"You called the Comet 1 Interim results dead on."
I did, but I wasn't smart enough to trade on it. Frankly I regret that decision. After looking at all the phase 2 data, my feeling is that Cabo treatment will add about two months to OS. The 10.8 month median survival for the expansion cohort solidified that expectation for me. The results will be out very soon and we will see where we go from there.
Looking at your valuation model, about a year ago I figured Cobi for $100m per year. That was assuming an even split in the market with the competing GSK drugs. I don't see MTC worth more than $40m per year. I think the other numbers are ballpark, but there are more VEGF TKI's in various pipelines. It also doesn't account for EXEL's management's proven ability to spend. Still, I agree with the thesis that it is very much under valued down here. I still have more confidence in Meteor and Celestial than in Comet 1. We'll see and I do have my fingers crossed.
'How are you handicapping it?'
65/35 it makes the primary endpoint.
"Do we all need to cross our fingers?"
Yes you do;
"I am assuming you wouldn't stay long if you didn't think Comet 1 would meet its endpoints."
That is a reasonable assumption.
"No One buying a company pays the short position. Those who "loaned" the shares for shorting, recall their shares."
Go back and reread the original thread. The concept forwarded was that a the acquirer of a company that also happens to have a large short position could somehow avoid covering that short if they eventually buy the company. I agree that a potential suitor is not responsible to cover the cost of the extra shareholder positions created by shorts. Those positions are paid off by the shorts covering when instructed to do so by their brokerages.
'"Let's say ABC has issued 1 million shares and XYZ has shorted 100K shares of ABC. There are now 1.1 million shares on the market held by "longs." Wrong! There are still only 1M shares. Its just at there 100K that have been borrowed. There is no "creation" of shares.'
Maybe the big numbers are hard to understand. ABC biotech opens its doors and has an IPO and issues 10 shares. Here are the lucky holders after day 1. Anderson holds 3 shares, Wilderguide 3 shares, Blackjack 3 shares, and Wildebeest 1 share. On day 2, Nomad borrows Wildebeest's share and sells it to Ernie. There are now long holders of 11 shares. On day 4, Nomad buys the whole darn company through a mandatory tender. He pays Anderson, Wilderguide, Blackjack and Wildebeest for their 10 shares and in addition he is forced to cover his short of one share so that Ernie can get paid for the one share he bought on the open market.
"Ernie, the version I'm reading is "NCCN guidelines 3.2014 NSCLC" which still lists cabozantinib for mutation-specific NSCLC treatment. Have I missed something? Where do you see this deletion?'
My mistake, thx for the correction. I looked a few months back and for some reason couldn't fine the listing. It is still the only drug recommended and it is a cat 2A listing.
NCCN is not going to contribute much to EXEL success. Cabo was listed for RET driven NSCLC in the 2013 version, only to be deleted in 2014. I thought one of the best NCCN opportunities for Cabo was in differentiated thyroid cancer, but this one didn't come through either. With Comet 1 reading out soon (in the next few weeks), I see very few of the prostate docs sticking their necks out until they see the OS result from that trial. We have a truly binary event coming and its either win or lose without any room in the middle. Making the primary endpoint will make any action by the NCCN irrelevant and if Comet 1 fails, there will not be many supporters left in the PC arena. I have my fingers crossed.
"erniewerner.......When you say...."Even if it were a material event (which is not generally considered to be the case).......are you saying that a buy-out offer......is not a material event ?"
I realize how contrary to logic this sounds. A formal buyout offer is material. Preliminary and hypothetical discussions are not. Think about it. In most cases, buyout offers are not publicized unless one party or the other decides to "leak" the information first. BMY's purchase of MEDX was finally agreed to and publicized only after an 8 month negotiation period. The very definition of material refers to the likelihood to influence stock price. Just knowing that firms are having exploratory discussions would definitely influence stock price, but many, perhaps most, of these discussions are not acknowledged.
"And the prohibiting of "selective release of insider information" is talking about selecting to only release information to any other individuals.............while not releasing that same information to the general public (stockholders) at the same exact time it is released to anybody."
Absolutely correct. A company cannot legally leak material information to any individual or entity. Material information, if released, must be done publicly. It is a common misconception that a company is obligated to promptly release material information, that is not the case. As I said, the main constraints on a company regarding treatment of material news are rules governing selective release and insider trading.
"I think that would be a material event and I'm going with yes. Just a guess though."
Even if it were a material event (which is not generally considered to be the case), company's are not obligated to promptly report all material events. Reg FD (fair disclosure) prohibits selective release of insider information and other regulations prohibit insider trading based on nondisclosed information, but companies can, and do, sit on significant news.
"What becomes of the short position if that short is also the aquirer of the shorted company?"
All long holders of a stock (including those whose holding resulted from the a short position) get paid off in the event of an acquisition. Let's say ABC has issued 1 million shares and XYZ has shorted 100K shares of ABC. There are now 1.1 million shares on the market held by "longs." If XYZ now acquires ABC for $10 per share, they will pay $10 million to the holders of the treasury issued shares and another $1 million to close out their short so that the holders of the 100k shares created by their short sale get their fair payout. Buying a company is not a way to escape the liability creating by a loss on a short sale.
"We'll have to wait and see, but the JCO has already published compellingly on Cobi, and the NCCN melanoma steering committee won't be far behind in making assessments."
NCCN can only recommend off label use of drugs that are already approved for other indications. Cobi has yet to be approved for any indication, so we will have to wait on the FDA for any commercial use.
"How much does this move share price any guesses. Wall Street has been pretty tough on this stock. It has been such a grind somehow in the face of good news I still worry."
CoBrim was pretty much a sure thing. It is still encouraging, and I like the fact that they mentioned the other 3 trial results still yet to come. I was worried they might be sitting on some bad news and were waiting for CoBrim to soften the blow. Here is what's yet to come.
EXAM medullary thyroid cancer survival data. My opinion is that there will be a nice survival trend that will fail to reach statsig.
Comet 1 OS data. Any statsig result will be appreciated. Any improvement over 2 months would be really good. This result will be out soon.
Comet 2. As the trial apparently has not reached full enrolment yet, this result will be out towards the end of the year. It has a very high probability of a successful result, but the regulatory pat is less clear cut than Comet 1.
They also mentioned anticipation for Meteor full enrolment. This trial exploits two strengths for Cabo. It has a good response rate and retains efficacy in patients previously treated with other VEGF TKI's. Meteor's primary endpoint is PFS and it does have an interim analysis. We could see results early in 2015 on this trial.
"Cobrim reached full enrolment of 500 in Feb 2013."
Sorry, that should be Feb 2014. Patients have been on trial in a range of 5-19 months.
"What is the number of events required to trigger interim analysis?" (CoBrim)
Unknown. Big Pharma typically does not publicly release this level of detail. Here is what we do know. Cobrim reached full enrolment of 500 in Feb 2013. Anticipated PFS for vemurafinib control arm is 6 mos and anticipated PFS for Vem/Cobi is 12+ mos. Trial started recruiting in Dec 2012 with enrolment likely backended. By now, I would guess 200+ events have been recorded. With this huge efficacy advantage, a highly significant p value would be evident with 100+ events. My guess is that the triggering event has occurred already. PFS analysis is complex and time consuming as it often involves independent review and adjudication of radiological data.
"Just a hunch, but I think we'll see either cobrim, comet or both read out in the next 30 days."
Better than a hunch, I think it a near certainty.
CoBrim will be an interim analysis for PFS. With a demonstrated 80+% objective response rate, it is inconceivable that there will be any result other than a highly significant PFS when the analysis is done. The only guidance is that Roche expects to file an NDA in 2014. They need the results in hand soon to make that deadline.
EXAM is overdue. The trial reached full enrolment in Feb 2011. At 6/15/2011 there had been 96 fatalities on trial. At 6/15/12 there had been 162 events. The final analysis will be triggered by the 217th event. There were 168 survivors as of 6/15/2012, over 2 years ago. The combined median OS at that time was 24 months. All surviving participants have been on trial a minimum of 40 months.
The Comet 1 interim topline was reported on 3/25/14, 3.4 months ago. That report was based on 387 events with the final to be based on an additional 191 events (578 total). EXEL's statistical assumption going into the trial was for a combined median survival of 8.5 months. If we make the sweeping assumption that the death rate is linear, that would be 56 deaths per month. Of course the survival curve is not linear, but for our purposes, it isn't far from it. The tail of the curve will be somewhat asymptotic, but we are not at that point in the trial yet. Simple math, based on these assumptions, would say that the 578th event will occur about 3.5 months after the 387th event. Even if EXEL's median OS estimate is overly conservative, a 10 month median would still place the topline announcement in July. There is nothing exact about this. The DSMB might take more or less time to report the result to EXEL and we don't know how prompt they will be in releasing the result. All said, I check the press releases every morning and I expect answers to all the major questions this month.
"...immunotherapy first then targeted therapies because immunotherapies grant a patient the chance at a durable remission and you don't want to deny someone the chance at that early on whereas absolutely everyone progresses on targeted therapies aimed at slowing cell proliferation"
That's the general notion for most indications and TKI's, but this may be an exception because of the very high response rates. Don't be surprised to see 80+% ORR from CoBrim. The thing about the immune therapy is that it not only prolongs survival, but is potentially curative. So the obvious question is whether reducing tumor burden with the TKI's first or in parallel might not improve on the immunotherapy complete response rate and increase the possibility of long term disease control and a possible cure for some patients. I'm sure I'm not the only one thinking about this. My guess is that there are already studies being planned to figure out the best sequence to use these very effective drugs optimally.
I think patients in an open label study are more likely to stay on drug and the rules are much more permissive wrt dose reduction and temporary discontinuation. Also the ph 3 had the requirement for progression within a 14 month window so patients with indolent disease were screened out. One of the things on which I based that guess is that the longest observed PFS duration for the EXAM Cabo arm at the PFS analysis was 19 months and patients were required to go off drug at progression. The minimum dose on EXAM was 60mg and the ph 1 allowed patients to go down to 20 mg. Any patients still on drug have been on trial for at least 40 months now. I'll stick with my guess that there aren't any left, but certainly acknowledge that it is a possibility that there could be a handful still on drug.
"Though I think Vemurafenib and Cobimetinib will pass CoBRIM, the potential market for it going forward in melanoma is going to be fairly limited."
The Ipi/Nivo combo will likely be the frontline drug treatment of choice. Remember though that those survival stats include a large number of patients who have progressed and are off treatment. Those with the Braf mutation will get one of the two Braf/Mek combinations. We don't have long term stats on the Ipi/Nivo patients, but I suspect a very high percentage will eventually progress and require a second line regimen. It will be interesting to see how the sequential and/or combination use of these drugs is worked out. It would not surprise me if the best potentially curative treatment would be the TKI's first to reduce tumor burden followed by the immune therapy.
"Take a look at the EMA review of Cabozantinib: WC500163705 dot pdf on page 58."
Thx. When the final analysis is done, that graph will be updated. The first 16 months will not change. From that point updated information on the censored patients will cause the curves to shift, hopefully with greater separation. As of 6/15/2011, 98 of the original 219 cabo arm patients were still receiving treatment. That was 3 years ago. It is unlikely (but certainly possible) that any are still on treatment. From 6/15/2011 to 6/15/2012 28% of the patents remaining on study died, leaving 168 patients still on study. 55 more deaths from among those remaining 168 (33%) will trigger the final analysis. It has been more than 24 months since the 2nd interim indicating that the curves have flattened out or EXEL has been sitting on the data for some time (or some of each). The report on the final topline result should happen any day now.
"Does anyone know what alpha was spent on the last EXAM OS interim? On another board, someone posted they thought it was highly likely only 0.0001 was spent on COMET 1 interim.--- followup responses were in disagreement with how the author came to this conclusion."
There would have been no alpha spent on the FDA mandated unscheduled interim. The spend for the first interim was variable based on the number of OS events at the time. The analysis was triggered by PFS events and the OS analysis was done simultaneously. There is still a full P=.04 or less remaining for the final OS analysis.
"""FWIW median OS on Cabo arm was 26 months and Placebo at 20 months at 162 events of the total 217 needed for final analysis."
If this ratio holds to the required event, will the trial have displayed statsig OS?
A 6 month comparative mOS survival advantage seems adequate to me...
Am I looking at this correctly?"""
Let's talk about this a little bit. First, I'd be surprised if they hit statsig. At the unscheduled interim they had a p value of about .25 and an HR of .825. With 50 more events to trigger the final, to hit statsig, that p value would have to come all the way down to .04 and in order to do that the HR would need to drop to something in the .70 range. It's possible, but not too likely. There have been two interims so far and there was a considerable improvement from he first to the second. I think there will be some improvement from the second to the final, but this is a small trial lacking the power to show moderate improvements at a statsig level. A 6 month improvement is definitely clinically relevant, but again, not enough to be statistically significant.
Because the first interim came about 4 months subsequent to full enrolment, we know that the first four months of the KM curves will show no treatment advantage. I'm working off memory, but I don't think there was a published KM curve showing the second interim OS data.
That the topline data is long overdue is a hopeful sign, but certainly not an indicator on which conclusions can be based. EXEL may just be sitting on the data waiting for an opportune time to release it. We should see soon enough.