Skint, you have a depressing tendency to play the man rather...

Skint, you have a depressing tendency to play the man rather than the ball when the facts don't fit.

I currently hold no stock, with a hold sentiment for those that might.

I would have thought any drug that can demonstrate improved cognitive outcomes where it claims to have a disease-modifying MOA would 'appear to work.'

What I posted was taken from the link to Alzforum that remains current. If you dispute what is posted on the site, take it up with them, not me.

Do you not think that Lily, etc, would have 'solid lab work,' as multinationals many times the size of Prana, to back their drug prior to investing $M's in a Phase III trial?

Hark back to PBTI all you like. The difference in ADAS-Cog you referred to was demonstrated only in the more severely affected patient subgroup. There is no logical reason why this should be.

How many patients continued in the 84-week extension? How many of these were part of the "more severely affected subgroup?" isn't it more likely that those who were doing better at the end of the double-blind phase stayed on to participate in the open-label extension, and thus represented a sample biased towards those who were doing well anyway.

I'd appreciate your enlightening us on what the difference IS between a post-hoc analysis and data mining, as you assert they are different.

Alois