ImmunoGen Inc (IMGN) 2014 Q3 法說會逐字稿

Good day, and welcome everyone to the ImmunoGen third 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 thatsummarizes our financial results for our quarter ended March 31st, 2014,the third quarter of our 2014 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 at ImmunoGen are included in our SEC filings which also can be accessed through our website. In our call today are our Chief Executive Officer, Dan Junius, who will provide an update on ImmunoGen, our Chief Development Officer, Dr. Charlie Morris, will discuss our wholly-owned compounds in more depth. and our Chief Financial Officer, Dave Johnstone, will discuss our financial results and guidance. Then Dan will review our anticipated events, and we'll open the call to questions. Dan.

Thank you Carol. Good morning everybody. We're making important progress on our three wholly-owned compounds and clinical testing on IMGN853, IMGN529, and IMGN289. Here we reported interesting and encouraging data from IMGN853 at AACR. You'll hear more about that from Charlie in just a moment. We'll have additional data at ASCO on IMGN853, plus the first clinical data from IMGN529. At the same time we have also advanced through a number of dose escalations with IMGN289. We expect to report the first findings on this compound in the back half of 2014. While this is going on we're also making good progress with our partners, I'm sure you are all aware of Kadcyla, where we are seeing good sales development and launches taking place in additional countries, and progress towards gaining additional indications.

We also have encouraging clinical data for Amgen's AMG595 reported at an oral presentation at AACR. Data on two compounds we have developed for Sanofi, SAR3419 and SAR566658 will both have data presentations at ASCO. And Biotest has expanded the BT062 clinical program to include assessments in two types of solid tumors. While this is going on, we continue to expand our technology portfolio to enhance our leadership in ADCs. I'll come back to you on the partner activity and the technology, but now I would ask Charlie to discuss our wholly-owned compounds. Charlie.

Thanks Dan. Good morning everybody. I'll start with IMGN853 and the data that we recently reported at the AACR. As you know IMGN853 is our folate receptor alpha targeting APC for a potential treatment for variant endometrial and certain lung and other cancers that highly expresses target. We have reported to you previously that during our ongoing Phase 1 dose escalation, we started seeing activity with IMGN853 at doses 3.3 milligram per kilogram and higher, with that does being given after 3 weeks, and then we encountered dose limiting toxicity at 7 milligrams per kilogram the DLT being a type of reversible ocular toxicity.

We also disclosed that based on analysis we conducted we obtained a study protocol to using adjusted ideal body weight, rather than total body weight in determining patient size for dosing calculations, and that we would report the data behind this change to AACR. What we reported to AACR was the data showing how we made this decision, and the initial impact from a tolerability perspective. When we compared the data from the first cycle of treatment for patients with and without ocular findings, we observed that ocular toxicity occurred in those patients whose IMGN853 [placer] levels exceeded specific thresholds. This means that a maximum concentration of C max just shy of 150 micrograms per milliliter and an area under the curve over the first 24 hours post-dosing of about 2,800 of the relevant units.

We also found that when dose was based on a patient's total body weight, there was a marked variability between patients and their IMGN853 blood levels, even though they were dosed at the same dose level. As a result an overweight patient would be more likely to experience ocular symptoms, than a patient of the same weight who happens to be taller. Having seen this, we used our clinical PK data to model over measures, a patient's size, including body surface area and adjusted ideal body weight. This modeling suggested that dosing according to an adjusted body weight approach should have the greatest impact on reducing that PK variability among patients. This in turn will be predicted to reduce the variability in both the C max and the Day 1 area under the curve, which would mean that we could keep more patients below the toxicity threshold.

At the time of AACR, we have treated 6 patients with a dose of 5 milligrams per kilogram using adjusted ideal body weight, and followed them for longer than the time at which we had previously seen ocular symptoms occur. The PK was consistent with our models, and sure enough none of them had ocular symptoms. In contrast to our previous experience of dosing according to total body weight, ocular changes were seen in 4 of the 10 patients receiving 5 milligram per kilogram. So early data but promising.

We also modeled alternative dosing schedules. Weekly, every two weeks and so on, to identify the schedule that would give patients the most exposure while keeping the peaks behold those thresholds that we just described. We identified that dosing weekly for 3 weeks out of every 4 should achieve the subjective, and have now added a cohort to our current trials to evaluate the schedule, and those cohorts are now recruiting. We're encouraged where we are with 853. We will report more data for the once every three week schedule at ASCO.

The next step to this schedule will be to begin the expansion phase, which we expect to occur this quarter. This will provide us information on the efficacy and safety in patients with platinum resistant ovarian cancer or endometrial cancer. This Phase will have more restricted patient enrollment criteria, including entry requirements for levels of target expression and the number of prior therapies. Once we have data on the modified weekly schedule, and decide whether or not that is likely to provide any additionally advantage, we will also look at lung adenocarcinoma with this schedule.

Now onto our IMGN529. You will remember this is our CD37 ADC for B-cell malignancies, there is another reminder, it is the same linker, same payload as Kadcyla. In addition, the antibody is designed to enhance efficacy having cell-killing properties of its own. This product is currently in Phase 1 testing for non-Hodgkin's lymphoma. We are please to have had an abstract accepted as ASCO, it will be a poster which has been selected for the highlight section. This presentation will cover what we've seen in the clinic to date, from both the safety and activity perspective, and we will clarify the modification to the administration strategy that we had previously referenced and have now implemented. As we are still in dose escalation, we don't expect to have MTD information at that time.

Our third clinical stage asset is IMGN289. The EGFR targeting ADC has the same linker and payload as Kadcyla, and this antibody once again is expected to contribute to efficacy. We have made good progress in the clinic to date. We've been able to advance the number of dose escalations since we started patient dosing in November, and believe that we're now at or around the anticipated threshold for clinical activity. We expect to report the first clinical data later this year, and are optimistic that expansion cohorts should begin accruing patients in the second half of this year.

With that, I will hand back to Dan, who is going to discuss recent progress with our partner compounds, and with our technology platform. Dan.

Thanks Charlie. Let me start with BTO62 as we have rights here to opt-in on codevelopment and co-commercialization of this compound with Biotest in the US. As a result we look at this somewhat differently than how we view our other partner compounds. The compound itself is a CD138 targeting ADC, the antibody itself is a targeting antibody as opposed to a potentially therapeutic antibody. The compound was initially being developed as a treatment for multiple myeloma. There was data at ASH that was very compelling in terms of its efficacy when dosed in combination with Revlimid and dexamethasone, this trial is ongoing. It's now also being evaluated for solid tumors. There's a study that's initiated in Europe for patients with bladder cancer and triple negative breast cancer. This is a monotherapy dose escalation study. It was initiated based on some very impressive preclinical data, and we look forward to seeing how that study evolves.

Turning to Kadcyla, last week Roche reported Q1 sales of 102 million Suisse francs, or approximately $115 million. This compares with sales of $250 million for all of 2013. Recall that Kadcyla was first approved in the US in February of last year. From the first quarter sales that were reported, the US continues to account for the bulk of these sales, Roche has indicated that they think they're at about a 43% market penetration in the second line and later markets in the US. But they're also starting to see increased activity ex-US. Much of this comes from the EU approval that came out late last year. These sales are principally in Germany at this point, where government approval is not required prior to launch for pricing and reimbursement. There was also a launch last week in Japan, another large market, so we would expect to see that reflected in sales as we move forward.

With Kadcyla there are a number of data presentations expected at ASCO. I think this is a natural evolution with an approved product, because there will be a number of physician sponsored studies that will have data on potential other applications of Kadcyla. In terms of Roche sponsored studies that will lead to registration filings, the next Phase 3 findings that we would expect would be from MARIANNE in the second half of this year. I'm sure you're all aware this is for first line HER2-Positive metastatic breast cancer patients, data later this year would lead to filing in 2015 according to Roche. There are multiple additional registration studies behind that. The GATSBY study continues for patients with advanced HER2-Positive gastric cancer, data and a registration filing would be expected here in 2015. There are two registration studies already underway in early HER2-Positive breast cancer, that would be KAITLIN in the adjuvant setting, and KATHERINE for residual invasive disease, there's also a third study in early HER2-Positive breast cancer, this is CHRISTINE in neo-adjuvant, which is scheduled to start this quarter.

Other clinical compounds with our partners, let me start with Amgen's AMG595, this is an ADC that targets the EGFRvIII. It's currently being evaluated for recurrent glioblastoma. This is a disease with high unmet need. There are very limited options available for these patients today. The first clinical findings were reported earlier this month in an oral presentation at AACR, and they were actually quite encouraging. It was data from the dose escalation stage across 24 patients, and it being dose escalation, I'm sure you appreciate that these patients would have gotten varying doses, different levels of target expression, different profile of prior therapies, and the like. But that said, there were three patients who had meaningful stable disease, there was one patient with a durable PR, and there was one patient with a near complete response that as of the presentation was ongoing for over 90 weeks, and I was told that the presenter indicated that's a relatively rare event in glioblastoma, and seemed quite encouraged by it. They did encounter their dose limiting toxicity with thrombocytopenia, which is consistent with what was seen with Kadcyla, given that AMG 595 has the same linker payload construct as Kadcyla, that's not entirely a surprising event to us.

In terms of other partner compounds, there are oral sessions for SAR3419, and SAR650984, both of those Sanofi compounds that we will be hearing at ASCO. Recall SAR3419 is a CD19 targeting ADC. What we'll hear at ASCO in the oral abstract session is data from the Phase 2 STARLIGHT trial. This is a trial that assesses 3419 as monotherapy for relapsed refractory DLBCL patients. For 650984, this is a CD38 targeting naked antibodies, so this is not an ADC, a naked antibody that's being evaluated in multiple myeloma. Data from a trial assessing it used in combination with Revlimid/dexamethasone was another study that's been selected for presentation in an oral abstract session.

Beyond that, for 984 there will also be a poster with updated data from the monotherapy Phase 1 trial. There was an oral presentation on this at ASH that again showed very encouraging data when they were dosing 984 for advanced patients with multiple myeloma.

Beyond those partnered, so that is upcoming partner activity, we also have been active, and had some data at ACR around our technology portfolio. Recall there are 3 components to an ACD, the antibody, the payload, and the linker. We think that our antibody expertise is expressed in a number of compounds, but maybe most clearly with 984, it being a naked antibody, that demonstrates our ability to generate antibodies, as well as our expertise in target selection.

We have disclosed previously that we developed a platform of DNA acting payload agents, and this is the complement to our existing platform of 2 billion inhibiting agents. At AACR we had additional data, and this was on the lead agent of this new platform, something that we're calling DGN462. In preclinical models it achieved the desired efficacy, while avoiding delayed toxicity, the data also illustrates the importance of the properties of the linker. The ADC constructs that we made with DGN462 and a cleavable linker had a wider therapeutic index, than those made with a non-cleavable linker.

So that again speaks to the importance of linkers as we looked at the overall construct, and it also leads to additional data that we reported at AACR on another new linker that we have added to our portfolio. This is a linker that we put in the cleavable category. Remember that we categorize our linkers as being cleavable and non-cleavable, and with the specific properties here is when the metabolite with the payload is cleaved inside of the cancer cell, that's the activity of releasing the payload in a form that's even more active with this particular linker, in animal models than those formed with our earlier existing cleavable linkers. So we think that's quite important.

At the same time that it was showing this higher level of activity, again preclinically it suggests a favorable tolerability, the same tolerability profile that we see with our earlier linkers. It may again extend the therapeutic index. I'm sure we'll have a chance to come back to some of this in the Q&A, but in the meantime, let me turn it over to Dave to cover our financial results.

Thanks Dan. Since we issued a press release this morning that outlines in some detail our third quarter fiscal 2014 results, let me just review the highlights, and then we'll discuss our updated financial guidance. Revenue for our third quarter was $6.9 million compared with $25 million in the same quarter of last year, keep in mind revenues in the prior year quarter included $11.1 million in amortization of upfront fees from Novartis, which were recognized in that quarter due to Novartis taking a license under our 2010 agreement with them. Novartis has taken three such licenses under this agreement, with a portion of the upfront fees recognized as revenue each time one of those licenses is taken. Prior year revenues also included a $10.5 million milestone payment earned from Roche, with the approval of Kadcyla in the US in February of 2013. Needless to say this was a non-recurring event, but marked the start of ImmunoGen beginning to receive royalty revenues from Roche on Kadcyla sales. For example, the current quarter revenues include $2.6 million in royalty revenue on Kadcyla sales made during the quarter ending December 31, 2013.

Operating expenses were $44.3 million, compared with $26.3 million in the same quarter of last year. One big item driving the year-on-year difference is a $12.8 million noncash charge to R&D expense, related to technology rights received, and to be received under the collaboration agreement we established with CytomX in January. The balance of the increase in spending reflects expanded clinical program activity, net of our 901 program, and an increase in manufacturing batches that were released to partners.

As a result, we reported a net loss of $37.5 million, or $0.44 per share, for the third quarter. And this compares with a net loss of $1.4 million, or $0.02 per share for the same quarter 2013 fiscal year.

We ended the quarter with $164 million of cash and marketable securities. As mentioned earlier we have updated our guidance for the 2014 fiscal year, which ends June 30, 2014. We now project revenue to be $60 million to $64 million, versus prior guidance of $71 million to $75 million. This reflects changes in projected timing of some partner related events. We now project our operating expenses to be $133 million to $137 million, compared to previous guidance of $140 million to $144 million. This is principally due to the timing of spending on some of our preclinical and clinical programs, partially offset by the noncash charge related to our CytomX collaboration. Taking these two together, we project our net loss to be $71 million to $75 million, compared to our prior projection of $67 million to $71 million.

However, we project our net cash used in operations to be $56 million to $60 million, which is lower than our previous guidance of between $64 million and $68 million. Our guidance related to capital expenditures is unchanged at $8 million to $10 million, we are now projecting ending our fiscal year with more cash than previously projected. We now expect to end the year with between $134 million and $138 million in cash and marketable securities, compared with between $124 million and $128 million as previously projected. We believe that our cash position combined with expected cash flows from royalties and other revenue sources is enough to fund us at least through our 2015 fiscal year, as we continue to invest in the development of our proprietary product pipeline.

With that, let me turn the call back over to Dan.

Thanks Dave. Let me run through the events upcoming over the balance of 2014, and then we'll be happy to take your questions. It's starting out as an active year and projects to be a very active year for ImmunoGen, and an exciting one because across our various development programs, we're at a stage where we're going to be getting the data that we need to understand what the potential is for these particular compounds. At the same time across the full spectrum of partner compounds, we're getting again into an active data generation mode. We're seeing from the license that we're taking over the past certainly 2013, those may lead to new compounds coming into the clinic, we think there's the opportunity for new IND filings. So a very, very active time for the Company. I think it is quite exciting for us.

Specifically on the ImmunoGen compounds for 85,3 as noted we'll have clinical data at ASCO, that data actually will be on May 31st, for those of you that are starting to build your calendars. We look to start the expansion Phase for 853 later this quarter, as Charlie said we're enrolling patients under two different dosing regimens, so that will give us some further insight, and we will have updated clinical data on 853 in the second half of 2014.

For IMGN529 we'll have our first clinical data on ASCO on May 30th, and the expansion phase on that compound should start sometime in 2014. For 289 we're planning to have our first clinical data in the second half of this year, and sometime over the course of this year once we have determined MTD, we'll then move into expansion cohorts for that particular compound. With partners Kadcyla we'll learn about sales development on a quarterly basis. We'll get insight into further potential for label expansion with MARIANNE data that we should hear about in the second half of this year.

For the seven other partner compounds in this clinic, we'll see quite a bit of data from Sanofi at ASCO, as I noted we'll see the Phase 2 single agent data on SAR3419, and an oral presentation on June 1st. For the CD38 compound 650984 from Sanofi in combination with Rev/Dex, we will hear about that again in an oral presentation on June 2nd. And then we'll see some updated data as a single agent. I beg your pardon, the oral presentation is the data as a single agent which would be on May 30th. The combo data would be a poster presentation.

There's also potential for data coming out from our partners across the other major conferences over the balance of the year, ESMO, EORTC, et cetera. Beyond those compounds, with clinical data we would hope to gain some insights in our partner's plans for development over the course of this year, as I noted there should be the potential for additional IND filings, and insight into new compounds with preclinical data, and then additional target licenses going out to partners over the course of the year. So a very active 2014 for ImmunoGen.

With that, I'll turn it over to Carol, and we would be happy to take your questions.

ThanksDan. We're about to open the call to questions. I would like to ask folks to limit their questions to one to two per person, until everyone has had a chance to ask their questions. You can then get back in the queue. Operator, we are now ready to open the line for questions.

Thank you. (Operator Instructions). We'll now go to Adnan Butt from RBC Capital Markets.

Good morning, thanks for taking my question. Two here please. First on 289, how compelling is the evidence that there's less in tox, and how much discontinuation is there with current drugs and does the EFR expression pattern change? That is on 289. And just a general technology question, what is CytomX technology bringing in that ImmunoGen maybe didn't have before?

I'll take the first part of that, and I may have to ask you to repeat the second part of the question. As things stand to date, the optimism around skin toxicity clearly comes from our preclinical information, in those preclinical studies the ability of the antibody and therefore the ADC to kill keratin sites, was much less than we have seen with currently marketed agents, including cetuximab and panitumumab, and there is also this very different release profile of cytokines in those same models, so those cytokines contributing significantly to the chronic rash past of the toxicity, which has been traditionally seen. As we have described before, we believe that this relates to a partial antagonist, to only a partial antagonist activity, which appears to allow enough signaling for skin cells to be able to resist some of the effects of the antibody, but enough antagonist activity for that to contribute to efficacy, as well to as deliver the payload. I sort of missed part of the question about the EGFR patterns. Could you--?

Sure Charlie. First generally how much discontinuation is there with, due to skin toxicity at this time with current drugs, and then does the EGFR expression pattern change with treatment?

Obviously we haven't got any data yet, as we mentioned on the call we intent to present data at the end of the year. I think I will leave to your interpretation the fact that we said that we have escalated through a number of dose escalations having only begun dosing in November. So at this point, we have not seen anything which would change our belief in those preclinical data.

Sure. The question was about current therapies that are already on the market?

Well, as I said, I mean I think the primary aim I think here is thinking with the pattern of skin toxicity. We are all very much aware of the chronic rash and the acne form aspects of the rash which has been seen with the antibodies in the TKIs. Not only that of course, but we are aware that we might run the risk of delivering to the same skin cells a toxic payload in the maytansinoid, but we believe that the EGFR expression is very different in skin, it tends to be lower antigen density per cell than it is on tumor cells, particularly if we look at things like squamous cell carcinomas, where you see very, very high expression of EGFR, similar to what has been described in breast cancer with HER2 and 3-Positive range, that Kadcyla is addressing.

So I believe if we can avoid the known skin toxicity around the rash, that will enable us to dose at levels to have impact of both the antibody and to deliver sufficient maytansinoid for cell kill. With the likelihood, though we have got to see this yet in the clinic, that that would not cause any additional or different types of skin symptoms, because we do not believe that there would be enough entry of maytansinoid in the skin cells to cause additional damage. And I guess it's a similar construct thought to historically we talked about people concerned about the expression of HER-2 on cardiac cells with Kadcyla. We have not seen reports of cardiac toxicity with Kadcyla. So we believe that there is precedent for believing that we can see this. And that's what we're testing right now, and nothing to date has given us cause for pause in terms of dose escalation.

So thanks Charlie. Carol, I had a question on CytomX,but I can come back to that later, unless you just let me have that?

I'll get to that now since it's on the table, Adnan. In terms of what does the CytomX technology potentially do for us, I see it in a couple of areas. One, for those targets that we may have chosen to avoid with our existing technology, our tubulin inhibitors, because of expression on healthy tissue that was potentially problematic, this would allow us to pursue those targetswith our existing technology. As I noted in the prepared comments, we also are moving along with our technology with the DNA acting agent, and here there may be even a more intense application of CytomX technology.

For everyone's benefit, recall the CytomX technology is the ability to mask the binding region of an antibody, and the mask is released only in a cancer cell environment due to some proteases that are released extracellular with the cancer cell. So you have a potential very specific application or targeting with this masked antibody using CytomX technology. And given a DNA acting agent is less discriminate in terms of cells that it is going to kill than a tubulin inhibitor,it may afford us opportunities to look at diseases that go at it with a more aggressive cytotoxin than we have with our tubulin agent, because the DNA acting agent is more potent than our existing portfolio. And again, may let us go at targets where there is expression similarly to what I referenced with tubulin inhibitors, where there is expression on healthy tissue as well. It's very interesting. We've been impressed with the preclinical data. We're working intensely with CytomX on the elements of the cross-license, to bring compounds forward both on our behalf and on theirs, and we'll see where it goes.

Okay. Thank you.

We'll now move on to Simos Simeonidis from Cowen and Company.

Good morning, thank you for taking the questions. Question for Charlie, first. You talk about the 853, newest data we have seen. Can you give us an idea how the efficacy and safety profile has improved now, may compare to Endocyte's newly approved drug in ovarian cancer in Europe?

I think it's really quite early for that kind of comparison. Clearly they have completed some much larger studies. As we have said, we remain in dose escalation I think we will get much more detailed information on efficacy as we begin to progress through the dose expansion cohort. I think one of the critical things to think about though is the size of the populations that we're addressing, because those clearly, although we're not sure quite how many of the patients which are addressed by having the maximum sort of effect on the Endocyte test, are overlapped with the higher expressions of the folate receptor. What we do know is that there approval is in a pretty narrow population, I think it's about 10% to 15% of the ovarian cancer population. In our hands the expression 3-plus, in that particular ovarian cancer group is about 50% to 60%. I think we are looking to address a broader population.

We're obviously very encouraged that we've seen as we previously reported single agent activity. And I think you'll see a little bit more about relationships between pharmacokinetics and activity as we get to ASCO. And that I think will also further clarify some of the decisions that we've remained around trying to get the dose right. So I think it's early for that type of commentary. It's good for them that they have that European approval, but I think right now the overlap is relatively small. And doesn't really change any of our plans for pressing forward with 853.

Thank you. Great, and a second one for Dan. Dan, Kadcyla certainly got out of the gate with a great start for the first couple of quarters. The last three quarters, even though we've seen good progress ex-US, in the US the growth has been less than the first few quarters, the last three quarters growth was I think Q-over-Q 14%. It was negative last quarter, and this quarter it was 11% approximately in the US. Are you concerned that there might be a flattening or plateauing of Kadcyla sales in the US?

Well, I'm not. I think you have to go back to the fact that all of the feedback that we hear, and not that we're out surveying docs on a regular basis, but the feedback from the docs is that this is a great drug. They like the fact that it has a very beneficial impact on patients, while being minimally disruptive to their lives. We still are in relatively early days here with a somewhat narrow indication or narrow label, I should say, relative to its potential. So I think that there is the tendency to scrutinize very early on and microscopically examined quarter-to-quarter, but I think you have to give this a little bit of time to get a real better understanding of its potential. I think the MARIANNE data is going to be very important. We'll see how expansion goes in Europe and other jurisdictions, but the potential for Kadcyla is so significant that I think for people to be expressing concern, given its therapeutic profile, it just seems to be misplaced.

Great. Thank you very much.

We'll now go to Cory Kasimov from JPMorgan.

Hi. This is actually Whitney on for Cory this morning. Two quick questions. First on the revenue guidance, can you give us any more granularity on the changes in partner activities that you mentioned, whether what they are, or maybe what's driving some of those changes?

Sure. I'm not going to say who the partners are, but the nature of the change is anticipated activities, in terms of either taking a license or reaching a milestone, a clinical milestone, that sort of thing. So these are things that we don't anticipate going away. They're just the timing of them has slipped out from anticipated being in the latter part of our Q4 fiscal year into the early part of our 2015 fiscal year.

Got it. And then in terms of the updates on the proprietary programs at ASCO, can you give us any more detail in terms of what we should be expecting on patient numbers, or cycles of therapy duration, anything like that?

I think ASCO is about six weeks away. I think you'll get more information at that point. For 529, it's really about where we are in terms of dose escalation and an update on things that we have previously indicated, in terms of the initial both signals of activity, but some other effects that we had to overcome. And with 853 the focus is really going to be again on PK relationships to the activity that we've seen to date. So I think it will all become clear hopefully as we finalize those ASCO posters.

Got it. Thanks for taking the questions.

With our next question we'll move to John Newman from Canaccord.

Hi guys, good morning, thanks for taking my question. It's a two-part question. The first one is if you can comment at all, obviously you can't promote here but if you can comment at all qualitatively as to whether or not you're seeing off label use of Kadcyla in front line HER2-Positive metastatic breast cancer? And also in terms of the MARIANNE study, I'm curious as to what the trial protocol allows for, in terms of dose reduction in the arm containing two biologic agents, and which one of the agents can be reduced, or if both can be reduced, and what not? Thank you.

It's a question that's probably more appropriate directed to Roche given that in terms of any off label use outside of the second line and later in metastatic that would be something that they would see. I guess I would just say anecdotally my understanding is that some of that given reimbursement is getting tougher and tougher. And in terms of the MARIANNE protocol, again that's under the control of Roche and we wouldn't have any comment on that.

I can send you what they have reported, but that would come from them. So I can send that to you.

Okay. Thanks.

We'll now move on to Mara Goldstein from Cantor Fitzgerald.

Hi. Can you hear me?

Yes.

Great. Thanks very much for taking the question. I just had a question on 853. I think as I go back to probably around the R&D meeting, where there was a discussion around activity levels for 853, and looking for possibly a 20% OR for continuation in ovarian cancer, and I'm wondering if given the change in the dosing, and what you're seeing if that is still the case?

I think our thresholds for determining what we would perceive to be proof of concept are not really going to change, I think historically we have tended to suggest that something north of 30% is likely to be of significant interest for further progressing. I think that would remain true. The aim of changing the dosing is really to maximize that opportunity for all patients. The variability that we saw dosing on a total body weight basis meant that some patients that were unable to continue with the dose which they had been assigned, we think more patients will be able to continue at that dose. So none of the ambition around the product changes, in fact it gets enhanced. Now the intent here is to have as few patients as possible, as reasonably possible with the ocular symptoms, but as many patient as possible above thresholds for activity. And I think it's been a purposeful and to date worthwhile process of trying to get it right. But none of the strategic intent has changed by what we're doing.

Okay thank you. And if I could just ask a quick financial question just to confirm that change in the revenue guidance is all cash as opposed to just recognition of cash?

No.

Okay.

The change in revenue guidance is, for example when you recognize licenses, it's simply amortization of upfront cash that we have already received.

Right. So how much then of the change in guidance, is all of the change in guidance essentially recognition of cash already received, or is it a portion of it?

It's a portion of it.

Okay. Are you able to share what that is?

No. We're not getting that specific because, but I think what you can determine out of that , is the fact that our cash usage is actually down. And the fact that our ending cash balance is also up.

Okay.

And the only really noncash expense we have added is the $12.8 million for CytomX. It's a fairly easy math exercise to determine that the majority of it is noncash revenue.

Okay. Thank you.

Sure.

And with the next question we'll now move to Joel Sendek from Stifel.

Thanks. I just have a couple of housekeeping things. Just to follow-up from the last question, if we back out the CytomX piece from the R&D number, would that be the kind of annualized run rate that you're on? As far as actually cash R&D spend?

Well, I would say the short answer is it depends. And it depends on sort of the clinical success that we see in our expansion cohorts with our three internal products. As you can imagine, if we see activity, then that's going to lead to further larger trials, and that leads to more clinical expense. In terms of the pure research side I think it is pretty much annualized.

Okay. And as far as the, I know you guys had a ton of data at AACR, so I'm wondering when the abstracts for ASCO come out, I know you're working on the posters now, but when the abstracts come out is there a possibility that we'll see any additional data, or is it going to be old stuff, as far as when you submitted those abstracts as far as the proprietary compounds?

I think what we want to do with the presentations is try to have the most up to date information that we can, bearing in mind that involves a certain amount of work to bring data in, and get it cleaned up, and everything else. So I think you should expect to see things more updated than appear in the abstract, though obviously there's always a slight inability to make absolutely real time. But we're trying to keep people as up to date as we can.

Just to clarify, I was wondering if what is in the abstracts themselves will be old relative to AACR?

Well, I think the answer is, it would be different, because the abstracts for ASCO really are intended to address different questions than the ones at AACR. So 529 for example we haven't previously presented any data. And we consciously, are obviously working with our investigators in the abstract office, try to address different questions with the AACR poster, than the ones that we're addressing with the ASCO poster. So I think it's different data, rather than sort of newer or older data.

Great. Thanks for the clarification.

And our next question will come from Michael Schmidt with Leerink Partners.

Actually this is Jonathan Chang, stepping in for Michael. Thanks for taking the questions. My first question is with regards to PD 062. Could you walk us through the rationale for testing 062 in triple negative breast cancer and urinary bladder cancer? And then maybe remind us what the preclinical evidence was for pursuing these specific indications? And then maybe more big picture, tell us what the time lines and catalysts are for 062?

Jonathan, it's Dan. Again, this is being advanced by a partner, so it's partner data versus ours, but some preclinical data, Carol, do you recall when that was published?

I want to say early last year.

So ACR a year ago was when the preclinical data was published around I think it was four different solid tumor indications. So it was bladder cancer, it was prostate, it was triple negative breast, and it was pancreatic. And we can point you to that afterwards, if you haven't had access to that data, but I think what you'll find is, and again it's preclinical, but it was quite compelling the suggestion of activity being seen in each one of those indications. So that's the underlying rationale, because the initiative to move then off of that clinical data into clinical studies was based on that. And the process took them until early this year to get those studies established and begin dosing patients.

With that, your question around time line, it's just going to be a function of dose escalation, expansion cohort, and what the data shows. And it's just too early since they're only dose escalating. The only caveat I'll give is that since they do have monotherapy data in multiple myeloma, it gives them some advantage with respect to starting dose, so you don't necessarily have all of the early low doses before you're getting into a therapeutic range, so that might compress the time line a little bit to move through a Phase 1 study in those indications.

Okay. Thanks. And just a second question on your DNA alkoloiding payload, DGN462. If you could just generally compare that to a tubulin binding payload, what are the differences and the advantages of each? That would be helpful. Thanks.

There's maybe two, I'll focus on two primary differences. The tubulin inhibiting agent will kill a cancer cell, or disrupt the cancer cell that will lead to cell death by preventing the cell from dividing. So it has I'll call it a discrete mechanism, that is only applicable in cells when they're undergoing cell division. That's important because you can look at targets as you're evaluating applicability for the technology, you can look at targets where there may be some degree of expression on healthy tissue, provided that either the level of expression and/or the nature of the cell is such that it's not going to be dividing. And you can accept that level of expression. The great example for that is HER-2. HER-2 is expressed on a variety of healthy tissue outside of breast cancer where it's applied with Kadcyla, but you've not seen toxicities, not seen issues with healthy tissue of any meaningful level with Kadcyla despite that expression of HER-2.

When you move over then and look at our DGN 42, there what you have is an indiscriminate killer in terms of a cell killer. So you don't have the luxury of looking at expression across a range of healthy and cancerous cells, and say I can apply this and not have the same degree of concern about healthy tissue expression. You would go into it with the assumption that you're going to adversely affect healthy tissue, and so therefore the relative degree of expression is going to play much more prominently in target selection. I reference CytomX. CytomX technology may afford us an advantage of using that as a means of avoiding the potential risk of healthy tissue expression, and utilizing our IGN technology, but otherwise without that type of a mechanism you're going to have to be very cautious about what the target selection is, and what the level of expression would be relative to healthy tissue expression.

The other differentiator is that our IGN family of which DGN42 is the lead indication is significantly more potent than our tubulin inhibitors. As we think about dosing, we think about a number of dimensions there, we would expect to be dosing patients at a much lower level with an IGN oriented compound, than we do with a tubulin based compound. We wouldn't need the level of material or the level of dosing to be able to get to a reasonable level of activity.

Okay. Thanks. That's very helpful.

Yes.

Our next question will come from Shaunak Deepak from Jefferies.

Good morning guys I just have a few questions on Kadcyla. I was interested in your thoughts on the recent [Nice] decision on Kadcyla, and I was curious if the 20% discount on Japanese pricing was in line with your expectations? Finally, I was hoping you would be able to provide a little bit of an update on reimbursement in pricing in Europe beyond UK and Germany?

Sure, it's Dan. These are elements controlled by Roche. I'll give some thoughts on it, but the interactions between regulators and reimbursement authorities, are all taking place with Roche. As it relates to the decision with Nice in the UK, and I'm sure everyone saw it, but Nice came out with a negative recommendation based on quality added life years for Kadcyla. And I don't know whether you say that was expected or unexpected. Given where Nice has come out on a number of therapies it becomes part of a process. I know that Roche in a quote that I saw expressed significant disappointment. But it is probably just early in the process of moving things through. And I expect there will be active discussions about how they can come up with an approach that will receive the endorsement from Nice.

There is an alternative mechanism to reimburse patients in the UK today. And so I don't know that it shuts off the avenue of access to Kadcyla for UK patients. The discount in Japan, the 20% discount, again I don't know that that's necessarily unexpected. Japan does look at the US as reference pricing, and so I think that the 20% discount doesn't strike me as being severe, but that obviously came out of discussions between Roche and the authorities in terms of what was going to be acceptable in Japan. So there's going to be more coming out around reimbursement as they move through other jurisdictions, and we'll just have to learn over the course of the coming quarters how that takes place, and will impact introduction of Kadcyla in a lot of other jurisdictions.

Okay. Thank you.

And we'll now move on to Matthew Roden from UBS.

Hey guys it's actually Andrew Peters in for Matt this morning. A couple of quick questions. The first one just on expectations for data later this year for 289, as most should we expect more of a safety update, or are we going to see some kind of early data? I know that it's unlikely that most patients will have seen MTD yet? And secondly, just a question on AMG595, if you could just kind of compare that, or how is it different than AVT414? I know there they have a bit of a broader development plan looking at solid tumors, as well as glioblastoma. Is there a rationale for doing that? Or have you spoken with Amgen on why they're limiting it to glioblastomas so far?

289, as I said we announced that we had opened the Phase 1 in November, we're going through dose escalation. I think we certainly will be planning to provide an update on safety, and obviously if we are seeing efficacy I think you can fully expect us to want to share it. But really it would be a safety update. Hopefully able to demonstrate that we have been able to escalate through without significant skin issues, which I think has been one of the primary focuses. I think we may be a temp TD by that time, so there may be some additional information, but it's a dynamic situation, and obviously there's more than one meeting in the second half of the year. The AORTC triple meeting, so exactly which meeting we still just need to exactly work out, but we plan to, we fully expect to submit something, so that we at least will be able to have some data during this fiscal year, this calendar year, sorry.

Andrew, on your question around 595, now we're getting into partner strategy and the like, and that's just something that we can't go beyond what Amgen has said, and frankly all that we've have heard from Amgen publicly is what was disclosed at AACR, so that's not a path that would be appropriate for us to go down.

Okay. Any differences between 414 and 595, is that also in that same bucket?

Yes, when you get into comparing now it even goes beyond comparing a non-partner's compound to a partner's compound, I think you would want, I would want them to be responding to that as opposed to me saying anything about it.

Okay. Fair enough. Thank you.

Our final question will come from Jason Kantor from Credit Suisse.

Hey, great, thanks for taking my question. On 853 just a couple of things, one I guess how confident are you that the new dosing strategy will get a, do away with the ocular toxicity, it seemed like from the AACR poster, you're still kind of very close to the exposure where you're seeing ocular toxicity?And then also I think in the past you've spoken of possible accelerated approval paths in ovarian cancer. Do you think that actually exists in the US just given other company's experiences with the FDA, and the need to come up with some survival data for approval?

This is Charlie. Well, yes, in terms of the ocular toxicity I think the key thing here is, I don't think it's going to go away, okay? I think we would be willing in the face of a highly active drug we would be willing to accept a certain degree of symptoms. But I think the problem with the previous dosing strategy was that it was becoming unpredictable who was going to get it. Here I think what we're doing we were able to exercise control, and I think you saw the data that says you were absolutely right, that we were pushing up against the line, having not seen any ocular toxicity we have actually said well, we should go one dose higher, see if we can get more in, and see what happens there as well. And then decide which is the right level and which should go forward. I think the key message I think from there is that if you use total body weight, we saw 4 grams of 10 and 5 milligrams per kilogram dose, because some patients were overdosed, frankly. And if you use just ideal body weight you get better predictability and therefore less ocular tox. Now it's a matter of just finding that balance between the two, and that's what we continue to strive to do. So I think it has to date achieved what it was set out to achieve.

I think one thing to, the note of caution on the data with AACR is the range of body weight of patients in that cohort was actually relatively narrow compared to what we've seen previously. So obviously we were highly encouraged by what we've seen, but perhaps haven't tested it at a particularly heavy overweight patient, or particularly underweight patients during that particular cohort, so there's more to learn but we're very pleased with where we are so far.

In terms of ovarian cancer, it's always data dependent. I think yes if we saw some kind of dramatic response I think there's always a question, a conversation to be had with the Agency, but we're also well aware that that's been tough without overall survival, let alone progression free survival in the past. I think we should also bear in mind that we have begun to see activity in endometrial carcinoma, which I think is clearly a highly unmet need with, if any product is indicated for it must be a very old product. So that's something where I think we really could have an accelerated path. So we don't close anything off for the moment. I think the main message from us is that we will always be on the lookout for that opportunity, clearly we would like to transition to being a Company in a pivotal development in the commercial company at the earliest opportunity, and we will continually be on the lookout for the opportunity to do that.

And Jason, just if I could add to Charlie's comment, as I think about what's been done in developing this dosing strategy, I applaud Charlie and his team, I think it's given us an important new tool that we're applying, looking at 853 and the dosing strategy, recall also we're looking at weekly dosing which has come out of some of the same work, which may afford us both the opportunity to calibrate dosing around mitox, and what is acceptable while having higher exposure to patients that may have a significant impact on efficacy. I think it's going to be very important there.

It also I think is provided us an insight that may have application as we think across the entire portfolio. It's not clear we're not ready to make a statement that this is the dosing methodology that we'll use uniformly, but for compounds that have a certain PK profile, I think it would be appropriate, and it gives us something, a sharper instrument as opposed to a very blunt instrument that in hindsight was what we were using with the dosing schedule given the parameters of 853. I think we have to learn more, but it has the potential quite frankly to be a very important insight, that will significantly enhance our ability to succeed with some of these compounds.

And with that I would like to turn the call back over to Carol Hausner for closing remarks.

Great. I would like to thank everyone for your interest in ImmunoGen, and if you have any subsequent questions, please don't hesitate to call. Have a very great day.

And that concludes today's conference. Thank you for your participation.