ImmunoGen Inc (IMGN) 2014 Q4 法說會逐字稿

Good day, and welcome, everyone, to this ImmunoGen fourth-quarter fiscal year 2014 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. Good morning. At 6:30 this morning we issued a press release that summarizes our financial results for our fourth quarter and fiscal year ended June 30. I hope you have all had a chance to review it. If not, it is available on our website.

During today's call we will make forward-looking statements. Our actual results may differ materially from such statements and 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 more detail; and our Chief Financial Officer, Dave Johnston, will discuss our financial results and guidance. We will then open the call for questions. Dan.

Thanks, Carol, and good morning, everybody. In the few months since our last quarterly update promising clinical data have been reported for several compounds with ImmunoGen's technology. We've made solid progress advancing our wholly-owned product candidates and there has been notable partner progress beyond reported clinical data.

ImmunoGen is recognized for our leadership in ADCs with multiple major companies advancing innovative therapies with our technology. There are now nine novel agents in the clinic through our partnerships, recently expanded with Sanofi's initiation of clinical testing with the CDC, SAR408701 which targets CCaM, a target found on a variety of solid tumors including colorectal, lung and gastric cancer.

Sanofi now has three ADCs in the clinic with our technology as well as the CD38 targeting naked antibody therapeutic, SAR650984. We are pleased to see a growing number of compounds with our technology demonstrating that they can make a difference for patients. First with Kadcyla which showed favorable levels of objective responses in early clinical testing and improved progression free and overall survival with fewer tolerability issues in its lead phase III trial.

This important medicine is gaining regulatory approvals and becoming available to physicians and their patients in an increasing number of countries around the world. Five other uses of Kadcyla are now being evaluated in phase III trials with Roche projecting readout on the next one, MARIANNE, later this year.

The oral presentation on bio test BT-062 at ASH last December showed how this ADC was making a difference for patients with relapsed refractory multiple myeloma. It generated a high level of objective responses when used in combination with REVLIMID and dexamethasone even against disease refractory to these agents.

And in June the oral presentation on Sanofi's SAR3419 at ASCO showed how this ADC, a product of ImmunoGen research, was delivering high response rates in patients with a hard to treat disease, relapsed diffuse large B cell lymphoma in a phase II trial.

Initial evidence of activity has also been reported for our lead wholly-owned compounds, IMGN853 and IMGN529, during their dose finding phase.

We are investing to determine as quickly as possible the therapeutic potential of each of our wholly-owned compounds. This involves not only promptly enrolling patients in trials, but also conducting the trials in a manner that provides a fair assessment of each compound, getting the right patients, the right dose, the rate schedule and so on. You will hear more of this from Charlie in a moment.

Our financials and guidance reflect this investment. Importantly, we believe that we have sufficient financial resources to advance our current clinical compounds to proof of concept. Iterating such data should markedly expand our options for pursuing each compound's full potential. I will now ask Charlie to discuss our product programs in more detail. Charlie.

Thank you, Dan, and good morning, everybody. I will start with our most advanced compound which is IMGN853 whose target folate receptor alpha is highly expressed on many ovarian and endometrial cancers. We've established the recommended phase II dose for 853 when administered once every three weeks and are now beginning its assessment specifically for the treatment of platinum resistant ovarian cancer and relapsed refractory endometrial cancer.

As you will recall, we saw evidence of [simulated] activity with 853 in both of these cancer types during dose finding. Its activity includes confirmed partial responses, CA 125 responses and long-term disease stabilization. And were seen in patients [whose tumors] highly expressed 853 to target consistent with its mechanism of action.

In this initial evaluation of patients will be prescreened to ensure a high level target expression as we believe this will be an important factor in patient selection. Analyses we have conducted suggest that administering 853 more frequently than once every three weeks may in fact further enhance its benefit. Specifically these analyses supported the schedule that we are looking at, 853 weekly for three weeks and then providing a rest week before returning to weekly administration.

We began assessing this modified weekly schedule earlier this year and are still dose escalating. We plan to choose between this schedule and the three weekly schedule for the next stages of 853 testing. We expect to report findings from these assessments in a package in the first half of 2015 most likely at ASCO. In the interim we plan to provide updates on program status as this decision is made about dosing schedule and the status of patient enrollment.

Needless to say there is a significant need for new and better treatments for both ovarian and endometrial cancers. We recently shared the initial 853 data with opinion leaders in the field and they were impressed with this single agent activity. We now have orphan drug designation for 853 in ovarian cancer and we plan to apply for it for endometrial cancer.

I will turn now to our IMGN529 compound. As reported at ASCO this year, we've also already seen clinical activity with 529, this time in the treatment of non-Hodgkin's lymphoma. As you recall, 529 comprises our ADC technology used with an ImmunoGen antibody that pre-typically demonstrated potent anticancer activity even without a payload.

529 started showing activity in the clinic at the dose of 0.2 milligram per kilogram administered once every three weeks. That is well below the active doses of other ADCs of our technology indicating the activity is likely due to 529's antibody component not expecting getting the benefit of its payload at such low doses.

At the time of ASCO we were at the 1 milligram per kilogram dose level and continuing to dose escalate. We believe we are now approaching doses where both the antibody and the payload will contribute to activity. We intend to report data findings for 529 at ASH. Around that time we expect to be moving 529 into assessment in disease specific patient population. Of course whether that's the case depends on when we encounter the dose limiting toxicity needed to establish 529's maximum tolerated dose.

For IMGN289 dose escalation is also ongoing. IMGN289 you will remember targets EGFR. Once we have established its recommended phase II dose we plan to assess 289 specifically for the treatment of patients who have cancers that are high expressers of EGFR and have limited options today. Specifically this would include squamous cell head and neck cancers, both squamous and non-squamous non-small cell lung cancer.

It is hard to predict when we will encounter the dose limiting toxicity needed to establish maximum tolerated dose, but it could be some time in the next six months given the relatively high doses at which we started.

Behind these preclinical compounds are additional early-stage product candidates. We reported on the most advanced of these, IMGN779, at the European Hematology Association meeting in June. 779 targets CD33 and is a potential therapy for acute myeloid leukemia. Of particular note, this is the first ADC disclosed that uses one of our new DNA acting payload agents which we call IgMs and has quite an exciting profile preclinically.

With that I will turn the call over to Dave to discuss the financials.

Thanks, Charlie. As Carol noted, we issued a press release this morning with our fourth-quarter and fiscal year 2014 financial results. I will review the highlights and then provide our financial guidance for our fiscal year 2015.

For the fiscal year ended June 30, 2014 we reported a net loss of $71.4 million or $0.83 per share, in line with our net loss of $72.8 million or $0.87 per share for fiscal year 2013. Our net loss for the fourth quarter of 2014 was $26.5 million or $0.31 per share compared to a net loss of $21.9 million or $0.26 per share for the same quarter of last year.

Revenues in our fiscal year 2014 were $59.9 million as compared to $35.5 million in fiscal 2013 with most of the increase driven by license and milestone fees and from royalties. For example, revenues from license and milestone fees were $39.5 million in fiscal 2014 compared to $24.2 million last year, with the increase primarily due to the greater number of licenses taken by our partners, Novartis and Lilly, in fiscal 2014 and the associated amortization of their upfront fees previously received.

Royalties on sales of Kadcyla increased to $10.3 million in fiscal 2014 from $600,000 last year. Operating expenses in fiscal 2014 were $131.4 million compared to $108.5 million in the prior year. This consisted of $107 million of R&D expenses compared with $87.1 million in fiscal 2013 and $24.5 million in G&A expenses compared to $21.5 million last year.

Keep in mind that the 2014 R&D expenses included the $12.8 million non-cash charge related to technology rights received and to be received under the collaboration agreement we established with CytomX in January of 2014. R&D spending also reflects expenses related to increased staffing required as we advance our wholly-owned pipeline.

We ended our fiscal year with approximately $142.3 million in cash and marketable securities, compared with $195 million as of June 30, 2013, and we continue to have no debt. Cash used in operations was $53.7 million in fiscal 2014, down from $60.3 million last year. Capital expenditures totaled $8.2 million in fiscal 2014 compared with $3.8 million in fiscal 2013. So our 2014 results were in line or better than our previous financial guidance.

Let's turn to our 2015 guidance. For the fiscal year ending June 30, 2015 we project revenues to be between $100 million and $105 million. Recall that our revenue is driven primarily by: one, cash milestone payments from our partners; two, Kadcyla royalties; and three, recognition of previously deferred revenue, mainly amortization of prior license fees and upfront payments. And we expect to see meaningful growth in all three of these areas.

We project operating expenses to be between $160 million and $165 million. Operating expense growth next year is directly related to advancing our wholly-owned pipeline and reflects both current clinical trial activities as well as material preparation for later stage trials.

With that we expect net loss to be between $60 million and $65 million, cash used in operations to be between $55 million and $60 million, capital expenditures to total between $7 million and $9 million and we expect to end the fiscal year with between $75 million and $85 million in cash.

We believe our financial resources combined with expected cash inflows from royalties and other revenue sources is enough to fund us through proof of concept for our lead programs into fiscal year 2016. So with that let me turn the call back over to Dan.

Thanks, Dave. So I think you can see that ImmunoGen has a lot in play this year on a number of fronts. We look forward to providing updates on our wholly-owned programs as well as partner compounds over the course of this fiscal year as we believe the ability to help patients with these therapies will become much better defined for a number of these agents during that time.

Let me recap the various activities. I will start with our proprietary programs.

For IMGN853 we are now beginning evaluation in platinum resistant ovarian cancer and relapsed endometrial cancer patients with well-defined profiles of target expression, prior therapies, etc., dosed once every three weeks. At the same time we are evaluating more frequent dosing.

We plan to provide a comprehensive picture of this therapy when we next report data, most likely at ASCO next June, 2015. And we expect to disclose on quarterly calls major progress such as which dose schedule we have selected.

For IMGN529 we're continuing to dose escalate and we intend to report additional findings at ASH in December. The potential is there to have the recommended phase II dose around that time, that would start us moving into further evaluation in disease specific patient populations.

For IMGN289 we are also continuing to dose escalate. Potential is there to have the recommended phase II dose in the next six months and then to start further evaluation in disease specific populations once we have that phase II dose. For 289 we are targeting ASCO 2015 for the first presentation of clinical findings.

And finally on the proprietary side for IMGN779, we anticipate presentation of additional preclinical data at up-and-coming conferences, the first will probably be ASH, and then with the IND filing sometime in the back half of 2015.

On the partner front, from Roche we understand that the readout from the MARIANNE study is expected later this year. That would leave the data presentation in 2015 and, if the data supports it, a regulatory filing in 2015 as well.

Similarly for gastric the data would come out a little bit later in 2015 but, again, assuming positive data, the regulatory filing would then be in 2015. And obviously we'll see on a quarterly basis how the sales profile is developing for a CAD filing.

With our other partnered compounds we expect more insight into the development programs for several of these with a presentation of the next clinical data expected at ASH for one or more partner compounds and then we would expect to see further data at subsequent medical conferences.

Beyond clinical data we look for additional compounds to move to IND that would then lead to clinical starts over the course of this fiscal year and we do look to see several additional licenses being taken by partners over the course of 2015.

So it portends to be a very exciting year for ImmunoGen giving us the opportunity to bring forward advances for cancer patients over the course of this year and beyond. With that let me turn it to Carol and we will move to the Q&A phase of the call.

Thank you, Dan. Operator, we are now ready to open the call to questions. I would like to ask our analysts to limit their questions to one or two per person until everyone has had a chance to ask questions. You can get back into the queue. Thank you.

(Operator Instructions). Adnan Butt, RBC Capital Markets.

Congrats on the clinical progress. I will start with a partnered compound, 650984. The partner Sanofi seemed to imply that there could be a potential faster path. Would you be able to go over some hypothetical scenarios, how it could be accelerated, and what is the anti-CD38 landscape? Also what the status is and when data is next?

And then a second question on guidance for revenues. Dave, would you mind breaking it out? Are you giving royalty guidance as well? Thanks.

Good morning, Adnan, it is Dan. For CD38, as it is with most -- with all partnered compounds, we don't get ahead of our partners. I am not aware of -- there was extensive data on the 984 CD38 compound at ASCO. But I don't know that Sanofi has laid out a roadmap from a development standpoint for that.

Your reference to the CD38 landscape, I know that there certainly there's a compound that has received breakthrough designation that is a competitive compound, a naked antibody targeting CD38. And there is one other compound that is being developed. I don't know whether there is breakthrough designation for that or not.

So while I am aware that Sanofi is quite excited about it, it is prominent at ASCO, not only with the data but also at their booth, as I said, I don't know that they have necessarily laid a roadmap out there. On the revenue side I will let Dave respond to that.

At this point we are not really breaking it out by component. But I think that we are looking forward to significant growth in the royalty area as well as the milestone and amortization of prior (inaudible).

Dan, would you know if they?ve applied for breakthrough designation?

Not that I am aware of, Adnan. I think that would be something that -- companies take different approaches on, A, whether they announce that they apply or whether they announce that they have received. I wouldn't be able to respond outside of anything that they would have disclosed.

Thank you.

Brian Klein, Stifle.

First on 853, just wanted to get a better understanding of the goal of looking at the weekly dose versus the Q3 dose. And it seems that you might be implying that there is some potential for better efficacy, but I'm just wondering how that parlays with the side effect profile that you are seeing so far. And then secondly, in terms of the data that you might present at ASCO, will we also get a chance to see any PFS data on 853?

Charlie?

In terms of three weekly and the weekly it really comes from the PK modeling and we presented elements of this both at ACR and ASCO this year. And there are a couple of aspects to that.

The peak exposures in cycle one seem to have quite a strong correlation to the incidence of some of the ocular side effects -- adverse effect. So that would've led to a lot of the work that we did around adjusting the calculation as well as to helping us identify the dose that we have gone ahead with.

The other piece that we saw though was a relationship between total exposure that was as measured by ADC zero to infinity seem to be correlating with the initial evidence of activity that we were seeing. So obviously by going through or testing a weekly schedule seems to have at least a theoretical potential to address both sides of that equation.

Because, generally speaking, you would expect to end up with a lower weekly dose than a three weekly dose, so your peak, your Cmax, would be lower so that would theoretically lower the risk for adverse events. But your AE, you would expect your AUC to become higher and therefore have again the potential to increase efficacy.

Now obviously we don't have data on that at the moment, so these are sort of theoretical discussions. We like what we have seen with the three weekly which is why we are continuing to explore that. But we believe that it should give ourselves also and patients the opportunity to have the best of both world if indeed the data show that. But we don't -- I don't know that the dose limiting toxicity on a weekly schedule for example would be the same as on the three week schedule.

In terms of data at ASCO, we will provide at that point essentially what we have, which I know isn't a very comprehensive response, but what responses we have, the duration of those -- and I think duration of response is probably at this stage where you -- phase I or in a phase I expansion, I think duration of response is perhaps clinically more informative than CFS in an entire population. But certainly we will provide an update on all information that we have at that point.

Great, thanks. And then maybe just one strategy question for Dan. In terms of the Kadcyla royalty, is there any thought about monetizing that to further fund your internal pipeline?

Well that is one of a variety of avenues we have as things evolve, Brian. I don't put it as our highest priority. You heard Dave comment about our ability to advance the compounds that we have with our existing resources.

I think we will continue to look at what the needs will be, what avenues are there, whether there are alternative means of advancing that. So it is nice to have that available to us as a potential funding resource, but it just has to -- we have to continually look at where that might fit in the mix and what our needs are.

Great, thanks so much.

Thomas Wei, Jeffries.

I had a guidance question and then a question on the whole dosing schedule and stuff. So on guidance, you had mentioned, Dave, meaningful growth in all three components including deferred revenue. I can't remember exactly what your revenue recognition policy is like. Usually it is upfront payments that are being deferred. So does that mean that there is new deals being signed here and would that be on the proprietary pipeline side or is that more technology stuff?

And then on the dosing schedules I am curious -- for the compounds that are moving forward right now, like 289, are you doing this work in parallel to look at Q3 weak, more frequent dosing? Or is this something that will only -- like a two weekly schedule would come up only if there is an issue with the therapeutic window at a less frequent dosing schedule? Thanks.

Sure, Thomas, Dave here. So let's talk about the deferred revenue. So what we -- in terms of revenue recognition, oftentimes what happens is people will pay us an upfront license fee and what we will do -- and the agreement is for a specified number of potential licenses that they might take and they take them over time. And so as they take them we recognize some of that deferred revenue.

So when I say that there is growth, what we are anticipating is existing partnerships taking licenses that they are entitled to under existing agreements and at that time we would then recognize the revenue that was deferred at the time of the signing. I hope that makes sense to you. So what it doesn't imply is any new deals. What we don't do generally is include future business development activities in our guidance.

And from the dosing perspective, the work that we did with 853 was clearly in response to a finding that we saw. You remember there was variability in the exposure, there was some clear relationships and correlations between PK and both some of the tolerability and efficacy findings. And we continue to do similar work with all of our compounds.

Now 289 you mentioned, 289 -- we actually have introduced that with a weekly schedule based on the preclinical efficacy information and also on the historic use of EGFR antibodies. So that is already on a weekly schedule. And we're using, as we have historically, total body weight. But clearly if we see the same kind of variability and it suggests an adjusted ideal body weight would be a better way to go then we will do so.

But I think what we have got to do right now is really await the data. We will collect those data, we will model the data, the PK relationships and make those decisions as we go. But for each of the other compounds as it stands today, for 529 we are only currently studying a three weekly schedule and for 289 we are only currently studying a 289 schedule -- sorry, a weekly schedule. But we shouldn't rule out anything if it suggests that it would improve things from the patient experience perspective.

Hey, Thomas, Dave circling back on the first part of the question. Just to put things in kind of a perspective in terms of recognizing deferred revenue. As of the end of fiscal 2014 on the balance sheet we had almost $60 million of deferred revenue so that is potentially upfront payments, license fees that haven't been recognized to revenue. So those will be recognized over time as licenses are then taken.

Great, thanks.

Andrew Peters, UBS.

Congrats on the progress this quarter. A couple questions on 853. The first, you mentioned kind of feedback from KOLs following the data at ASCO. I was just curious, based on your conversations, if you think there is -- what the acceptable and if there is an acceptable level of the ocular tox kind of as the dosing work continues.

And then you mentioned kind of choosing which does going forward. Can you just elaborate that? It seems like that you are continue or you are starting to dose in patients or continuing to dose in patients an every three weeks dose. What are the decisions that you're going to be making? What are the sorts of parameters that you are looking at? And is there a possibility that you look at both kind of dosing profiles going forward?

Charlie?

Thanks, Andrew. I think in terms of an acceptable level of the ocular tox is sort of the double edge question because it will generally depend on the level of efficacy to some extent drive the level of toxicity that you are willing to see.

Obviously we were encouraged to look at a different dosing schedule based on the rate that we originally saw in the 5 milligram per kilogram where it was five patients had some degree of finding at that level, we wanted to bring that down and based on the information we presented at ASCO we appear to have done that. And we would certainly like to have the minimum amount of grade 3 and higher that we can.

So it is not a direct answer to the question, but I think obviously we feel that we need to get the balance right. The good news is it is a reversible effect, the levels that we are seeing at the moment are clearly different than when we were using total body weight and we are encouraged by the data that we have.

In terms of selecting a dosing schedule, it is really going to come down to again that balance of how does the therapeutic index appear to be, what is the balance of safety and efficacy. So we are proceeding with at least exploring initial cohorts for a three-week schedule because we like what we have seen and our investigators like what we have seen.

We believe that the three weekly is viable, but, as I responded before, there's data there that, if these PK relationships are true, suggest that the weekly could have other advantages. But clearly we don't today know what the effect of weekly is and we haven't tested that.

So as we move through I think we will certainly look at the efficacy and the tolerability of that weekly schedule as we proceed. And when we have enough data to feel that we can make a selection of one over the other we will do so.

Great, thanks.

Matthew Harrison, Morgan Stanley.

I have two pieces. So first on 853, can you just tell us what the dose that you're going to use in the phase II portion is? And then separately on the Sanofi compounds, I don't know in the past if you've mentioned, but can you tell us if the royalty rate on the naked antibody is different from the ADC compounds and sort of what that might be? Or can you expound upon the single -- the sort of percentage -- I think you have mentioned before single-digits I think is what you said but maybe you could clarify. Thanks.

Hi, Matthew, it is Charlie. We have not disclosed at this point what the dose that we are taking forward is, that would be a part of the package that we present later. But you will remember that we had looked at the 5 and 6 milligram per kilogram doses on the adjusted ideal body weight when we were at ASCO, so it sort of -- it narrows your range.

Matthew, it is Dan. On the royalty question, we noted in our press release this morning some information on this and let me take you through it. The royalty rate is not different than those on our ADC. So under the agreement Sanofi royalty was defined for compounds that were developed there and I think the guidance that we have given is it's a tiered royalty running from the mid- to high-single-digits.

The nuance for the CD38 compound is that that is the one naked antibody that was developed under that particular research agreement. And the reason that is important is that under the research agreement as ImmunoGen developed technology or developed IP an obligation was to assign that IP to Sanofi.

Why it is important for the CD38 compound is that doesn't fall under the umbrella of our ADC technology. And so for many elements of our ADC technology we have IP that would cover development, it may cover a composition of matter, for example, for the toxin, it may cover various process elements, it may cover precursors.

And so, we have a lot of IP that covers ADCs whether they are proprietary to us or to partners. When you have a naked antibody you only have one molecule. And so, despite the fact that ImmunoGen invented that particular antibody, we assign the composition of matter patent to Sanofi as part of that agreement.

Now, why is that important? It is not unusual under any of these license agreements for royalties to be reduced to the extent there is not IP that is wholly owned by one party to cover the molecule. And so, that then comes into play for that 984 compound with Sanofi because we assigned that composition of matter patent to them, they now own that and therefore the provisions of the agreement reduce the royalty that is available to ImmunoGen.

Again, it is unique because of the fact that it is a naked antibody, it has no impact on our milestones, but it does on a net basis, as we indicated this morning, take us down to the potential of a low-single-digit royalty on that particular molecule. Does that get at the issue for you?

Yes. No, thank you for the clarification.

Michael Schmidt, Leerink.

I have two. One on 289, I'm sorry, on 529. So at ASCO you presented some data on patients with treatment of steroids in addition to the drug. And I think you still saw neutropenia later in the cycle at the 1 milligram per kilogram dose. And so I was wondering where are you right now in the dose escalation? Are you beyond the 1 milligram dose? And how confident are you that neutropenia later in the cycle isn't limiting in terms of dose escalation?

And on 853, I guess at the ASCO data with the adjusted ideal body weight dosing there was grade 1, 2, ocular events as well at the 5 and 6 milligram dose. So I was just wondering in summary how confident you are that the adjusted ideal body weight method actually helps in terms of improving the therapeutic index of that drug? Thank you.

In terms of the neutropenia so, yes, steroids were introduced because of these unusual early drops in the neutrophil count which, because it was so early, was really very untypical for some of the cytotoxic like effects. So we hypothesized that was likely a cytokine related redistribution of the neutrophils and therefore to cover any early release of the -- sorry, of cytokines introduced to steroids just for a couple of days after dosing as well as to cover the potential for infusion reaction.

And that certainly seemed to have a positive effect and we were able to dose escalate further than we had before. We did then see some later changes in neutrophil count at a more typical point in the cycle, if you like, sort of around the date 8 to 10 as we got to that higher 1 milligram per kilogram dose. And we certainly believe that we can dose escalate beyond there given that we have the option and the use of growth factors which is Neupogen to keep us going.

So we certainly are progressing through dose escalation and are comfortable that we have neutropenia and the modern age tends to be a manageable affect. We will continue to manage it.

With 853, I mean the fact of the matter is of course if you dose high enough you will see the ocular effect. So it was never going to manage it all the way away. And what we were trying to do was to make the level more predictable so that you could bring patients in sort of a more targeted exposure.

I think if you recall the data from ASCO we certainly appear to have done that. The rate was lower, we had originally, as we mentioned earlier, a 5 milligram per kilogram ideal of total body weight so quite a broad range of exposures with some of the patients who were perhaps more overweight having ocular toxicity with 5 out of 10 patients. With the adjusted ideal body weight method it had been just one of the patients who had had a grade 1 effect.

So I think there's clearly something very different. We've clearly controlled that breath of exposure. But I think what we're able to do with this is to make this -- we are not going to -- it is not going to completely disappear or certainly not without going into what might be potentially therapeutic ranges. I think what we are able to do is to make this a more controllable and therefore manageable effect.

Let me at just maybe one or two things to that, I mean maybe said differently. What we have been able to do with the adjusted ideal body weight is address the interpatient variability that exists when we are using total body weight. And to add to that, Charlie referenced earlier and in his comments that we met with KOLs on this topic, and they understood very clearly what we did and actually were very complimentary about the approach that was taken to allow us to be more consistent in what we were delivering in terms of the therapy across the range of the patient population.

So we recognize that the dose limiting toxicity in this particular instance is eye tox, but what we have been able to do is to give us greater capability to control it across a range of patients and, therefore, to potentially get ourselves a better therapeutic index.

Okay, thank you.

Jason Kantor, Credit Suisse.

My question is on the guidance. The revenue guidance seems really quite high relative to what you've been doing the last couple of quarters. And I'm just wondering what specifically can we look forward to? I mean how much visibility do you actually have on this? Should we expect that a lot of that should be coming in in the first and second quarter of the fiscal year because you have visibility on some big milestones that you might be getting? Can you give us an example of what types of milestones you might be receiving over the next 12 months? Thanks.

Sure, Jason. So we don't break it out by quarter. But in terms of visibility, a lot of this is driven by existing partnerships that we have. So, for example, of the three components to it in terms of milestones, milestones are generally driven by clinical progress with our partners, and we have some insights as to where they are and where we would anticipate them to be.

So, for example, if there was a milestone on the start of either filing of an IND, the start of a Phase II, start of a Phase III, et cetera; we have some idea generally about when those would take place. So that provides us some confidence. In terms of the amortization of deferred revenue, those are generally, as I said earlier, associated with taking of licenses. And we have some idea there as well, working with these partners, what their anticipated timing is.

And then the last big component is the Kadcyla royalties. And that is based off of a relatively conservative outlook on projected Kadcyla revenues for the next -- what turns out to be our fiscal year, so the second half of 2014 into 2015.

Thank you.

Jason, I would just add, look, it is probably -- you are looking at quarterly. As you saw in fiscal 2014, it is probably going to be lumpy. These things don't come out smoothly. So it is really tough to be precise quarter by quarter, so that is why we are not going to forecast. I would like to make your model smoother, but it is going to be lumpy.

John Newman, Canaccord.

Just had a question on 853. I know you have gotten a lot of questions already, but can you give us a sense maybe qualitatively what would cause you to make the decision to go to a once-weekly dose versus once every three weeks?

And also in terms of the data next year at ASCO, what would you need to see with a once-every-three-week dose to take the program forward? Would you be comfortable just looking at, say, complete response rate? Would you need to see PFS? Just curious as to what you would look to see. Thanks.

So qualitatively between weekly and three-weekly, I think what we would be looking for is at least one of better looking tolerability or better looking accuracy, which is kind of obvious but I think important. I think weekly schedules are not unappealing to an ovarian cancer population. There is a lot of weekly Taxol being used. So I don't think if we switched to a weekly schedule that that would be any significant disadvantage relative to what we have today.

And clearly it is not every single week. There is a day one [815] out for a four-week schedule. So I think if it looks like we have similar accuracy but lower -- we are not getting into that range of exposures where we have seen the ocular findings, that would bring us over. And I think if we -- then when we have a sense, I mean it is not going to be a randomized comparative study here, but if there is a sense that we are getting high levels of activity then that would also make us consider the weekly.

And there may even be situations where you may have combinations where one is on a more intermittent schedule than another. So I think having data on both is never a bad thing anyway.

And I am completely blank on the second -- oh, the second part of the question -- the kind of levels of activity.

I think historically if you get north of 30% in ovarian cancer you are starting to have a very interesting drug, if you get north of 40% you have got a very, very interesting drug and perhaps one that you might want to take forward rapidly. But those are just -- but you want durability of those responses as well to bring things forward. So I think you need to look at the entirety of the picture.

Endometrial cancer -- the response rate for second line drugs today are really quite poor. The [amateur] -- if you don't cycle in the second line you are looking at 10% response rate. So even at 20% to 30% type response rates I think you have got an interesting drug. If you higher than 30% you've probably got a very interesting drug.

So obviously we will continue to look at that. Again, I think a response is one thing, a durable response would be very important as well. But I think we would take the totality of information, tolerability, response rate and durability of -- PFS before making a decision to invest solidly in a pivotal program.

But the good news from everybody's perspective of course is now that we have gone into the growth expansion portion of the ongoing study we are getting closer and closer to having that information and being able to make those decisions. So I think it is an important time for 853 in its development.

Great, thank you.

Cory Kasimov, JPMorgan.

This is Whitney on for Cory. I guess just following up on the revenue question, I'm just wondering if you can quantify at all maybe the revenue per license you expect to recognize or just give us any granularity there. And then I guess same question for some of the INDs you expect to be filed over the fiscal year, if you can just quantify milestones per IND.

No, first of all the amount of milestones that we would -- well in terms of revenue recognition let's do that first on a per license basis. Even if I could define it varies by the agreement depending on how many license that they have had and how many licenses -- so there is no one answer.

So for example, last year in 2014 we recognized about $30 million worth of deferred revenue. So I think that you will see that if you were to look at next year you will see growth from there. I think that is about as granular as we can move on that.

And in terms of milestones, milestones also vary depending on if it is a filing an IND, filing a -- or various phases of the clinical progress. So it is really hard to provide any sort of firm guidance there. And I don't think that we'd feel comfortable doing it anyway.

Got it, thanks.

Mara Goldstein, Cantor Fitzgerald. And Mara has disconnected. At this time we have no further questions.

So I would like to thank everyone for their interest in ImmunoGen and if you have any additional questions don't hesitate to give us a call. Have a good day and a good weekend. Take care.

This concludes today's conference. Thank you for your participation.