Clovis Oncology Inc (CLVS) 2013 Q3 法說會逐字稿

Good day, ladies and gentlemen and welcome to the third quarter 2013 Clovis Oncology Incorporated earnings conference call. My name is Dominique, and I will be your operator for today. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session.

(Operator Instructions)

I would now like to turn the conference over to Ms. Anna Sussman, Senior Director of Investor Relations. Please proceed.

Thank you, Dominique. Good afternoon and welcome to the Clovis Oncology third quarter 2013 conference call. You should have received the news release announcing our third quarter results. If not, it is available on the website at www.clovisoncology.com. As a reminder, this conference call is being recorded and webcast and may be accessed live on our website during the call and will be available in our archives for the next several weeks.

The agenda for today's call is as follows. Patrick Mahaffy, our President and CEO, will discuss the highlights of the third quarter and provide an update of our clinical develop and programs. Then Erle Mast, our Chief Financial Officer, will cover the financial results for the quarter in more detail and comment on the outlook for 2013. Patrick will make a few closing remarks and then we will open the call for Q&A. Andrew Allen, our Chief Medical Officer, joins us for the Q&A session.

Before we begin please note that during today's conference call we may make forward-looking statements within the meaning of the Federal Securities Laws, including statements concerning our financial outlook and expected business plans. All of these statements are subject to risks and uncertainties that could cause actual results to differ materially from those described in the forward-looking statements. Please refer to our recent filings with the SEC for a full review of the risks and uncertainties associated with our business. Forward-looking statements speak only as of the date on which they are made and Clovis undertakes no obligation to update or revise any forward-looking statements. Now I will turn the call over to Patrick Mahaffy.

All right, thanks, Anna. Welcome, everybody. Thank you for joining us on Halloween evening. I know many of you have children waiting to start trick or treating, or maybe you want to start trick or treating, so I will get started right away.

We made a lot of progress during the third quarter on clinical development programs, I'll start with 1686. 1686 is our targeted covalent inhibitor of the mutant forms of EGFR for the treatment of non small cell lung cancer. 1686 targets both the initial activating EGFR mutations, as well as the dominant resistance mutation, T790M, and as such has potential both in newly diagnosed patients as well as in those dosed patients who failed their first TKI. Because 1686 spares wild-type or normal EGFR, it has the potential to cause a much lower incidence of toxicities such as skin rash and diarrhea than normally associated with other EGFR inhibitors.

As you may be aware, earlier this week we presented promising clinical activity and safety data from the ongoing 1686 Phase I study at the 15th Annual World Conference on Lung Cancer Conference in Sydney. Based on these encouraging data we intend to accelerate the timing of the initial registration trial for 1686 in second line T790M+ non small cell lung cancer patients who most recently failed an EGFR directed therapy. That is now slated to begin in first half of 2014. This will be the first pivotal study of our broad and global clinical development plan. We intend to provide a full outline of this development plan early next year.

We used our time in Sydney to meet with our 1686 advisors and investigators who are very enthusiastic about this drug and about our clinical development program. That enthusiasm is driven by continuing encouraging results from the Phase 1 study. We presented data at World Lung from the 900-milligram BID cohort of the original free based formulation, as well as data from the first cohort treated on the new hydrobromide salt formulation. In the 900-milligram BID cohort six RECIST partial responses have been observed to date in nine evaluable T790M+ patients, 67% objective response rate. Eight of the nine evaluable patients, or 89%, experienced tumor shrinkage greater than 10%. There's been no evidence of systemic wild-type EGFR-driven toxicities to date.

These patients were heavily pre-treated. Eight of the nine patients have progressed on a prior TKI immediately prior to enrolling in the study and six of the nine patients received two or more lines of TKI therapy. As a reminder, these patients were dosed with the free base formulation which does not have the same attractive pharmaceutical properties as the new hydrobromide salt formulation. In August, we transitioned development to this new hydrobromide salt formulation and we presented preliminary PK and safety data from the first cohort at 500-milligrams BID. We were very pleased with the PK and safety results to date. This cohort demonstrated far greater exposures than expected, and with reduced variability and no rash or diarrhea. It is now cleared the 500-milligrams BID dose and the 750-milligrams BID dose cohort has commenced. Dose escalation is ongoing as the MTD has not yet been reached.

We expect to select the recommended Phase 2-3 dose by the end of the year and to then initiate the Phase 2 expansion cohorts to assess efficacy in second line T790M+ non small cell lung cancer patients as well as in the first line EGFR non small cell lung cancer patients in early 2014. We also intend to initiate our Japanese Phase I study in early 2014. Following a productive meeting with the PMVA, the Japanese FDA equivalent, we will be able to initiate our Phase 1 study at, at least 500-milligrams BID and therefore we do not anticipate that this study will last long. We will then be in a position to enroll Japanese patients in our global pivotal study. We will obviously include centers in other Asian countries as well.

In addition we recently announced an agreement with QIAGEN to develop a companion diagnostic test to identify the T790 and resistance mutation in patients with EGFR-driven non small cell lung cancer. As you may know, QIAGEN's test is already FDA approved to detect EGFR mutations, including T790M. This really complements our accelerated development plan for 1686 by potentially allowing for a supplemental PMA filing alongside the NDA.

Lastly, on 1686 I want you to understand that we understand the great desire for information regarding the progress of 1686. Obviously ASCO will be an important venue for this, but that doesn't occur for 7 more months. We view several events between now and then as opportunities to provide additional updates. These include the JPMorgan Healthcare Conference in January and either the IASLC Targeted Therapies Conference in mid February or the European Lung Cancer Conference in late March. In any event we will make sure that updated results from the ongoing study will be provided to you prior to ASCO.

Turning now to rucaparib. Rucaparib is our oral potent small molecule inhibitor of PARP1 and PARP2, which we are exploring as monotherapy in a maintenance setting for ovarian cancer. We had encouraging data presented over the past few months at the European Cancer Congress, the ESGO Conference, and the Triple meeting. We selected 600-milligrams BID as the recommended Phase 2-3 dose for rucaparib based on exposure, manageable toxicity and clinical activity.

From a safety standpoint rucaparib is well-tolerated at this dose, with most adverse events low-grade and manageable, important for a drug intended for use in a maintenance setting. No patients to date have discontinued rucaparib due to a treatment related adverse event. Rucaparib also demonstrates attractive PK properties, including low interpatient variability and predictable plasma drug concentration maintained over a 24 hour period after BID dosing. We've also seen encouraging activity, including eight RECIST responses in ovarian, breast, and pancreatic cancer patients in this Phase 1 study.

In germ-line BRCA ovarian cancer patients, we've seen one RECIST complete response, two RECIST partial responses, and two CA-125 responses. All of these patients are ongoing with maintained responses. These responses have been observed in both platinum-sensitive and platinum-resistant disease and overall 70% of ovarian cancer patients with germ-line BRCA mutations treated with rucaparib achieved disease control as defined by complete response, partial response or stable disease for greater than 24 weeks.

Yesterday, we announced the enrollment of the first patient in the global ARIEL2 study at a US site. ARIEL2 is a single arm open label study designed to identify tumor characteristics that predict sensitivity to rucaparib using DNA sequencing to evaluate each patient's tumor. We expect to initiate ARIEL3, the global Phase 3 registration maintenance study by the end of this year. ARIEL3 evaluates platinum-sensitive ovarian cancer patients with efficacy analyses prespecified in populations defined by deficiencies. First in BRCA, and then in other DNA repair genes.

Now, let me turn the call over to Erle to discuss third quarter 2013 financial results and guidance for the rest of the year.

Thanks, Pat. Afternoon everyone. Our full financial results are included in this afternoon's press release so I'll direct my comments just to highlights for the quarter. We reported a net loss of $20.3 million or $0.68 a share for the third quarter of 2013 and the net loss for the first 9 months of the year was $55.3 million or $2 per share. Research and development expenses totaled $16.1 million for the third quarter and $44 million for the first 9 months of 2013. R&D expenses increased slightly over the comparable periods from last year, primarily to expanded development activities for both rucaparib and for CO-1686, partially offset by the termination of the CO-101 program late last year.

General and administrative expenses totaled $4.3 million for the third quarter and $11 million for the first 9 months of 2013. These amounts represented increases over the comparable periods in 2012 primarily driven by higher stock compensation expense and some third-party professional fees. Finally, total operating expenses for the third quarter included $2.8 million of share-based compensation and $6.6 million for share-based compensation expense for the first 9 months of the year.

Our cash burn from operations totaled $16 million for the third quarter and just under $48 for the first 9 months of 2013 and we expect cash burn for the full year to be approximately $66 million. As of September 30, we had $356.6 million of cash and 30.2 million outstanding shares of our common stock. I'll turn the call back over to Pat for some closing comments and then we will open it up for Q&A.

Thanks, Erle. Let me talk about our anticipated milestone for the last quarter of 2013. For 1686 we plan the following, to select the recommended Phase 2-3 dose by the end of the year, to initiate the Phase 2 expansion cohorts to assess efficacy in second line T790M non small cell lung cancer patients in late 2013, and then first line EGFR non-small cell lung cancer patients in early 2014. And, most importantly, to accelerate the clinical development program in order to begin the initial registration study in the first half of 2014.

Turning now to rucaparib, before the end of the year, we intend to initiate the pivotal Phase 3 study, ARIEL3, in platinum-sensitive ovarian cancer patients. In summary, we are very proud of what we've achieved this quarter and look forward to continuing to advance the clinical development programs we have in our portfolio. With that, thanks for joining and we will now open the call to your questions.

(Operator Instructions)

Ravi Mehrotra of Credit Suisse.

Thank you for taking my question and congratulations on all the progress you've made this quarter. Two questions, if I may. I know the clinical profile of 1686 and AZD-9291 is just emerging, but perhaps you can just give us an apples to oranges comparison of the two.

And secondly, could you also give us more color on the null and valuable patients from the Phase 1 data you just presented? Thank you.

Yes to both of those, so first with regard to comparing in an apples to oranges way, or apples to pears for you, Ravi. AstraZeneca to us, several caveats here, right? First of all this is very early days for both programs, and so it is hard enough, and these aren't clearly in directly comparative studies. In fact, I would argue they are in slightly different patient populations as I'll describe here in a moment.

A couple things that I think are emerging that are demonstrating some of the characteristics of the compound as reported to date. It has a very long half-life, at least 50 hours, maybe longer. That has some benefits in terms of accumulation, and perhaps demonstrating efficacy at relatively low doses. It has some possible issues as well particularly in managing side effects when they emerge with a drug that is going to persist for as long as it does. Our half-life as you know is quite a bit shorter which I actually think in this case is an advantage.

We appear to see in the reports they made so far that they've got a pretty significant amount of variability in their exposures. That doesn't really surprise us, kinase inhibitors are notoriously poorly absorbed and as you know we had to fix that problem with 1686 and have very successfully with the hydrobromide salt formulation. They appear to be doing some work on a new formulation if clinicaltrials.gov is to be, is reporting it accurately. We'll see if they are able to overcome that variability as well as we did.

I think that the most evident difference is that even at these relatively low and initial doses at which they are setting the drug, they are seeing a greater incidence, in fact proof positive of EGFR-driven toxicities, particularly rash. How that will ultimately emerge for them at the dose they select I cannot say. But I think so far where we just don't see those types of side effects, particularly rash, I think that has the potential to be, if plays out over time, an advantage for 1686.

In terms of efficacy it is self evidently efficacious in both T790M+ patients and in patients are not T790M+. I think the reason for that activity in T790M negative patients relates to the patient characteristics in their trial. The majority of patients that have been reported to date -- and this will change over time as they move into Europe and into the United States. The majority of patients that they have treated thus far have been treated in Asia and primarily in Taiwan, where treatment practice is to immediately treat with a TKI a patient upon initial diagnosis with an EGFR driven tumor. But then to follow with chemotherapy, not to rechallenge with a TKI.

So since 40 plus of the initial patients that they reported on were treated at one center in Taiwan, I think that there's a possibility that what you are seeing with that drug, in addition to activity against T790M, but in some of these patients you could be seeing a retreatment effect. As you are all probably aware, it is quite common for patients who fail or another TKI go on chemo, then to be rechallenged and rechallenged successfully, a recent publication shows about a 35% response rate to these patients who failed chemo and the TKI and then are rechallenged with a TKI.

In our case, you know that of our responders only one followed chemotherapy, all of our patients by and large have been immediate TKI failures. So, it is the most rigorous test of a T790M inhibitor where you have to show real activity against T790M to have provided the benefit to the patient that we have provided.

More will emerge over time. We are in a race. They are a very able competitor with an active drug. I've been in races before, I'd just say I like our chances. Andrew, would you add or subtract, particularly if you need to subtract from any of that or anything else you would say?

The only addition would be on the issue of wild type inhibition, that the cell line that we have used and published publicly as the basis for assessing activity of our drug against wild type EGF receptors is the A431 cell line. And the reason we and many others have used that cell line is that it is a cell line that contains amplified wild type EGF receptor genes, and is driven, the proliferation of that cell line is driven by the wild type EGF receptor protein product. And therefore, it is very sensitive test of wild type EGFR inhibition.

That's why in our presentations we tend to show that even at the full efficacious dose of 1686 we have minimal inhibition in a xenograft of the A431 cell line. And we showed that in Sydney with around 30% inhibition of tumor growth in a xenograft system of the A431 cell line.

AZ have themselves published data with their molecule suggesting 78% inhibition at a 5 milligram per kilogram dose, which is not their optimally effective dose when they are looking for efficacy. And yet they showed fairly meaningful inhibition. So, non-clinically we have a suspicion that there is less of a therapeutic window for that molecule versus mutant forms of the EGF receptor than perhaps we have.

The LoVo cell line is perhaps not the most useful cell line to look at because it is a K-Ras mutant cell line, and as I think we all know, those lines are typically not driven by EGF receptor. So, showing low activity of a molecule against a LoVo cell line does not say a great deal about its EGFR wild type inhibitory activity.

We have done some work against LoVo in our own hands and our phosphate EGFR, or at least our GI 50 is around 2.3 micromolar. Afatinib is about 500 nanomolar and AZ's data suggests they are at 480 nanomolar with a phosphate EGFR IC50. So, this is technical, Ravi, and as you say, it's apples to pears, but in LoVo, I'm not sure we're learning a ton apart from to see that certainly our molecule has a very big therapeutic window, but I think the A431 cell line is more useful.

Okay. And then Ravi to your second question, I will just remind you and everybody that this is a Phase 1 study, so this is not a clean Phase 2 population where clearly absolute second line good performance data. These are patients who have been on five or six prior therapies, they are very ill in many cases and in every case I will point out very brave to participate in these types of studies.

The non-evaluable patients in our trial included one who refused consent prior to her first scan, I will remind you that was in the poster at ASCO. Two unfortunately died of their underlying disease before they could even get to their first scan. One we found out, actually a person who had -- was demonstrating a pretty good response, we found out was seeing two separate oncologists. One the investigator, one a private oncologist, and therefore was on additional anti-cancer medicine at the time she was on 1686. Quite frankly, as an aside, she is lucky she didn't suffer fairly serious consequences from that decision.

And finally, one of the patients enrolled without measurable disease, so 9 out of 14. That actually is a pretty good hit rate, and again I think it is very normal in a Phase 1 that this type of characteristic emerges. I will give an example, AstraZeneca reported on 89 patients who were enrolled in the trial, but reported only on 18 T790M+ patients, and results on a total of 35 of their 80 patients -- of their 89 patients.

So I'm not saying anything negative about AZ, I'm just saying in an apples pears analysis is not uncommon to see that you cannot really evaluate every patient enrolled in a Phase 1 study.

Nothing that scary there for us. Thanks for taking the questions.

Thanks, Ravi.

Charles Duncan of Piper Jaffray. Charles, please check your mute function.

Hi, I think Charles may have dropped off. This is Roy, I'll take a quick speculative question. Have you guys thought about doing any next generation sequencing in lieu of maybe the more standard type sequencing, try to discover some early T790M clones? Thanks.

Andrew?

Is your question about looking for T790M at baseline? You're referring to its sensitivity of next gen sequencing with ultra [deep] sequencing or--

Yes, right.

Yes so obviously, it is a very interesting question. We're thinking it is obviously in the front line context, is that your question?

Yes, I guess it would be mostly front line.

So in front line there is much interest as to whether patients can have T790M detectable before they have begun any form of therapy, the so-called de novo T790M+ patients. The reason that interesting is that there is a little bit of data that if patients do have detectable T790M at baseline they do poorly on standard front line TKI therapy. And there were data presented at ASCO this year suggesting that the median PFS in such patients was only around three months or so on erlotinib, that was data from Memorial Sloan Kettering.

Now, in using early techniques for sequencing, which are relatively insensitive, such as [Sanger] sequencing which has a 10% sensitivity threshold, you find very few patients have measurable T790M at baseline, meaning probably less than 1%. With more sensitive sequencing at Memorial and the US Lung Cancer Mutation Consortium have published on this, you see a figure of around 2% of patients with baseline T790M.

There are several groups around the world who published much higher figures using different technologies, and there is a very live debate in the committee as to whether some of these technologies have a problem with false positives or whether in fact they are reading correctly, they are just much more sensitive than sequencing approaches. We obviously are very interested in this point because if there is a sizable group of patients who are going to do very poorly on erlotinib because of T790M driven resistance, then clearly that's a group that could do very well on 1686.

So it matters to us a lot as to whether these patients really exist and really can be found. So, we are working in this area but at this point we haven't got any information that we can share.

Okay. Just to follow-up on that. The groups that I've seen publish the higher figures, it seems like all the studies were done in Asia, is that correct? And is that a possible factor, maybe of something specific to Asians?

Several of the groups have been in Asia but one well-known group is the Spanish group in Barcelona led by Rafael Rizal. They've also published a series with this quite high de novo T790M rate. So, I don't think that it is an ethnic phenomenon, at least at this point, it is too early to draw that conclusion.

Okay. Thanks and then one question on 1686. So you are at 750 mgs BID but still dose escalating, so and for that 750 mgs, BID is definitely not the maximum dose? And do you have any idea we might reach it?

We cannot say that. We are still in the DLT window timeframe for the first patient treated at 750 mg BID. The 500 mg BID dose was unbelievably clean. We've said that there were no rash or diarrhea or any GI side effects, there were really no side effects. This is a really well-tolerated drug. So there's nothing at 500 mgs that tells us, yes, we are close and that it is probably going to be 750mgs.

We are seeing really good exposures, and if we continue to see those really good exposures, which we anticipate, at 750 mgs or even at one higher dose, we have said consistently -- well, post ASCO that we need to be at 200 nanograms per mill or better for a 16 hour time period, or greater.

We obviously achieved that with the first patient treated at 500 mgs BID and by a wide margin, I don't have the PK data yet for 750 mgs. I would anticipate it would be better and at some point it may be that we just get to a dose because we've got great exposures and great activity and not great toxicity.

Okay. Thank you.

Cory Kasimov of JPMorgan.

Hey, good afternoon guys. Thanks for taking the questions. First one I have is on how we are going to know your drug works in patients with primary mutations? We've seen some data from AstraZeneca's drug there and it appears it is working based on the data. So, how do you know your drug works and when will you have data that you can show, you can demonstrate that? And I have some follow-ups.

Okay. I will give a quick answer and then Andrew as to timing and why we think it will and then Andrew, if you have anything to add. So, I'll remind you that we have exclusive data for both T790M+ tumor models as well as for activating mutation tumor models where we show we're as active as Tarceva or afatinib in a variety of comparative studies and in a variety of in vivo models.

I will just assert that since the T790M data translated absolutely 100% and beautifully from animals to humans, that will happen for patients with the activating mutations as well. This is a very active drug and its behavior has been well predicted in animal models.

As to timing where you will see these patients we will start treating patients with, treatment naive patients with the activating mutations in early 2014. I would imagine that we will not have -- I don't know, we may have, might have some data at ASCO, more likely data later in the year. At some point in 2014 we will be able to demonstrate initial results from an initial group of patients treated with activating mutations of EGFR.

Okay, great. And then with regard to your initial registration study, do you know yet how you're going to define what a second line patient is in terms of time since failure of their last therapy, the TKI or chemotherapy?

We do.

Okay.

I'm trying to figure out if I want to say anything about that right now and I think I do not.

Okay.

We've thought about this a lot. We have a very extensive clinical development planned but I think I will reveal the details of that full program early next year.

Okay. In that case I will ask another question.

I will tell you one thing, Cory, just to be really clear. At a minimum we are going to evaluate patients who have immediately failed a PKI consistent with the patient population that have done so well on the drug in our Phase 1 dose escalation study.

Okay. That's what I assumed. And then lastly, there's obviously a tremendous amount of buzz around the PD-1 class and a lot of data coming there now in lung cancer as well. And expectation is going to be used very broadly in that space so how do you see the role of 1686 evolving in this treatment paradigm?

What are the response rates we've seen so far in lung cancer for these immunomodulators?

Are you asking me?

I'm asking you to make the point.

They are going to a very broad patient population so you're -- it is in the 20s.

And we are in the 60s. And we don't have side effects. So what I think is going to happen is patients who have EGFR-driven tumors are going to be treated orally with a TKI because they can get great progression free survival with acceptable side effects. And in the case of 1686 I think we have a very good chance of being a very significant player in T790M+ patients who failed a PKI.

I think all of the animal data suggests that we have a very good chance of emerging as a very important first-line drug in patients with EGFR-driven disease. And given these response rates and the durability of these response rates that I have seen in the first line setting with TKIs, I just don't believe that the majority of patients are going to go away from them unless something miraculous happens in that response rate for the immunomodulators, but I will let Andrew add or subtract. Anything you'd add or take away from that, Andrew?

As you know, Cory, the higher response rates and the better efficacy outcomes with the immunomodulators appear to be in smokers and the story that's emerging is that perhaps that's because these patients have more mutations. So the oligomutated adenocarcinomas of the lung are perhaps not in the group of highest responders, obviously, particularly because these are non-smoking, often young people that get the mutant EGFR lung cancer. So, it is kind of the opposite corner of the matrix, if you like, from that corner where the immunomodulators are showing their best efficacy.

I would echo Pat's point that when we've spoken to physicians about their treatment choices, a well-tolerated effective first-line therapy that's targeted to the disease the patient has is obviously a very compelling treatment choice. And often the more aggressive in the therapy but carries a very different risk-benefit balance may be reserved for progression. Obviously the interesting question of combination arises and that obviously is something we are looking at in the nonclinical setting at this point.

All right, great, thanks a lot for taking the questions.

Thanks, Cory.

Brian Klein of Stifel.

Great. Thank you for taking my questions. Just have two. First I'd like to ask about the two patients who had responses at ASCO and then subsequently progressed.

I know it is a small end, but did you see a similar resistance pattern develop in both of those patients and do you believe that what you are seeing in terms of a resistance profile will translate over into a larger patient population? And then I have a follow-up question.

The answer is we don't know what caused them to progress. Remembering, these are six time, seven time in one case, patients. To really know we would have to mandate that these very advanced patients undergo a post failure biopsy and that's by and large impossible to mandate in a clinical study.

We may at some point get feedback from investigators, not necessarily for these two, but for others who then go on to subsequent therapies that are directed at a specific mutation. We don't have any information on those two. We did report some data based on in vitro analyses of what we think resistant mechanism could emerge. Andrew, do you want to talk about that?

Yes. In our online cancer discovery paper which is in public domain now as well as in some of the posters we have been presenting recently, we've described that in vitro in several different cell lines we see -- have seen epithelial mesenchymal transformation, EMT, as the basis for acquired resistance to CO-1686. It is a hypothesis that, that will be what we will see in patients. However, of course cell lines in patients often don't behave in the same way, and in particular, with cell lines, an individual cell line can be very hard wired to reproduce and predictably develop resistance always through the same pathway in a manner that's very different from a heterogeneous multi-clonal context that we see in patients.

So it is a hypothesis that we are going to be examining in patients when they progress if we are able to get biopsies, but as Pat said, getting those biopsies at progression can be very challenging in lung cancer patients. And in one of the patients who did lose progression, the progression was with a further CNS metastasis, which obviously is non-biopsiable, typically, and certainly in this case was not biopsiable. And that was a patient who, the patient we presented at ASCO who had an initial CNS response to drug and then developed a secondary progression within the CNS as well, so it just adds to the complexity of getting material from such patients.

I think one thing to note about one of those PRs, I think it was that patient, did have this formal sign of progression, but many other signs of great clinical benefit and actually remains on drug.

Great. Thank you. And then just a brief follow-up question on dosing and the new formulation. I know that you guys are employing PK parameters to determine whether you have sufficient drug exposure. Would you consider a three-time a day dosing regimen?

The evidence is we don't need to. We don't want to need to, to be clear. Because it is more of a burden for patients, two times a day is actually quite easy, especially because we don't require fasting, which can make it a little bit harder, but we don't require that for 1686. Three times a day gets to be a little bit more of a burden but there's no evidence from the PK we've seen to date that that's going to be required, so I think we'll stick to BID.

Great. Thanks a lot.

You bet.

Peter Lawson of Mizuho.

Patrick, just on rucaparib, when is data expected from ARIEL2 and what are the endpoints or top line data that we would see?

Andrew?

Yes, the data from ARIEL2 will start to become meaningful at the end of 2015. We may or may not be very public with those data because obviously what we're doing in that study is refining our genetic signature that we will be deploying prospectively in the ARIEL3 study. Which we will be running concurrently, although a little bit behind because it is a bigger trial. And obviously therefore it may not be something we wish to immediately announce publicly. So that's the timeframe and that's the way in which will be approaching and using those data. I'm not sure it will be visible to you in a real-time.

Got you. And then just considering the recent competitive data is there any change in how you are viewing the competitive advantages for rucaparib?

Andrew?

Yes. As we look across the spectrum of PARP inhibitors, some very interesting data have emerged recently that I think have changed the positioning of the different products. If you look simply at efficacy, and I'll exclude veliparib here where there is less data and I think everyone recognizes it is the least potent of all the drugs through whichever mechanism of action. But if you look at rucaparib, olaparib, and BioMarin's product, in terms of efficacy and responses in germline BRCA mutant patients, disease control rates, they all look quite similar.

Tolerability, one difference that's emerged is alopecia with the BioMarin product. And that may relate to emerging difference in the activity of that drug, which appears to be less through enzymatic inhibition of PARP and more through the so-called DNA trapping affect, which makes it behave a little more like a TOPO I inhibitor than it does a straight PARP inhibitor.

And that may be why we are seeing pretty aggressive myeloid suppression and alopecia with that product which isn't being seen with the other classes of PARP inhibitors, and obviously that may be important in women undergoing chemotherapy, for whom obviously that's a very important issue.

The PK data that have emerged and there's not lots of public data, but certainly we like the profile of rucaparib enormously in that when patients are dosed orally with 600 milligrams twice a day they have a very reproducible and tight plasma exposure, which is obviously very good from both managing toxicity, but also ensuring that you have efficacy in patients. So we like the PK profile and certainly it is amongst the best, if not the best of its competitors.

Then finally I think a clinical development strategy obviously that's really differentiating the different programs, but hopefully that gives you a window on how we are viewing the core attributes of the different drugs that are in the space now.

Great. Thank you and you may have mentioned this, I think you did. When do you expect to see Phase 2 data for 1686?

I want to make sure I heard that right. We do we expect to see Phase II data from 1686?

Yes.

Well, I would imagine that we will have data from the expansion cohorts. They are open label, so we can report certain things along way but you will certainly see data emerging from those expansion cohorts before the end of 2014.

Got you. Thank you so much.

(Operator Instructions)

Yaron Werber of Citi.

Yes, hi, thanks for taking my question and also congrats on the great data. I had two questions. One relating, I don't know if you can help us understand a little bit, how you're thinking about the hydrobromide salt dosing. It sounds like you went up 50% from 500 to 750 and you are still not seeing any DLT. So how are you thinking about the next dose and do you need to get there? And then I have a follow-up.

We wondered, not with any great amount of thought, we knew that if we had seen sort of meaningful toxicities emerge at 500 we would've had considered going to an interim dose, not going up the full 50%, we saw no toxicities. So I guess I'll make the same claim about 750. If 750 appears to be as clean in this group of patients as it was at 500 then I think we would probably go up in a 50% increment, that would be the most likely. We may just land on 1,000, because that's kind of an easy number for us given the strength that we're using right now to provide patients.

Okay, and thanks for that. How are you guys thinking about progression free survival, sort of durability of disease? It sounds like you are still not, you're over 181 days now, you still have two patients responding from the initial four at ASCO, but one of them is at the low dose, one of them is at the 900. How are you guys thinking about durability? Of response?

Two or three things I will say. One is the patient is at -- just to be clear he's not really on a low dose, he's at 600 mgs BID, so he dose escalated, so at some level I wouldn't call that a low dose, just for clarity. Two, we think it matters a lot.

We put out that PFS data because we got asked all the time and we are trying to be as transparent as we can with investors. But I will also point out four patients is hardly a data set. So, I think these expansion cohorts are going to give us a much clearer view of the true PFS.

We may see more emerge as we seen efficacy on the dosing cohorts of the hydrobromide salt that gives us a little more visibility because we will enroll more patients at each dose now than we did for the hydrobromide, except for the 900 mg BID, for the free base, except at 900 mg BID.

I've always said, we've always said that for a second line drug the target we would have would be minimum five months, more like six months PFS. I guess the good news is we have 181 days for the first four patients, but I don't know that that predicts anything. We know how important it is. Andrew, anything you would add?

I would just echo Pat as you say that in this patient group PFS is a difficult statistic to evaluate because you don't really know what control would look like with these multitude pre-treated patients. We do know that afatinib plus cetuximab has been a source of interest to physicians and there's a pivotal trial running of that combination right now as the best that we currently have that's in late stage for patients with acquired resistance to EGFR inhibitors.

And in the Phase 2 data presented last year the median PFS for the active combination at full dose was 4.7 months. So I think we can obviously see that five or six months with this agent would be a win for patients but we just don't have enough data to draw meaningful conclusions at this point, although certainly what we are seeing is encouraging.

Okay and Patrick, if you don't mind, it's Halloween, so I'm going to ask you this question and I'm hoping you're going to be willing to answer it. There's been a lot of discussion maybe on Bloomberg in two articles talking about M&A, that Clovis was up for sale. I think a week later there was an article that came out saying that the Company is not going to be acquired, there was no interest or, and I cannot remember exactly the words that were used. I know we are in a public forum but I don't know if you can comment on any of those -- any of that speculation? Thank you.

I guess you are going to go trick-or-treating this Halloween. I know the answer now.

There waiting for me outside.

Yes. Look, you know I cannot comment. And everyone knows I cannot comment, I think everyone probably knows I'm frustrated at the fact that I cannot comment. I will say that, that was a pretty bad week for us apparently, is one. And two, I think it is unlikely that any CEO in this industry would think he could charge a buyer $750 million per PR. Good luck with the kids.

So it sounds like the Company wasn't for sale, right? There was no process going on, I don't know if you can comment on that.

I've given you all I can give you, Yaron. I have a lawyer in the room who's looking daggers at me right now. You know we're all trained to say we cannot comment on that, it is just frustrating that we are trained to say we cannot comment on that. But I cannot comment on that.

Thank you.

Marko Kozul of Leerink Swann.

Good evening. Thanks for taking some of my questions and look forward to the maturing data sets. First one I have is on the regulatory front. We don't often see promising compounds for a single line indication.

I definitely wanted to ask you in terms of regulatory precedent or your understanding of regulatory requirements, how does this work if you file first for an accelerated approval? Does that preclude somebody else from coming in and filing at the same time or until you have an outcome to your application? I guess if you could give us some color on your understanding of that aspect, that would be my first question.

I will be really clear Marko, it, it does not preclude someone else. So, if we were to get an accelerated approval on day 5, whatever that is, or if we were to submit for an accelerated approval on day five, and if someone else -- say a company who started with an A submitted on day 15 or day 20, I think that they could get accelerated approval.

They would have to confirm, as we would, the data that was the basis of their accelerated approval. But we don't think there is -- we don't think there's that type of benefit to the first approved drug that it would preclude anybody else from getting approved without say a comparative study against us. Over time that would occur.

I will give a recent example. Both Tarceva and afatinib were approved in sort of a similar time frame in studies they ran against chemotherapy. The first one approved was Tarceva, it didn't force afatinib to go and run a head to head against chemo. That will be an analogous situation here. First matters, but it doesn't matter that much.

Sure, so if company C had an approval on T790 first, then company A would have to run plays against the compound from Company C?

I think that, for we and our known competitor now, I would imagine we will run similar development programs on a similar time frame and both, if the data are positive, be approved on the basis of those trials. If someone wanted to start a trial subsequent to those approvals against T790M+ patients they would have to choose we or that other Company as a comparator arm.

Okay, terrific. Second question, resensitization effect, or a retreatment effect, do we know what is more important here? Is it the TKI holiday length, or prior number of therapies or do you have any understanding of what is more important in the context here?

Andrew?

There is very little published on this topic, Marko. So there's no hard data to answer your question. The gestalt is that it's time off of TKI that matters, and that's where you are giving the original maximally fit clone which just carried the activating EGFR mutation time to be reassert itself and grow back to become the dominant clone. That's what then underpins the observed retreatment effect.

The published literature as we represented in Sydney suggests a response rate of around 30%. But in the conversations we've had with physicians most of them feel that number is higher than they see day to day, and more typically they see clinical benefit with stable disease, but not such a high response rate. But in terms of whether it is the number of prior therapies or the holiday, I think the feeling is it is holiday, but we don't have good data to make that a clear definitive statement.

I appreciate that. That was helpful. A question on safety. Two questions here actually. With emerging rate of [GI toxicity] of -- at least a case of diarrhea for one compound and nausea or decreased appetite for another. Do you have any feedback from physicians as far as which of those, if they had to split hairs they would prefer to see in the compound? And then the second question is does this provide any thoughts about combinability effects-- if that becomes part of that story down the line?

I'm going to answer this in a way that isn't exactly what you asked, but one of the things that we are going to be really interested in seeing is whether some of the GI tox that we did see, not really in a dose related way, for the free base, it doesn't show up for the hydrobromide salt.

As you know, for the first patients treated, and it was three, so this is a small end, we saw no GI toxicity. No diarrhea, no vomiting, no anything. And at some level it is an early possible confirmation of our thesis that the reason we saw some of that GI tox with the free base. As we discussed this earlier, the drug was not well absorbed and at higher doses in particular had a reasonable amount of drug that was in the vicinity of your belly.

So it is not entirely surprising we may have seen sort of a local GI effect from the free base. Not a systemic one, we didn't see that, but a local one. So it's one way of saying I'm not sure that the data we have for the free basis is going -- with the diarrhea is likely to be the same for the hydrobromide salt.

As to whether a patient would prefer to either not be very hungry or to have grade one vomiting episodes, or to have diarrhea, I'll just answer by saying they would prefer to have none of the above. I don't know that I could make a judgment about, particularly in the case of grade one versions of that, about which was worse. All are bad if they emerge as grade 3 toxicities, all of those, particularly the vomiting and diarrhea would be, they are hard. Andrew, do you have a view that one is better than the other?

No. Just personal judgment, but I agree with you that trying to avoid them all is the key.

All right, so that brings me to the last one which is, you had an interesting AACR abstract that I think got little attention where you showed that you essentially shut down the EGFR receptor, immediate progression path entirely. Can you talk about the impact and what that suggests about the drug and future development?

Andrew?

Sure. Well spotted. When we develop and try to engender resistance to 1686 in vitro, and I mentioned we've done this in several different cellular backgrounds now, we do find this EMT pathway, as a mentioned earlier. What we also find is that the EGF receptor pathway remains shut down. And therefore resistance is emerging through an alternative bypass set of pathways. And we're obviously characterizing those right now because that clearly might suggest a combination that could be useful in that setting or maybe even earlier in the treatment course.

It's obviously very encouraging to us to see this, because one of the questions when we started the work was will acquired resistance for CO-1686 be driven by some further change to the EGFR pathway itself, will there be amplification of the gene, or will there be a secondary mutation, tertiary mutation, in the EGF receptor gene that drives resistance to 1686.

We've obviously looked for that very carefully and have sequenced the gene in these resistant cell lines. And we're very happy that we don't see that, and indeed there does appear to be a need to bypass the EGF receptor pathway completely in order to develop acquired resistance to the drug.

So very encouraging in that it seems as though this pathway has now shut down in this cells, which augers very well for the utility of the drugs in the treatment of mutant EGFR lung cancer. But obviously, as ever, we will need to combine with something else down the road in order to really turn this disease into a chronic illness, which of course remains our goal.

Terrific. Thanks for taking the questions. Appreciate it.

Thanks, Marko.

This ends today's question-and-answer session. I would like to hand the call back over to Ms. Anna Sussman, Senior Director of Investor Relations, for closing remarks.

Thank you very much. We thank each of you for your interest in Clovis today. If you have any follow-up questions please call me at 303-625-5022. The call can be accessed via replay at clovisoncology.com in about an hour and will be available for 30 days. Thank you, and have a very happy Halloween.

Thank you for your participation in today's conference. This concludes the presentation. You may now disconnect and have a wonderful day.