"We have no information other than that the trial failed to meet 30%."

This is not strictly correct (although it certainly may be true that apabetalone did not achieve a 30% RRR).

What we know is that apabetalone did not increase the time-to-first-MACE at P<0.05.  We don't know what RRR was achieved (that was not the endpoint).  Remember that this was a (relatively) small, short CVOT trial with very sick patients.  Although Bear pointed out the problem with comparing the time-to-death profle with that of other trials (with other durgs and other patient groups) the time-to-death profile posted on Twitter yesterday (and linked from this forum) was very telling.  Even if apabetalone does cause a good RRR, the trial may not have run long enough to detect an extended time to event with the required degree of certainty.

If, for example, between 10 and 20% of the MACE events took place randomly in the apabealone and placebo arms of the trial in the first few weeks or months of the trial before apabetalone could make sufficient physiological changes to achieve MACE prevention or delay, those early deaths wiould delay the separation of the placebo and treatement time-to-death curves that was required for the successful outcome to be demonstrated with "academic rigour." 

Jupe