I don't know when Bear eats, sleeps or works!  A great response to Cabel's question.  Here's a bit more simplistic response that I drafted that may help folks confused about the basics to understand:

Half the patients are on placebo and half are on apabetalone.  So, at any particular point in time, the overall event rate is the average of the two (placebo and apabetalone) event rates.  In your example of an 11% event rate in the placebo group, the event rate in the apabetalone group would be 3.4% to provide the 7.2% average.  That is close to a 70% RRR at this particular time, but it is unlikely that this differential (if it even exists now) would have been place throughout the trial.  Even if apabetalone is effective, many MACE events would be likely to take place in the apabetalone group early in the trial before the physiological changes induced by the drug had time to change the adverse course of events in some of the high risk patients.  Also, I would not necessarily expect the differential to be so great this close to the end of a relatively long trial because many of the placebo patients who were destined to have a MACE likely have had it by now since the group is so high risk – so the placebo rate may also be dropping.

A more reasonable (but still wildly optimistic) guess would be that apabetalone achieves a 50% RRR over the time span of the trial.  In that case, one-third of the MACE events (167) would have occurred in the apabetalone group and two-thirds (333) in the control placebo group.  As Bear has pointed out, this result needs to be statistically significant for the trial to be considered a success and the test for significance will be dependent on a number of factors in the experimental design.  However, I a pretty sure the trial was designed (i.e. the end-point of 250 MACE events was chosen) to ensure that a 50% reduction in MACE would be significant.