It's official. You can ingnore clinical trial results from R&D and very early stage studies. Don't be investing ten's of millions of dollars in upfront payments. That's for sure.
A recent study confirms that 90% of early positive studies failed to contnue with such results in late stage trials. In fact, it appears that 98% of follow-on studies that previously saw an early stage "big effect" failed to reproduce it in susequent trials.
The reason almost certainly? Small sample numbers. You can't do adequate science with very small samples. The random positive responses tend to distort trial results until you build the numbers in late stage trials. Surprise, surprise given all the late stage crashes that we've seen over the past several years. I'm sure I could knock you over with a sledge hammer about now. See Fierce Biotech.
Posted by Bruce Lehr Oct 25th 2012.