"I keep wondering if this is a potential risk for BETonMACE. The primary endpoint is the Time to the first occurrence of Major Adverse Cardiovascular Events (MACE). Could this be a problem? Does it help us to meet the endpoint or not?"

Time to first occurence of MACE is standard protocol for analysis of cardiovascular outcomes in clinical trials. Nothing special about BETonMACE in that regard. 

"Because, the way I see RRR calculated here is with events. 150 events on placebo, 100 on ABL, 33% RRR. Fine. But thats juste relative events on each side. Is no events = x amount of time on the trial? Will just people that had an event will count on the RRR calculation? Maybe I missed something."

You need the time component in order to generate the Kaplan-Meier plots that show the event accumulation in each group over time. Take a look at the publication of any major cardio outcomes trial (for example Figure 1 in EMPA-REG OUTCOME). What you want to see is the two groups separating over time. Knowing if the groups separate earlier vs. later in the trial can also be informative for interpreting the clinical results. I think the statistical analysis in CVOTs takes into account the separation of the curves over time. It is not just reliant on a static number of events in each group independent of time. You need that time component. Remember, patients are going to be on drug for between ~9 months and ~38 months. 

Often times, the %RRR is reported at the time period corresponding to median drug exposure in the trial. For example, the event rates, hazard ratios and %RRRs shown in Table 1 in EMPA-REG OUTCOME correspond to data at this median drug exposure period of ~ 2.6 years. However, as shown in Figure 1A one can see that the Kaplan-Meier plot goes out to 48 months at which time the curves separate even more. But not all patients in EMPA-REG OUTCOME were followed out to 48 months.

I hope that helps!

BDAZ