Biotech Values

[terry hallinan]

DewDiligence - not KERX,

As with the issue of interim looks (#msg-8530993), this is a matter of degree.

I would challenge you to find a single credible source that would defend the FDA's allocation of p values in the case of interim looks.

As a matter of fact, a DSMB is empowered to look at the data at any time it feels called on to do so. It must have that power in order to properly monitor the safety of volunteers. The safety of volunteers is written into its very name.

You offer nothing but anecdotal evidence for your conclusion in the case of Biomira. They were unsuccessful at the time of the interim look and therefore the trial was almost surely a bust.

It did indeed turn out disastrous but just maybe that had something to do with the design of the trial. Even if Biomira had not been so heavily penalized by the math illiterates at the FDA in the allocation of p-value, the trial was still a dud.

Let me offer you a bit stronger anecdotal evidence to the contrary. The CEO and other executives were buying before the results were known. You think they did not have a real statistician or two to consult? Please note I didn't say they had a competent one.

The DSMB advised the trial should continue. Why did they advise such a thing if the trial was surely doomed according to the deep calculations of Dew Diligence? Some trials have been discontinued in such a case.

Long before the trial ended, before the interim look as best I can recall, a knowledgeable Ph.D. involved in numerous trials advised he had sold all his shares and advised others to do so after getting a look at a supposedly purloined copy of the trial design that was easily available. Wish I had listened of course.

Without any doubt there is a reason for penalizing formalized interim looks at the data but the numbers are all out of whack as is usual with the FDA. The idea is that the more times one looks, the better chance there is to find a point at which the data is statistically significant without regard to the ultimate truth. But the DSMB can take a look any time it has a hankering to do so though discouraged from doing so.

Just quibbling, Dew, over a pet peeve.

Sports statistics are probably more intensively studied than just about anything and are not always terribly meaningful. I got tired of being told how poor John Elway's statistics were. I would have bet on him nevertheless without regard to much of anything else.

Statistics are most meaningful to those who know least about them.

BTW I got a picture for you that might help with your analysis of biotechs. You might notice folks were smiling for the camera. They had reason to. Came a fur piece many of them did, including some from Germany.

http://www.ragingbull.lycos.com/mboard/boards.cgi?board=BIPH&read=77630

Picture of bagholders ...

http://groups.msn.com/TheBiophanAttic/shoebox.msnw?action=ShowPhoto&PhotoID=30

Those "bagholders" seem to have done rather well, including the CEO pictured at the top left. The picture was taken before the annual party - er, meeting of Biophan. There are plenty more pictures of the party - er, meeting.

Since that annual party - er, meeting some have left and others have perhaps learned to smile less because the price of the stock has been cut in half. Still most are still smiling I reckon. This old fool is.

Have you got a statistical analysis handy that will tell more than that picture? That would be worth a fortune.

Even a onetime photogrammetrist is stumped.

Best, Terry