Exactly what I said, and the remaining are all treatment arm, rather than randomized. The 17 missing would have all had to have been placebo had the trial continued enrolling placebo patients. Instead, the DSMB “randomized” them all to the treatment arm, and that, I believe, is meaningful. I have changed my ethical interpretation slightly, but not much. When you know you no longer have a placebo arm, when everyone has crossed over, and the measure you used to allow crossover is confounded, to randomize to placebo arm when you are now going for survival is unethical.
The regulator does not necessarily have to explain that. Effectively there are different terms that can be used but some terms might be wrongly used by shorts, and it is likely that regulators knew that to allow for an adaptive trial to play out properly like this, they can’t unblind to sponsor or the doctors managing the trial. Hence the mystery.
The randomization to placebo at that stage clearly would not be appropriate, delaying care and giving those patients potentially what appeared to be a lesser chance at long-term survival, for no actual benefit. At the time recurrence was not a concept for the trial to measure. Plus these were newly diagnosed so delaying their care as an experiment so that you could test only recurrence would also be unethical.
So the partial halt.
News 
Market Data 
Discover 
AI
