This is not a recommendation to sell. Far from it. I am heavily invested. I am also bursting with confidence that we will have truly successful results. However, I feel somewhat of an obligation to at least mention some of the things that could go wrong and urge a little caution.
First, I can afford to lose money. That has not always been the case. I remember years back after making quite a chunk of change on a particular stock, then succumbing to the temptation of putting a lot of that money down on a stock I had not personally researched. The company was going to have a binary event. I put the money on it because a trusted friend told me it was a "sure" winner.
Well, the next day the FDA issued a CRL and I literally lost $70,000 in about 15 seconds. For many people that might represent 4-5 years of savings, loss of a college fund, delayed retirement, no vacation for several years, forget the new car, or telling your son or daughter that he/she cannot go to the college you promised. It truly is a sick feeling. At the time I really could not afford that loss. Initially I was not even mad. I was so stunned I was numb. About a week later I could not stop saying "How could you be so fk-ing stupid" a thousand times a day.
Okay, what could go wrong? Somethings we just don't know. We don't know how many patients have been censored and from which groups. The K-M estimator incorporates censored patients. Censored patients affects the numerator initially and the denominator in the next time frame. Since censored patients are eventually subtracted from the numerator and the denominator, it tends to decrease the fraction. Why? The numerator gets hit first. Now, you might say no big deal both control and treatment groups are subject to censoring. True but we do not know the breakdown of censored patients.
Second, and this bothers me a bit more than most things. The trial is powered at 80 percent.
Beta = 1-P (P = powered percentage), Beta = percent of Type II error.
Beta, or the Type II error criteria is 20 percent. That means there is a 20 percent chance we might incur a false negative. That is the analysis states the trial failed when in fact the data was good. We could have increased the power of the study but that would have required more patients, more money and more time.
The trial is being conducted at 79 different sites. I really don't see a problem here, especially since Celsion appears to have been bending over backwards to insure the quality of the trials and data. However, the logistics are dauting and you never know what might go wrong with so many different settings, labs, oncologist, technicians etc.
On the positive side....Well, there are just too many things and you heard them all before. However, I cannot resist mentioning the $75 million shelf registration. The company has literally pushed all its chips in on this one. But they have not even touched the $75 million. Even the "intelligent" shorts do not have an answer for this one. I personally have never seen a small biotech so confident. To me it is a statement of supreme confidence.
I will continue to examine anything I may have overlooked. I also want to thank you for your insights, thoughts and sharing your experience and knowledge. I have learned an incredible amount on this message board.
Be vigilant, don't stop doing DD and don't risk more than you can afford to lose in 15 seconds. This truly is an upcoming binary event.
Mgmt said from the time they approved the shelf they were not going to do anything with it until after the news was released. The fact that they have stuck to that and have been buying in leads me to believe positive results are ahead.
Sentiment: Strong Buy
dcup. One little thing to mention but we don't actually know that any institution wanted to buy any shares from the shelf at recent prices. From what we've read from Adam F Wall Street doesn't care about CLSN at this point
Sentiment: Strong Buy
They could have easily sold 12 million shares at say $6.50 per share. Hedge Funds would have immediately shorted the heck out of the stock from the mid $8s knowning they could just turn in their stock to cover when the pps hit $6.50 a share. Hedgies would have loved it.
They could have and still could raise the money if they wanted to bad enough. Of course shareholders would be up in arms.
Great post, D, and thank you for your honesty and continued due diligence in which you have shared with this message board. It appears that you are providing a caveat to your previous post, "Kaplan-Meier and Simulation models, model revisions, odds of success etc," and, more or less, simply restating Sia's statement from a few months ago that "anything can happen." GLTY.
Yes, Sia's "Anything can happen," Avii77's "Shyt happens," Alark13's "There are things we don't know, we don't know" and coutless others' "Don't risk more than you can afford to lose (in 15 seconds)" are all cautionary tales and reasons to use a little prudence even under the most promising circumstances. No matter how you 'slice it or dice it' this is a binary event. The trial results will have a major impact.
To the long-longs out there, the guys who have been battling out there for 5 to 10 years or more, make sure this is a victory for you no matter what happens. You have been 'taking hits' for so long and now are way ahead, don't let this turn into a nightmare. I am not going to be so arrogant as to tell you what to do but if you are up now, let's say $300,000 or $400,000, why not make sure you end up $200,000 ahead, even if the 'worst of the worst' case scenarios occurs. Cheers and GLTA.
“Second, and this bothers me a bit more than most things. The trial is powered at 80 percent….. the Type II error criteria is 20 percent. That means there is a 20 percent chance we might incur a false negative”
While it is true that the trial has a 20% chance of FAILING to detect a stat sig result (at the 5% level) if the TRUE efficacy of the drug is a 33% improvement over RFA alone, you might find some solace in learning that the trial has only a 10% chance of failing to detect a stat sig (SS) result if the true efficacy is a 39% improvement over control.
5% chance of failing to detect a SS result if the true efficacy is a 45% improvement
1% chance of failing to detect a SS result if the true efficacy is a 55% improvement
0.1% chance of failing to detect a SS result if the true efficacy is a 69% improvement
0.01% chance of failing to detect a SS result if the true efficacy is an 81% improvement
This is a powering statement, it says nothing about the chance of the trial succeeding because we do not yet have an estimate of the true efficacy (if any) of tdox.
But, reiterating the last example above. IF the true efficacy of TDOX represents an 81% improvement in PFS over RFA alone, the trial design (observing 390 events, 1:1 allocation) allows for us to detect that (stat sig at the 5% level) in the MEASURED PFS with a chance of a false negative of only 0.01% (as opposed to a 20% chance of a false negative if the true efficacy is only a 33% improvement in PFS)
I haven’t read all the posts here, but I am pretty sure your expectation of the efficacy of TDOX is in the neighborhood of a 100% improvement (or more?) over RFA alone. If that’s how you feel, you shouldn’t be worried so much about the false negative issue (but focus instead on that 100% improvement estimate).
Regarding your comment:” . I am also bursting with confidence that we will have truly successful results.” Been there, done that. You gave wise consul: Shyt happens. This biotech investing is not for the faint of heart. And for retail investors, not appropriate for anything more than discretionary funds (not risking college funds or your mortgage payment)
Thanks for putting things in perspective. I do think ThermoDox will provide a large improvement over RFA alone, especially so if control is anywhere near 12 PFS. Even if control is 18 PFS the results should be significant. Thanks for the comments and analysis. Excellent perspective. Cheers.
WHILE CAUTION IS A GOOD THING AT TIMES....I have a very good feeling about Celsion....very good......stay long....even after the positive data.....good things will happen
Sentiment: Strong Buy
Thanks for sharing your story and your model/intellect here at the YMB.
Wish I could have invested here much earlier like most seem to have done -- and be riding 'free-shares' to the event, that's the way most of these should be played. But the money available was tied up in ACAD and that worked out wonderfully; a portion of my earnings there have been placed here at the 7s range (since I couldn't buy sooner). Nevertheless, I am optimistic that this will play out, as management, the stock chart and the models suggest. If not, I will lose all I was willing to lose. Next time, I'll try to find out about an opportunity like this sooner. Good luck...!!
Thanks Dcups, I have posted myself next to my computer for that very reason. We do not stand a chance if a negative binary occurs. We - investors - will not be the first to know.
In your analysis did you account for the fact that the trial is skewed in favor of Tdox. i.e. greater than 3 cm and 4 tumors? Do we assume that a larger trial population is part of the data set? I am thinking that the control is 12 - 14 months and active is 40+. ??? 380 events occured mid October ??? I also remember that there was some confusion as to how many in the population actually evented or were misdiagnosed or dropped from the trial. Resulting in a slight delay in announcing 380 events. Never heard if any of that was true or not.
I suppose if we tried to take all these outlyers into account we would just be more confused.
Sentiment: Strong Buy
Whilst I agree with you about the shelf, I think there's a more compelling issue, possibly the issue that stopped them from touching the shelf... The DMC reviewed efficacy, and then suggested to the company, that they might want to take another look soon, and that there really wouldn't be much alpha spend. In the literature, for the stat model that CLSN is using, (Lan DeMets), is a curious statement, that if the stopping boundry is close at interim, that the DMC can, (perfectly validly!), suggest further looks, EVEN continual assesment, without risking the integrity of the trial... That's WHY they're not touching the shelf, and that is why I'm over-exposed. The stopping boundary at interim was MUCH tighter that at final. (You see DCups, I'm as bullish as anyone!).
Sentiment: Strong Buy
bty, I thought MT indicated that the request came from CLSN mgt for interim look..or do you think that was a ruse to keep the illusion of being blinded. I always thought they were getting pressure from large investors as to the reason...Nate August in particular.
never invest more then you can afford to lose in any bio ahead of a binary event. I have read BioTech sage's analysis and I have read yours and others and have decided the risk/reward ratio leans enough for the bull case to hold through data.
And yes I agree, The 75million shelf registration is a VERY powerful indication that the company either is A. Insane or B. KNOWS the data will meet the 33% threshold and beyond.
Unlike most I discount insider buying considering I have seen many bio companies simply do it just to attract the stock screeners and all the hot money that comes from a high insider buy / sell ratio
GLTA we shall know in 20 days max
Sentiment: Strong Buy
your point is well taken about insider trades being used to attract hot money and raising pps... the problem with the CLSN story is that having done that (if in fact that was what they were doing) they then failed to utilize the higher pps to raise funds in front of the binary event in order to protect their jobs and the company... management acted foolishly and irresponsibly in this case... unless they have a very good reason to be confident... I usually discount insider buys somewhat as well... but this one has been a little different...