blade re DNDN briefing docs:
The conclusion in the FDA's briefing docs were actually quite favorable to DNDN, stating that all things considered, it appeared that the chances of the survival benefit achieved in 9901 being a false positive could not be 1 in 1600 (because not primary endpoint), but instead could be estimated as closer to 1 in 40. Also in the conclusion was the statement that all of the other endpoints in the two trials, while not stat sig, trended favorably in Provenge's direction. The stock price reaction the day the docs came out was a gain of approximately 8-15% as I recall.
The relative favorability of the FDA statistician's briefing docs toward Provenge was one of the factors in my decision to load up prior to the panel meeting, and hold all of my common through the final decision. It appeared that despite all of the biostatistical rules that were broken in the Provenge BLA, CBER was favorably inclined to approve. They could have savaged the application along the lines of what was done to the applications of DORB, ALTH, ABT's Xinlay, and IDMI...but they didn't.
