BTW, if the number of dropouts is higher (closer to typical given numbers that Bear has provided in the past) and 2000 patients stayed in the trial, the reductions in MACE for the apabetalone group at placebo rates of 7% and 8% drop to 26% and 47%, respectively.  This is due to a drop in total patient years (from 4950 to 4500). Hence, my concern about trial size and duration. 

The pre-trial assumption was a placebo MACE rate of 8% so all is good unless there was an unusually low MACE rate in the placebo arm of the trial.  This is unliekly given the very sick nature of trial participants.  And, to be clear, even a 26% reduction in MACE (and its associated effect on time-to-MACE) may be significant, it is just less likely to be. 

Jupe