"The other thing is: What on earth happened after 24 months??? Up untill 24 months, there is a stellar performance of Apabetalone, but right after 24 months, the event rate in the treatment arm takes off..."

Yep, at face value that is an uncomfortable looking curve.  However, recall that median dosing was 26 months so the number of patients (in both arms) that got beyond this point is much smaller than it was in earlier time periods.  Enrollment into the trial took place over time but on those curves, every patient is present in the first time period but the ones that enrolled later and spent less time in the trial aren't present in the later time periods.  See how the 'number at risk' numbers on the bottom of the slides decline by time period.  So the curves become less reliable as time goes on.  More patients and/or more time would likely have produced a smoother curve...

I totally agree with your point about the CKD population accounting for most of the MACE result.  Although I have aready pointed out that the 50% RRR needs to be viewed cautiously, it is pretty enticing.

Jupe