN program, and look forward to begin treating patients with this first IGN.

Before Charlie takes you through our programs in more detail, let me previously touch on some of the progress being made by our partners. Exciting Phase Ib findings with Bayer's ADC anetumab ravtansine in the treatment of mesothelioma were reported at the World Conference on lung cancer last month, as discussed at our recent analyst event. Of particular note is the durability of some responses. With five of the 16 pretreated patients with this deadly cancer on treatment for over a year, and three still on study after more than two years. I'm also pleased to report that both Sanofi and Amgen have additional ADCs with our technology, and that patient dosing has started with the first ADC from our collaboration with Lilly. So we're making very solid progress on our wholly-owned pipeline, as are many of our partners. With that, Charlie will now discuss our programs in more detail.

Thanks Dan, and good morning everyone. I will start with the lead program of mirvetuximab soravtansine. As Dan mentioned, the triple meeting abstracts became public last night. If you've had a chance to read abstract C47 and C170, I'm sure you'll understand our continued excitement about this program. The C47 abstract provides initial findings on the association between the level of folic receptor alpha on patient tumor cells and response to mirvetuximab, using data from the 22 patients with platinum resistant ovarian cancer in the Phase I expansion cohort reported at ASCO in May. At the time, the abstract was submitted to the meeting we had response rate data for 17 patients as we had at ASCO, and so that is what in the poster. The poster will be presented on Sunday, November 8, and will include updated data for all 22 patients, including updated response and duration of treatment information for all the available patients.

As noted in the abstract, this analysis divided the patients into three groups based on level of target expression. High expresses have moderate to high levels of folate receptor alpha on at least 75% of their cancer cells, medium expresses on 50% to 74% of cells, and low expresses on 25% to 49% of cells. Notably, about half of the treated patients were high expresses, and more than three-fourths were medium or high expresses. In the abstract objective response on treatment with mirvetuximab was reported in a remarkable 8 out of 10, or 80% of the high expresses, one of the five medium expresses, and neither of the two low expresses.

This is consistent with our hypothesis that higher expresses may have a higher response rate, and explains why FORWARD I will be limited to high and medium expressing patients, with stratification between the high and medium groups. This approach will give us considerable flexibility around our registration strategy, as we continue to build patient data. Our findings to date indicate that not enrolling low expresses excludes only a minority of patients, while ensuring that we retain the opportunity to provide benefit to a majority of ovarian cancer patients. Our success in the enrollment of patients with ovarian cancer encourages us to think that FORWARD I could be recruited in good time.

As Dan mentioned, we completed enrollment in the 40-patient cohort ahead of schedule. And enrollment in our other ovarian cancer cohort, the 20-patient cohort, for patients requiring biopsies before and on treatment, also is ahead of schedule. In fact, we dosed the 20th patient yesterday. We're now gaining safety, efficacy and durability results for the full 40-patient cohort with platinum resistant ovarian cancer, and plan to discuss these findings when they mature with the FDA. We also expect to report them at ASCO next year.

The other mirvetuximab abstract at the triple meeting, abstract C170, relates to our interest in rapidly moving this ADC into assessment in combination regimens, and the design of our FORWARD II trials. Our preclinical assessments indicate that the addition of mirvetuximab to Avastin, to carboplatin, to Avastin and carboplatin together, and to doxil, has the potential to meaningfully enhance efficacy. For example, we found using mirvetuximab with Avastin had a more than additive impact on the activity models of platinum resistant ovarian cancer. The full data will be disclosed on November 8. You can see from the abstract that these preclinical models suggest that mirvetuximab in combination with agents like carboplatin and Avastin, could potentially exceed the efficacy of current standard of care combinations. Of course, this requires that the drugs can be safely combined, and this will be evaluated along with initial assessments of efficacy in the FORWARD II study, which like FORWARD I, are on track to begin by the end of 2015.

While I spent considerable time on mirvetuximab, I should note that we're also making meaningful progress with our other pipeline program. We submitted the IND for Amgen 779, and expect to begin Phase I testing in the first part of 2016. We are on track to begin clinical assessment of IMGN 529, using combination with Rituxan this quarter, and to start clinical testing of coltuximab ravtansine in a combination regimen in 2016. In all, we're making solid progress advancing our strong pipeline. On that note, I'll turn the call over to Dave to discuss the financials. Dave.

Thanks Charlie, and good morning everyone. So as Carol noted, we issued a press releasethis morning with our first quarter fiscal year 2016 financial results, and I will review the highlights, and then also review the guidance for our full fiscal year of 2016. For the quarter, revenues in our first quarter of our fiscal year were $14.9 million, as compared to $13.2 million for the same quarter of last year, and of that, revenue from licensed and milestone fees was approximately $6 million, which includes milestones earned from Lilly and Amgen, with advancement by each of a new ADC to clinical testing. Our revenue in the current period also includes $5.7 million of non-cash royalty revenue on sales of Kadcyla, for the three months ended June 30th, 2015. As a reminder, the royalties pass to the purchaser of the royalty stream, and we report non-cash Kadcyla royalty revenue, that is partially offset by non-cash interest expense resulting in non-cash net income.

Finally, our revenues also include $3.1 million of clinical materials revenue and research and development support fees. Operating expenses for our first quarter of fiscal year 2016 were $43.5 million, compared to $35.1 million in the first quarter of last year. These consisted of $35.1 million of R&D expense for the first quarter of this fiscal year, compared to $28 million in the same period last year. And $8.3 million in G&A expenses this year, compared to $7.1 million in the same quarter of last year. Driving the change from previous fiscal year are greater third-party costs related to the advancement of our product candidates, especially mirvetuximab, and higher personnel-related expenses related principally to recent hiring.

For the first quarter of fiscal 2016, we reported a net loss of $33.7 million, or $0.39 per share, compared with a net loss of $22.3 million, or $0.26 per share for the same quarter of last year. Cash used in operations was $31.4 million for the first three months of fiscal 2016, compared with $18.9 million in the same period of last year.

Our capital expenses were $3.4 million in the first quarter, compared with $1.7 million in the same period of last year. And lastly, we ended the quarter with $247.8 million in cash and cash equivalents, compared with $278.1 million as of June 30, 2015, and we continue to have no debt. Let's go over our fiscal year guidance. It remains unchanged from what we issued on July 31st of 2015. Specifically, revenues are expected to be between $70 million and $80 million, operating expenses to be between $175 million and $180 million, and net loss to be between $120 million and $125 million. Cash used on operations between $100 million and $105 million, capital expenditures to total between $13 million and $15 million, and we expect to end fiscal year 2016 with between $165 million and $170 million in cash. We firmly believe that development of our own portfolio of compounds will result in the greatest creation of value for our shareholders, and we're making significant advance the to advance mirvetuximab soravtansine as aggressively and expeditiously as possible. With that, let me turn the call back over to Dan.

Thank you Dave. As you heard from Charlie, we have a lot going on, and look forward to continued strong progress and a steady news flow. Of particular importance are our anticipated mirvetuximab events, starting with the presentation and patient response data stratified by level of target expression on November 8 at the triple meeting. We intend to report the findings from the full 40-patient cohort of patients with platinum resistant ovarian cancer at the 2016 ASCO meeting. We're likely to have additional presentations on mirvetuximab at ASCO, such as the findings in the patients with endometrial cancer, and/or from biomarker assessment of tumor biopsies.

We'll have to see how these data mature, and where we are at the time of abstract submission. And of course, we expect to start FORWARD I and FORWARD II by the end of this year, and to meet with the FDA about an accelerated approval pathway in the first half of 2016. We'll also be reporting findings from preclinical assessment of mirvetuximab and combination regimens for ovarian cancer at the triple meeting, and for IMGN529 and coltuximab ravtansine in combination regimens with B-cell malignancies at ASH. We expect to begin testing of IMGN529 in combination with Rituxan this year, and assessment of coltuximab in a combination regimen in 2016, both in pretreated diffused large B-cell lymphoma patients.

Finally, we plan to advance our IMGN779 ADC into patient testing early next year. We're looking for marked progress on the partner front as well, including two of our partners starting registration enabling trials in 2016. So we're executing on our product plans, we look forward to providing you updates as these programs progress. And with that, I will turn it over to Carol, so that we can take your questions.

Thank you Dan. We are about to open the calls to questions. We ask that these be limited to one to two questions per person, until each analyst has had a chance to ask a question. Operator, we are now ready to open the lines to questions.

Thank you. (Operator Instructions). I'll pause just a moment to allow everyone an opportunity to signal. We will go first to Jessica Fye with JPMorgan.

Hey guys, thanks for taking my questions. A couple on mirvetuximab. First, do you see the mix of high and medium folate receptor alpha expressors you enrolled in that ovarian cohort you're talking about, do you think that's reflective of the overall patient population? I know they're small numbers, but what gives you confidence that those won't swing around too much? And then second question is, can you walk through the decision to enroll both mid and high expressors in FORWARD I one and not just high, and is there a target number of patients for each level of expression, or do you take them how they come? Just thinking about how the mix of those patients could affect the results of the first stage of the study? Thanks.

Thanks, Jessica. I think the first question is the important question. What we have been doing is tracking not just the patients who go into the study, but obviously we do get patients who come in for screening, and for various reasons don't make it into the study. I would say that the mix of patients who have gone on to the study has been reflective of the overall screening pool, if that makes sense. So I think in that larger cohort of patients, yes, I think there is a consistency between what we're seeing, and I think this kind of if you like, 60/20/20 mix between the high, medium, lows, is probably going to be about 50/25/25 roughly, is roughly how we expect things to pan out.

In terms of making decisions, in each individual data set at this point, there are relatively small number of patients. And clearly there's high activity in the high expressors, you'll see more about that in a couple of weeks time. There's clearly activity in the medium expressors, and we just want to have more opportunity to understand that in greater detail. We're not trying to limit specific numbers of patients by those categories. We'll take what we get. But we will stratify to make sure that there's balance. And one of the upsides of the two stage design, of course, is that at the end of Stage I of FORWARD I, we've got an opportunity to take a look and say, do we want to keep those medium patients in, or as another evidence of benefit we'll just stick with the high. So it enables that decision to be sort of kicked down the road, without necessarily impacting our ability to make comparisons to the physician's choice.

Okay. Great. Thanks. Can I just squeeze in one follow-up on that. The one patient that you had in that mid expressor group who was a responder, I know you had some unconfirmed partial responses that were just over the threshold, was that mid level patient kind of one of these borderline responders, or did they have a more significant response?

November 8th.

Thank you.

Nice try.

We'll go to Matthew Harrison with Morgan Stanley.

Great. Thanks for taking the question. So one just to follow-up on Jessica's questions, can you tell us how many patients you've screened, just so we have some idea of the sample size you guys are looking at? And then, just can you help us think about your meeting with the FDA. I think you said you completed the 40-patient cohort enrollment in August. So when do you think you'll internally have data, and how has that set you up for a timeline in terms of meeting with the FDA, and do you need to have that, would you send that full data set into them before requesting a meeting, or how should we think about those order of events? Thanks.

So, in terms of No. 3, I apologize, I don't have the exact number. It's around 100 that have been screened in total, at various points along the way. On the FDA front, the whole idea has been to allow that data set to mature, so that we have the patients who were coming in, remembering that we recruited less patients in August, we'd like to give those patients the opportunity to respond and get some durability to their individual responses, before we take that data down to the Agency. The plan there would be to have all of that first stage of the next study begins, but we would have their feedback, their input, and their thoughts on the second stage, long before that second stage begins.

So just to be clear, is there a certain level of duration that you're looking for in that first 40 patients before going to go meet with the FDA?

I think it's more a matter of knowing what it is. I think to start to see things as particularly interesting, I think we'd like to at least be thinking about at least a six-month meeting duration of response. We get some sense as you'll see in a couple of weeks time, on how durable the responses are, but it's early to be thinking definitively about what that is, so I think taking a more complete data set, and having a real sense of where that is will help us enormously in our conversations with the Agency.

Okay. Thanks very much.

We'll go to Andrew Peters with UBS.

Thanks guys for taking the questions. Just a couple. I guess, first, to follow-up on Jessica and Matthew's question on screening, so if I heard you right, you said you had screened 100 patients. I just wanted to understand, I guess what's the primary cause of the screen failure that you're encountering so far? And just on folate expression levels, I just wanted to understand, is there any reason that expression would change as patients, I guess by previous therapy, I just want toward if you'd expect kind of the folate receptor expression to be similar as you move into earlier settings? Thanks.

Yes, I mean, in terms of screen failures, without necessarily bearing down into all of the details, in the Phase I population, as you know, you're often dealing with sick patients who have a variety of other things, and you have your set entry criteria, so we get a lot, you get a certain number of patients who come in and they will get their screening, but then you find during the rest of the screening process, that maybe their chemistry doesn't work, or their baseline hematology doesn't work, so it's a variety of things. Some patients decide they don't want to go into a clinical trial after all. Some even progress within the screening period. So there's a variety of things that can happen, and there's no sort of fixed pattern to it, other than fairly typical for what you see in a Phase I setting.

The expression change is a really important question, and that's really one of the key questions that we're asking in this biopsy cohort that we've talked about. That will give us the expression to look back to the archival tissues, on which we've been basing our entry decisions so far, and say well, does that appear to correlate with what we see at the time of a fresh biopsy, two, three, four, however many years later it is. Those data obviously are not mature at this point. I would hope that we would have enough information on that to also submit that to ASCO. But I think intuitively you would think, well, the high expressors seem to be doing very well. That makes kind of intuitive sense. So I think we're probably going to find that it's a stable phenol type, but we'll have data to confirm that

Okay. Great. Thank you.

Simos Simeonidis with RBC Capital Markets.

Good morning. Thank you very much for taking the questions. On the slides that Carol sent us before the call, when you are describing the entry criteria for FORWARD, there's a line here that says raise the FRA cut-off. Can you talk about what that change was, from what to what? Is it that before you were not using low expressors, sorry, that you were using low expressors, and now you're just going to use medium and high?

Yes. So previously it was the 25% of at least 2-plus. Now we're just raising that to be 50%.

Okay.

It will include the medium and high, but those two groups will also, we will stratify the randomization, so that we have an equal distribution between arms, that we think should be particularly important in Stage I of course, because what you don't want to do is find that your efficacy is influenced by that distribution, and it gives us another opportunity to take a look at the medium patients and say, are we seeing enough. But the lower expressors, the 25 to 50, of which you'll notice there's not many in the study anyway, those we think we will leave those out even starting now.

Okay. Great. And then one for you, Charlie. Without getting too technical I guess, talk a little bit about the assay used, and how much variability do you think there would be in these assays, when scoring cells for low, high, and medium expression, depending on how the biopsy is excised? What I mean by that is, when you're looking at, I guess cells, the chemistry, and you're scoring tumor cells versus cells from the adjacent tissue, is that a concern, or do you think that the difference would be pretty much the same, or similar across samples?

At this point we believe it would likely be the same. Again, it's another important question, and indeed it's one of the reasons, I think, to not just focus entirely on the highs at this moment, is to say, well, given that there is indeed, with IXT there is, if you like, a subjective element, people will look at it, they'll look at the density of staining, and make their interpretation of whether it's 2-plus, is it 3-plus. So I think the patients that are maybe close to the line, if you like, around that 75% level, I think it's important that we understand how they do, before sort of excluding that medium group of patients. It should be reproducible. Part of the testing that happens obviously, hopefully when we get around to having a commercial product, part of the process there is to educate pathologists on the use of the diagnostic, and we think it's important that we're working with a very experienced diagnostic group, working with Ventana, that's a collaboration that is going well, and we think that's an important, going to be clearly, especially having made these decisions about the cutoffs, it's going to be an important part of the success of the product.

Great. Thank you very much.

We'll go to Mara Goldstein with Cantor Fitzgerald.

Thank you very much. I've got a question, and I guess you're going to tell me this is a November 8th kind of answer, but with respect to the responses that you're seeing and the expression levels and durability, are we going to be able to look at patients and duration of treatment according to the stratification?

What you will see is, if you remember the swim lengths that we've used previously, you'll see those swim lengths laid out by high, medium, and low expressions. So it won't be that you'll be sort of looking at a median. I think the individual numbers in each subset are very low to be talking about medians, but I think you'll have an expression, because you'll see the data split out according to the level of expression on those swim lengths.

And just to take it back to Simos' question for a quick second, so if you think about it for a second, and when you're looking at HER2, for those persons who appear on the margin between 2-plus and 3-plus, they often go onto fish diagnostics, just to confirm treatment. Is there any opportunity to do the same with the folate receptor, in terms of really being able to identify which of these patients on the cusp might really be those that should be in the high group, versus the moderate group?

Since at this point we are getting, we're including both, at this point we have not looked at that. I think it's an interesting question to see whether, would those be patients, for example, who might be candidates at that point, rather than to use the archival tissue, might you use, ask for a fresh biopsy at that point, to see how it looks today. I think that's a story that will continue to evolve, particularly if we continue to see what we would think were important differences between the medium and the high group. But we haven't looked at that at this point.

Okay. Thanks very much.

And we'll go to Christopher Marai at Oppenheimer.

Hi, good morning guys, thanks for taking the question. So in keeping with the theme on the call here, we'll touch upon the folate expression. I was wondering if you could comment on any regulatory discussions you've had regarding the measurement of the folate receptor expression. Obviously it seems like you'd be looking to enrich a patient population, or a trial based on those levels. Is the FDA okay with your measurement techniques, and how would you implement that in a real world setting post-approval, let's say, or even think about it with respect to the label?

Well, at this point, that will be another part of the discussion with FDA as we go forward. FDA clearly reviews all of our protocols, and has an opportunity to comment as we go forward. And there's nothing that we're doing here that I think anybody would consider anything else than a relatively standard approach to how we're seeing oncology products being developed. So I don't think that there would be any particular concern there. But that would be part of the discussion with FDA. But clearly thinking about, as you move through the various stages to having something defined as companion diagnostics, clearly one of the things that will be happening is that we will be launching a diagnostic at the same time as we launch the product, if we are successful enough to have the pleasure of being in a launch situation. And so I think the two will go together. But I think it will be an important part of success, and it enables you to define a population who, if we continue to see anything like this level of activity, that would define a population that would be very important.

Chris, to reemphasize Charlie's points, the FDA does have exposure in the protocol submission to the full scope of what we're doing, and therefore they can see what the screening technique is that's going to be used, and have the opportunity to comment on it at that point. So I do think they have that exposure. And hopefully that addresses the concern that you raise.

Okay. And then with respect to the diagnostic, in terms of development, where are you at, and maybe when do we expect to hear an update from that?

In terms of where we are, as I say, we're working with Ventana. You go through the various stages, where we have, if you like, a lock on the design of what we hope will be a diagnostic, at this stage it's officially diagnosed as an investigational use only diagnostic, but that would be intended to not need any further tweaking, if you like, prior to being hopefully approved at some point to the companion diagnostic to go along. So we would anticipate this would be a part of what would be in the prescribing information, would be to say that it will be patients defined as the appropriate level of expression as defined by this particular diagnostic.

Right. And then remind us, you're using this diagnostic in the FORWARD trials?

Yes.

Then finally, just one last question we'll throw in here, thinking about the competitive landscape, obviously Opdivo is something that's being investigated in the folate setting. I'm wondering, how are you guys looking at potential complications with IO type therapies? Thanks

It's something we're exploring. I mean, we're looking at the full range of existing therapies and potential emerging therapies, so I think we want to make sure that we look at the full range of opportunities of how these things may benefit patients. So that is certainly on the radar screen, Chris.

Great. Thanks for taking the questions.

We'll go to John Newman with Canaccord.

Hi, thanks for taking my question. So my question is just with regard to the design of the studies going forward. I'm just wondering if there's been any changes to any sort of cut-off that was in the expansion cohort, whereas you set a time limit in terms of the fastest amount of time that a patient could have progressed, and if they progress faster than that, they are excluded from the studies. I'm just curious if there's been any changes there, just so that you screen out people that might have relapsed, or progressed really, really quickly in the study going forward?

The key criteria will be that we are targeting patients who have had three or four prior lines of therapy, and are otherwise eligible according to a variety of criteria, including epithelial, ovarian, we are not including clear cell, we're not including some of the more low-grade ovarian tumors, but in terms of rapidity of prior progression, there wasn't actually anything specifically defined in the current expansion cohort. So I think the population that you're going to see is going to be largely similar. They won't be as heavily pretreated as they were obviously in dose escalation in the prior studies, but other than that, it will be typical of what you would see in a resistant population.

Now, what we have said in prior studies was that the traditional definition of platinum refractory, those were excluded, those patients who basically don't respond to platinum refractory. They are thankfully a relatively uncommon, if not rare population, and I guess the sad thing is, that they don't generally get to be fourth or fifth line patients, because they do very badly on subsequent therapy, so I don't think that population is likely to be a candidate, unless they are one of the fortunate minority who have had the opportunity to respond to subsequent therapies.

To emphasize, John, there wasn't anything in the criteria that looked at the rate of patient progression. If there's a suggestion that we were looking for a patient population that was not, the disease was not progressing as rapidly, we're getting a quote, less sick patient population, that isn't anything that even would have come up. It was simply a matter of a number of criteria, a number of prior therapies, et cetera, but the time it took for the patient to progress is not something that would be evaluated at all.

Great. Thanks. That's very encouraging. Thank you.

We'll go to Boris Peaker with Cowen and Company.

Great. Thanks for taking my question. I just want to know about the FORWARD II study. Will you specifically select high expressors there, and in general, how many patients do you plan to include in each combination arm for that study?

So for FORWARD II, we're actually not changing the cut off for that. That will remain at 25%. The data that we have suggests the ability for activity across, so we will continue to look at that low, medium and high, and obviously we have the opportunity to change that according to what the data show. When it gets to the expansion piece of that, it's 24 patients in each of the Avastin groups that we were planning to treat. I'll remind you that was one group who had not had prior Avastin exposure, and one who had had prior Avastin exposure. And although we haven't necessarily discussed what other expansions may occur, I think that will be determined by what we see in dose escalation, but I think it would be similar size cohorts.

Got you. And will you have some data from that study at ASCO?

No. It would just be too soon. We're not anticipating recruiting patients until toward the end of this year. The number of patients, and with the best will in the world it's always a slow start, because you've got to get your IRB approved, and everything else going, so I don't think there will be anything that will be particularly helpful to present at ASCO for that one.

Great. Well, thank you very much for taking my question

We'll go to Michael Schmidt with Leerink Partners.

Good morning, and thanks for taking my question. I just had one on FORWARD I. I notice you are also enrolling a once every four weeks dose cohort in the source part of the study. And my question is, can you just remind us of what you've learned in terms of the PK relationship and associated toxicities, and sort of what side effects could be reduced, I guess, with the less frequent dosing? Is that mainly an attempt to lower the ocular side effects, or are there other adverse events that might benefit in terms of lower rate from this less frequent dosing? Thanks.

The thought here is to take the opportunity to say, how do we get the best possible experience for the patient. Patients having a good experience, responses, in fact, the tolerability is good, we're not seeing any high rate of discontinuation. The eye effects are reversible and manageable. But if we can have less and retain this activity, we'd like to do that. So the concept behind the four-weekly was really more about recovery time. Based on the half-life of the product, which we've got around four to five days, you would expect to be, at the fourth week would just give you a little bit more freedom, if you like, from treatment, so that would enable the corneas a longer time to recover, in theory, so the hypothesis there is that may reduce the amount of ocular, i.e., reduce the amount of blurred vision. Clearly with the data that we've got, and the safety and tolerability, efficacy we've got, we would be more than prepared to go forward with the three-weekly schedule, but we just wanted to have that look to see if the four-weekly could provide us an even better. We're not going to give up on efficacy here. It's too important to us. So we'll take a look at that at the end of the first stage, and make our decision at that point.

Okay. Great. Thank you so much. And congratulations on the progress.

Thank you.

Thank you.

We'll go back to Simos Simeonidis with RBC Capital Markets.

Yes, a quick follow-up. On November 8 are we going to see data on the 17 patients, or could we see data on the 22, the five that were not evaluable, could we have enough information to see data on those?

Yes, there will be data on all 22.

Okay. Great. And finally, since nobody's asking Dave anything --

Come on. Even to wake up.

Can you tell us what the milestones from Lilly and Sanofi were for entering the clinic?

Okay. Sanofi was actually received in October, so that will be actually recorded in next quarter. So between, Lilly was $5 million for this quarter, for the first quarter, Amgen was $1 million, and we're looking for the Sanofi one to be recorded for next quarter. So you'll see that then. That was $2 million.

Okay. Great. Thank you very much.

Sure. Thanks.

That does conclude today's question-and-answer session. At this time, I'd like to turn the call back over to Carol Hausner for any additional or closing remarks.

Great. We'd like to thank you for your interest in ImmunoGen, and if you have any additional questions, please don't hesitate to call. Have a great day. Take care. Bye.

And this does conclude today's conference. Thank you for your participation.