It's fun to play with numbers. But keep in mind that:

1) The placebo event rate is unknown. EXAMINE provides the best example but that is just one trial and its event rate may not be reproducible. BETonMACE has additional low-HDL requirement that may affect the placebo event rate to an uncertain degree.

2) The apabetalone effect size is unknown. Could be 0% RRR. Could be 50% RRR. BETonMACE was modeled for a 30% RRR but we don't know what the actual %RRR will be. 

3) Discontinuations/Dropouts are unknown. This greatly impacts patient year estimates.

4) Enrollment updates are vague. Because patient enrollment is reported in "waves", one does not know how many patient were added towards the beginning of the wave or towards the end of the wave. One could over- or under-estimate enrollment and patient years based on assumptions.

5) Total MACE events are unknown. We know it will be 250 or greater. We don't know much beyond that.

So have fun with numbers, but keep the limitations of your assumptions in mind.

BDAZ