Madrigal Pharmaceuticals Inc (MDGL) 2013 Q3 法說會逐字稿

Good day, and welcome to the Synta Pharmaceuticals third-quarter 2013 earnings conference call. Today's conference call is being recorded and webcast. At this time for opening remarks, I will turn the call over to George Farmer, Vice President of Corporate Development at Synta. Please go ahead, sir.

Hello, and thank you all for taking the time to join us today. With me are Safi Bahcall, our Chief Executive Officer; Vojo Vukovic, our Chief Medical Officer; Keith Ehrlich, our Chief Financial Officer, and Ilker Yalcin, our Vice President of Biostatistics.

This morning we issued a press release the reported results for the third quarter 2013. This release can be found on our website SyntaPharma.com. Before we begin, I would like to point out that we will be making forward-looking statements based on our current intent, beliefs, and expectations which are subject to certain risks and uncertainties. Additional detail can be found in related SEC filings also available through our website. I will now turn the call over to Dr. Bahcall. Safi?

Thanks, George, and thank you all for joining us this morning.

Vojo just returned from the World Conference on Lung Cancer in Sydney, where one year follow-up results from the GALAXY-1 lung cancer trial were presented. Let me start by turning the call over to Vojo for a clinical update.

Thank you, Safi.

We were very pleased by the reception in Sydney with the results presented by Dr. Ramalingam, the principal investigator of the study. The results are the most mature we have today from GALAXY-1, with one year follow up since enrollment completion and just over 65% of survival events. They show very clear survival separation in the pre-specified population selected for Phase 3 for well over one year follow-up. The results are very motivating to those of us who have worked with lung cancer for many years, because patients in the [south] setting, meaning after first-line treatments, generally have very limited options and live, typically, 7 to 8 months. For a drug to increase survival by 30% to 40% is very exciting.

The strong survival signals I'm going to test with this level of mature data is the key take away for all of us and our investigators. We reviewed some of the findings in more detail in a call from Sydney last week, and since we haven't had the opportunity to address additional details in the Q&A session late in this call, I would like to make just two additional remarks and then turn it back to Safi. First, the goal of this very large Phase 3 trial, with close to the 400 patients enrolled in total, was to de-risk Phase 3 by identifying the best patient operation and ways to optimize the operational plan. We are not aware of any oncological program that had this much data in the same setting, with the same trial design, and the same countries, heading into clinical trial. We feel very fortunate to have such a wealth of data and to be able to use this to optimize Phase 3. I am referring both to the selectional Phase 3 population, which as very nicely confirmed again at Sydney, and to the identification of our patient profiles from certain countries which led to our decision to no longer enroll patients from Russia and Ukraine. We are aware of many trials that have failed the transition from Phase 2 to Phase 3 because of the operation risks expanding to many untested sites and countries.

We've already begun to add new centers in North America and Western Europe to GALAXY-2 and are pleased with the very high level of interest we have seen. Survival rate has the goal stated in oncology and positive survivor data from a large randomized mod center trial with the most positive modulator we know. With our new operational plan, we expect over 75% of sites in GALAXY-2 to be from Western countries.

Secondly, the results from GALAXY-1, as well as other trials, have clearly established that ganetespib is clinically active, and its safety profile is very encouraging. Our increased confidence in the activity of ganetespib supports additional investments to mitigate Phase 3 execution risk, including adding patients to decrease risk from imbalances or statistical fluctuations. Our internal projections estimate the hazard ratio for the GALAXY-2 population to be the 0.6 to 0.7 range. With 700 patients, GALAXY-2 has over 85% power to detect any hazard ratio below .75. We should believe that's a comfortable margin to show meaningful survival advantage. We intend to review an updated statistical plan for a 700-patient trial size with our lead investigators and data monitoring committee and hope to finalize this change very soon. This change, it will be expected interim analyses from GALAXY-2 in the second half of next year and final beta in the first half of 2015. Safi?

Thanks, Vojo.

With regards to the ganetespib program, as Vojo mentioned, the trial to date have shown the ganetespib is clinically active with encouraging safety. Our number one goal is to honor our commitment to patients by bringing this drug to successful clinical outcome and regulatory approval. The objective of GALAXY-1 was to optimize GALAXY-2 in order to deliver on that goal. The three points Vojo mentioned are excellent examples of that strategy -- patient selection, operation, and trial size. I would like to add that the extra insurance from increased trial size and from refocusing enrollment all Western countries is important to us both as people committed to delivering this drug to patients and as into shareholders. We all recognize the potential value at stake in opening the envelope on a positive Phase 3 trial in lung cancer, and we recognize the benefits of extra steps to ensure this positive outcome.

Before turning it over to Keith for financials and then to questions, I would like to briefly mention our HDC program. As a reminder, HDCs are conjugate molecules that link ganetespib or other Hsp90-inhibitors, with small molecule cancer drugs using the tumor accumulation properties of Hsp90-inhibitors to improve the delivery of those payload drugs to tumors. We've been receiving a lot of inquiries about this program from both the pharmaceutical community and the investment community due in part to the high interest in ADCs, antibody drug conjugates. One of the most common questions has to do with whether we intend to develop HDCs with older established drugs such taxanes or platinums or with newer and investigational agents. We've generated HDCs for both and we can say that feedback from our partnering discussions has been a high interest in taking their newer or investigational agents, for example, MEK inhibitors, proteasome inhibitors, PI3K inhibitors, or CDK cell cycle inhibitors, of which there are many in the industry, and using the HDC technology to generate a competitive advantage, and in the case of certain pharma companies, to use HDC as a next-generation lifecycle management opportunity. We are all, at center, very excited about the broad potential and broad partnering opportunity of this program. And now to Keith.

Thank you, Safi, and good morning everyone.

They were no revenues recognized in the third quarters in 2013 or 2012. In the third quarter of 2013, our research and development expenses were $17.6 million as compared to $11.7 million for the same period of 2012. Our third-quarter general and administrative expenses were $4.2 million as compared to $2.8 million for the comparable period of 2012. Our net loss in the third quarter of 2013 was $22.5 million, or $0.33 per basic and diluted share, as compared to a net loss of $15 million, or $0.25 per basic and diluted share, in the same period of 2012. As of September 30, 2013, we had approximately $53.4 million of cash resources on hand. Based on our current operating levels, we expect our cash resources will be sufficient to fund operations into the second quarter of 2014. Certain new activities contemplated for 2013 and 2014 will be conducted subject to the availability of sufficient financial resources. I will now turn the call back over to Safi for concluding remarks. Safi?

Thanks, Keith. This concludes our prepared remarks. Operator, we will now open the call to questions.

Thank you, Dr. Bahcall. We will now be conducting our question-and-answer session. (Operator Instructions) Thomas Wei, Jefferies.

Just a clarification on how many centers were involved previously in GALAXY-2? And what was the geographic mix and how many new centers are being opened with the potential expansion of enrollment?

Let me turn that over to Vojo.

Sure. In GALAXY-1, we had approximately 60 active centers. And in GALAXY-2, we're obviously planning a very significant expansion. As we mentioned in the quarterly reports and the press release, there will be some operational adjustments going on which will result in a new number of centers, but what we're really targeting is having a very similar geographical distribution in GALAXY-1 and GALAXY-2.

Was the original number centers in GALAXY-2? Was it 60 as well?

No, in GALAXY-2 we had, obviously, much larger number of centers because the study was much larger compared to GALAXY-1, and the original number centers was 114.

Okay. And you know how many you'll be taking that to?

These plans are still under discussion and we're still working on that, but we should have a similar comparable number of centers after we complete the operational adjustments.

In these new centers that you're bringing on, can you say -- are these just different academic centers that you did not use previously? Or are you now having to -- have you gone through most of the usual players on the academic center side and you're going to be reaching into the community center level to get these new sites?

So we are expanding in the West, as we have said, and there we expect to have a good mix of both established credible academic centers and very productive, experienced community centers.

And then just lastly, from a funding standpoint, any update on how things are progressing on a partnership for the ganetespib now that World Lung is over and you've had a chance to get some feedback from perspective parties there. And what other options do you have besides equity financing to help fund the GALAXY-2 trial? Thanks.

The discussions are -- we are in advanced discussions. They are going well. We were certainly -- a lot of us were waiting for the World Lung data. I think, as Vojo mentioned, we are very encouraged by the positive reception in Sydney to the data in the lung community. And I think that will have a positive impact, although I think most expectations on the data are consistent before and after that meeting. Beyond that, I couldn't really give any guidance on timing. We've never really guided on timing discussions. In addition, we continue to look at all -- we always look opportunistically at all options.

Thanks.

Mike King, JMP Securities.

A couple of quick ones if you don't mind. Vojo, thanks for the updated power information on the upsized trial. I appreciate that. I just wanted to ask if you have formally submitted a protocol amendment for the patient numbers and whether the pivot to the Western centers is going to also require its own change in protocol or is that a routine change.

As we mentioned, we are discussing and planning to discuss the plans about those operational adjustments, including potential increase in trial size with our data monitoring committee, with our steering committee, and we expect to finalize these changes very soon. In terms of will that require a formal amendment of the protocol, we don't think that it will require a formal amendment of the protocol because the protocol allows for a dynamic change in trial size, so those changes can be implemented, for example, in the statistical plan.

Just wondering, I was also going back and reviewing the curves -- the GALAXY-1 curves from World Lung and just noticed there is a substantial number of censored events in both arms but a lot more in the ganetespib and docetaxel. Can you give us exact numbers that were still censored at World Lung and I'm just wondering also, have you projected out when you hit -- if the progression events happened at the same rate, what the final hazard ratio will be when you cross that 70% threshold?

Unfortunately, I don't know the exact numbers in terms of how many patients I exactly censored, and that's obviously a [Danik] variable. But the percentages that you're looking for are in the presentation in slide deck. And so, we expect for these additional events to occur over the next couple of months and, as we said, we expect to conduct the final analysis in the very early part of 2014.

I'll get back in queue, thanks.

Gene Mack, Brean Capital.

Vojo, I wondered if -- or maybe Safi, you could just give us a sense of how the -- I'm assuming that the GALAXY-2 trials got some built-in interim looks either for futility or whatever. And I am wondering if you could just give us an idea, from the outset, when those looks or at what point, in terms of event those looks might occur and if there's statistical penalties for any sort of material things that we might want to just consider once those events occur and what may be stopping criteria, if any, there are for each of those looks.

We certainly have defined very specifically at what point the interim looks will occur. That was discussed and agreed upon with the regulatory agencies. We don't disclose those exact points. In terms of the procedures regarding office spend and fiscal penalties, that's all discussed and defined and agreed upon with regulatory authorities. We are following standard procedures and standard methodology.

I'll turn it over to our head of biostats. Ilker, you have some comments?

Fiscal penalties have been incorporated in the sample size calculation, so yes, we will have the penalties but it is accounted for in the sample side.

One thing, Gene, that we have guided in the past is that when companies elect to do -- this is run, the GALAXY-2 is a typical, pretty plain, vanilla Phase 3 trial with interim looks and a final alpha in the original design of .042. So what you can gather from that and what we've commented in the past, companies could put interim looks at more aggressive time points. For example, very early in the study like 30% to 40% would be pretty aggressive and pretty early, or companies can put the more conservatively. And what we've got in the past is that our interim analysis are in a more conservative time point, and next reflected in the final alpha being pretty close to .05.

Great, that's certainly helpful. And then, I wonder -- I don't know if you reviewed this in the beginning. I got on late. But can you give us an idea of how much ENCHANT data might be at the San Antonio breast cancer conference? Have you guys just got an idea of how much would actually wind up being presented? How many patients?

So I'll take the first crack at that. We will be presenting initial data from ENCHANT-1 at San Antonio, and the exact scope and content will be discussed over the next couple of weeks with the investigators when we finalize the presentation.

Okay, great. Thanks.

Joe Pantginis, ROTH Capital Partners.

Thanks for taking the question. Quick question is first -- is the 700 number for patient enrollment for GALAXY-2, is that a firm number at this point?

So to 700 number is part of our plan, as we mentioned before, that we're discussing over the next month or two with our investigators, and that's pretty much it. That's our thinking.

And the more important question is if you look at -- we're talking about the fundamentals here so I don't want to invoke the stock price, so obviously there's been some volatility in the understanding of the data or the views on subpopulations, however you want to describe it. So I guess I would ask -- but you've obviously gone through a very methodical approach to get where you are today in answering the questions of the right population, so I guess while this volatility remains maybe on the Street, Vojo, can you comment on your initial thoughts that you shared with us at the World Lung conference when you were discussing the data with the physician community with regard to what the physician community's views are regarding the approaches that you took in the subpopulations that you looked at?

That's a really great question. We've certainly, throughout the program, discussed quite intensively with the investigators around the world the approach. Just as a reminder, we set out from the beginning to identify the right patient population with a drug-like ganetespib which is in Hsp90-inhibitor with a broad spectrum of activity. Our task of finding the right patient population was a little bit more difficult compared to very narrowly targeted drugs such as PKIs for example. And so we've consulted, quite frequently with investigators, and we've taken this trial through a broad methodical approach to identifying the right patient population. The concept of our drugs showing the best activity in chemosensitive patients is a very logical and, I think, very strongly supported point. I think once we explained the approach, the investigators are on board with that. And some recent scientific information and I think some preliminary part of that has been presented at the World Lung conference has provided, also, some of the aspects of the underlying biology about why, for example, ganetespib works better in the chemosensitive patient population. So, I think it's really coming together and the approach that we are taking, I think, has really gathered a lot of support in the lung cancer oncology community.

Thanks a lot. That's very helpful.

George Zavoico, MLV & Co.

Thanks for taking the questions. I have a couple of quick ones I think. First of all, with regard to increasing the number of patients and to the -- your costs and your burn, you're saying you have funds into the second quarter of 2014. Does that include the extra component of patients?

Will as we disclosed -- what we said was that based on our current operating burn, that cash would last into the second quarter. Obviously, as Vojo said, we are continuing to work on figuring out exactly the sizing and that, so it would be difficult to forecast beyond that.

Okay. And can you provide an updated at all, right now, as to how many patients are already enrolled in GALAXY-2?

We haven't provided enrollment guidance in the past, and we're not providing it now. Just to clarify, because I don't (inaudible), we intend to go forward with the 700 patient trial. Cost of patients and the incremental cost of a new patient added in cancer trials is typically in the $30,000 to $40,000 in external patient costs.

Per patient, yes, okay. With regard to the HDC program, can you provide any guidance whatsoever with regard to when you might be able to get into pre-90 and identify a lead compound? Or it still too early to tell?

We are in preclinical development now. We hope to have an IND filed within the next -- could be 12 to 14 months.

Okay. Good luck for that. I hope -- I'm waiting to see which one you pick and for what indication. It will be really interesting. And then, final question with regards to the GALAXY-1 data that you presented at the World Lung Congress, in the median overall survival, from your numbers on May 2013, you had ganetespib benefit of 4.3 months. It was reduced to 3.3 months just now at the World Congress mainly due to the docetaxel alone arm -- the median overall survival apparently going from 6.4 to 7.4 months. The GNB -- the G plus D arm stayed about the same at about 10.7 months. Is that largely due to the Russian and Ukrainian patients or do you have any other explanation for that increase in the overall survival?

It is normal in large clinical trials to see some fluctuations as it gets along. Our confidence in the numbers that we currently have is quite substantial. We have a very robust data-tested 5% of survival events. We should also remind ourselves that point estimates are not as precise as overall hazard ratio estimates. But then also, specifically referring to your point about Russia and Ukraine, it is quite feasible that patients from the two countries have contributed to this effect because those patients came on late in the trial, and therefore, the results from these patients are seen later in the trial.

And this explains, obviously, your decision to increase the number of patients to account for these sorts of unpredictable events that might occur in GALAXY-2. So I appreciate that and the added insurance that you're adding to the outcome for GALAXY-2. Okay, thanks for taking my questions.

Brian Klein, Stifel.

So first, we saw in GALAXY-1 that there was a large regional difference in terms of the survival between patients treated with docetaxel alone versus docetaxel plus ganetespib. Can you provide us with your assumptions now for the phase 3 program in terms of what you expect median survival would be for the docetaxel alone arm? And what you anticipate the combination will demonstrate?

Sure, I'll turn it over to Ilker. I do want to address both points. The regional differences as seen in our trials, also seen in quite a few other trials, is really behind those strategy that was developed a couple of years ago in the GALAXY-1, GALAXY-2 program to be able to tease this out and get exactly this kind of information. There are many examples, unfortunately, of trials that see these kinds of regional differences at the end of Phase 3 and we wanted to go -- part of GALAXY-1 and part of the point exactly was to go into all of those countries early, identify that early, and we did start to see some of those patterns as much as six or nine months ago and have been following it closely, which led to the decision recently to terminate enrollment in those two countries. And you're right, one of the features is that in those countries, we did see a lot of enthusiasm for the drug, especially towards the end of the study, and we dig a lot of long-lived earlier-stage disease patients and that did lead to patients who had much longer median survivals being enrolled from those countries, which means that you need a lot more follow-up to see a difference. If a patient is going to live a year and a half or two years, you need at least 1.5 or 2 years to begin to see a difference between two treatment arms. I think the powering -- Ilker can talk to. The powering, as you know, doesn't depend on the median, but in terms of the timeline, anywhere from 7 to 9 months median in the control arm would be consistent with the timelines that we provided. Ilker, do you want to add some points?

No, I agree that statistical powering is primarily based on the hazard ratio of 0.75 regardless of the median. So, medians could be 7.5 to 10 or 8.5 to 11.3 -- should provide the same power.

I guess the question really involves now that you are pivoting towards more Western enrollment, when we look at the GALAXY-1 data, you showed Western Europe and the US survivals of 3.5 to 5 months for the control arm and then about of 2- to 3-month benefit. And when you looked at Eastern Europe, you showed seven months median versus 10 months with a combination, so I am just trying to understand, now that you are pivoting which was the Western world, you expecting that the medians will be closer to the 3.5 to 5 months or will they still be at around 7?

I think those numbers are off. Are you thinking of progression-free survival? Because there is no survival time that's around three months or four months. Progression-free survival is around three months. We've been talking about overall survival.

Okay, all right. Just wanted to double check on your expectations for --

What we're talking about overall survival, so you're right, patients in the West were more standard advanced disease patients. The median survival in the West was in the neighborhood of six months in the hazard ratio that we've seen in the last was about 0.5.

Thank you.

Median survival in the East was closer to nine months or 10 months, and the hazard ratio is much less mature there. Because they live so long, there is relatively less follow-up. So a median survival of six months and a hazard ratio of about 0.5 in the West and then with a median survival of about 9 to 10 months and a hazard ratio in the 0.8 or 0.9 range in East. Does that answer your question, Brian?

Yes, that does answer it, thank you. My second question is now that you've upsized the trial, can you give us a sense of a number of events that need to occur before you unblind?

Yes. Our head of corporate communication is saying no, no, no. We've consistently had a policy where we don't provide the exact number of events. I think what we have said in the past on this question is they are pretty typical plain-vanilla assumptions for powering Phase 3 studies, and we're well in line with what you see in a typical Phase 3 trial in terms of follow-up and number of events. Is that all right?

Thank you for taking my questions.

Robin Davison, Edison Investment Research.

Just a quick one, please. It just occurred to me that giving you are undertaking this protocol amendment and expanding the study, have you given consideration to excluding any basically mutation types? For example, KRAS, where certainly the data wasn't good there for OS in the last data set. Is it possible or have you considered it?

Hi Robin, I'll turn it over to Vojo and Ilker if they have any -- I think, while one should be careful about the KRAS -- what you're seeing in KRAS is there isn't an enormous effect but it's underpowered to see if you have the same effect. It's only 63 patients, so with 63 patients your power to detect a very big effect which is closer to what you're seeing with a high LDH group. But you are not powered, so there's no evidence one way or the other that it would do better or worse than the overall. Ilker, did you want to add anything?

No.

Vojo?

No.

Other than that -- the GALAXY-1 program, the strategy that was developed a couple years ago was to gather information on patient population operational plan using GALAXY-1. And I think it's fair to say that we've got all the information that we were looking for and that we need to optimize the Phase 3 trial, and that's what we've done.

Right, so there would be no other changes to the size of patients that are recruited apart from those we've discussed right? The evidence of the increase in size and the change in the center profile if you like.

That's right. The GALAXY-1 data -- that's correct, Robin. The GALAXY-1 data are near final and they've given us the information we were looking for to optimize the Phase 3.

And just one other one quickly, actually. Did you collect PD-L1 biomarker data or can you? Could you even? To see whether that has any effect?

No, we're not collecting that information. Obviously, PD-L became interesting for the lung cancer community way after we finished the design of the original GALAXY-1 study, and also, technically it would be quite challenging to actually get this information.

Right. Okay. Will expect. Thanks very much.

Mike King, JMP Securities.

Thanks for taking the follow-up. A couple of questions. You guys had stated that you believe that ganetespib is clinically active, and I feel like there is a constituency of investors out there that don't buy into that. So maybe just give you an opportunity to kind of defend that statement with concrete examples.

I'll take a crack at that. I think we've shown consistently for the last couple of years CT scans in our presentations of tumors that shrink on ganetespib form of therapy and that these remain under control for a long period of time. There's a couple patients, I think, in our early trials who have been on treatment with the drug for more than three years now. So I think once you demonstrate that the drug has monotherapy activity -- and I think we provided that evidence in abundance -- I think that question about clinical activity, I think, is answered. What you also see in the GALAXY program is, in the randomized testing, is the increase in response rate, an increase in progression-free survival, and we see a very strong survival signal as well in the subset of patients that's defined prospectively. So, I think we have pretty substantial evidence about the drug's activity in the clinic.

Ilker, do you want to add any comments?

Well, working in large pharma over 15 years and Synta about now close to three years, in order to understand what is not clinically active, you have to see a lot of compounds. I put two drugs in the market, now it's over $6.5 billion, and I also killed so many drugs. You know when you look at the data, from the body of evidence, if something is clinically active or not active. I have no doubt that ganetespib is clinically active -- I mean, looking at the data from many different perspectives, we know (multiple speakers).

Actually we have Iman El-Hariry here who's working on our breast cancer program. Iman, you frequently get these comments from breast investigators. You have anything you want to add?

Thank you. Thanks Mike. I think -- I often look at my first 20 years or 15 years of industry experience and working in large pharma, seeing a thing in agent activity and is not an end of one or end of two that we have actually a very good number of seeing clinical activity in different tumors starts including lung cancer, breast, and others; it's been very encouraging. And it really is in line with the science that we started to become very conversant with on the HSP science and the division of client proteins. So when we took to our investigative [teams] there is huge interest in taking this drug in all subtypes for instances in breast cancer and given the level of evidence that and the level of activity that we are seeing. So this is really very encouraging and, like Ilker, I also worked on drugs that were approved and came to the market, and comparing the reason with my past experience, I think the level of activity I'm seeing far exceeds that what I have seen in my past experience in any of the drugs that I also have taken to the market.

Okay. Thanks for that. The other contention, and Safi, I think we've discussed this before in private meetings, but the other contention is that all of your activity has been in the ALK population. I wonder if you could speak to that in whatever level of detail that you are able.

I think it's true Hsp90 inhibition has very clearly been demonstrated. Now, several Hsp90 inhibitors in ALK-positive lung cancer, but we have CRs or near CRs in triple negative; we have pretty remarkable responses in mutant BRAF lung, pretty remarkable results in renal, and melanoma, and colorectal. So as Iman and Vojo said, it is a little bit difficult to grasp because people are very familiar now with ALK inhibitors for ALK-positive lung, or EGFR inhibitors for mutant-EGFR lung, or BRAF inhibitors for BRAF melanoma. So this is a compound that's kind of different because it does do that. It does see the kind of activity that you are seeing there with durable objective responses that last year or in some cases three years now. But it is broad. It's not just ALK lung. When you have a CR in triple negative breast or durable responses in gastric -- HER2-positive gastric, or HER2-positive -- it's more broad. So it's like a collection of those drugs acting together because you are targeting a network of proteins, not just one protein. (multiple speakers)

Sorry, finish your thought. Go ahead.

It just ties back to exactly what Vojo was saying a few minutes ago. It really underlies the strategy that we developed a couple of years ago. If you are developing an ALK inhibitor for ALK-positive lung, you've got to know your patient population. You don't need such a large exploratory trial to identify.

I guess it'd make my question clear. The responses -- the argument is that the responses in GALAXY-1 are largely in the ALK-positive patient population.

Oh I see, yeah. It's the other way around, actually. The patients without positive lung or mutant EGFR lung tend to live very long, median survival is -- can be a couple years. So it's very hard even with EGFR inhibitors to see a survival difference in EGFR lung. So when you have patients that live two years or longer, you won't see any separation very early on because they do well in both. So what a lot of people think is that we may eventually see a tail in the curve that favors the ganetespib -- that was mentioned at the last ASCO meeting. People expect to see a tail in the far -- when you wait two years or three years, you'll see a tale in the far right of the curve and that will be your 5% that is ALK-positive lung. But as Vojo mentioned, there's a very strong separation right away in the first two months, three months, four months, five months, and six months, and that can't be due to ALK lung. The ALK type of patients are all in the far right, the 5%, that live very, very long, and there is generally no separation in the early part. Does that answer your question, Mike?

Yeah, but were taking up a lot of time so we can continue the rest off line. Thank you.

Thank you. That does conclude the question-and-answer session. I will now turn the conference back over to Dr. Bahcall for concluding remarks.

Thank you all for your time and attention today. That ends our call.

Thank you, ladies and gentlemen. This does conclude today's call. You may now disconnect.