Koo,

You said: Should you have said that a successful topline is achieved if a statistically significant increase in time to the first occurrence is achieved, regardless of the overall %RRR?

Yep, indeed I should.  With respect to the parameter used in the determination of statistical differentiation between the ABL and placebo groups, I suspect that it is time-to-occurrence since that is the commonly stated endpoint.  However, Bear has me convinced that there is pretty much a standard format for analysis of CVOT trials that is employed in most studies.   Topline success may be determined by increased time-to-occurrence, but the companies may choose to report a variety of numbers, including %RRR.  After all, it sounds much better to be able say the the treatment reduced the risk of MACE (or number of MACE) instead of the treatment increased the time to a MACE. 

I also agree with Bear that knowing if we have exceeded the expected 'patient years' would be an excellent way of knowing whether  the trial really is 'taking longer than expected' or whether real-life factors such as slow enollment or high dropouts have just extended the timeframe for the required 250 MACE.  And perhaps there is yet another time factor here - how many 'forum posting hours' do shareholders need to put in for each 'first-in'class' drug that makes it to market?

Cheers,

Jupe