George,

I won't touch on 2-4, but I will comment on #1. 

"Time. To get a positive BoM, placebo needs to get to 143-150 MACE before end of dosing (250 events). Using EXAMINE rates, we're already there. BoM rates should be equal or worse than EXAMINE, that potentially leaves RVX with a buffer."

Really the only accurate way to compare event rates between BETonMACE and other past trials, including EXAMINE, is to know how many patient years BETonMACE currently is at. And we don't know that. I appreciate your efforts though. I've gone through several similar exercises on this myself. But without a solid number on the # of patients years in BETonMACE, it is really tough to compare to other trials. 

BETonMACE was modeled to give 250 events with 3600 patient years based upon 2400 patients dosed for a median of 1.5 years, which would give overall (placebo and apabetalone groups combined) event rate of 6.94 events per 100 patient years (250/36). However, the AHA poster from November 2018 stated "With an assumed primary event rate of 7 per 100 patient years in the placebo group." What is odd is that both assumed rates are ~ 7 per 100 patient years but one is for combined groups and other is for placebo only. I'm not sure what to make of that. BETonMACE could currently be under or over 3600 patient years. If they haven't yet surpassed 3600 patient years, then not yet hitting 250 3-point MACE events could mean they are still on track with the original projections from a few years ago.

EXAMINE is a good model, but it is only one trial. Times change. Standard of care changes. EXAMINE started enrolling almost 10 years ago in October 2009. There would be a lot more confidence in the reproducibility of EXAMINE's event rates if there were more trials with patient populations like EXAMINE (diabetic with ACS event w/i 90 days). EXAMINE placebo group had an event rate of ~8.1 events per 100 patient years. The closest comparator is ELIXA that enrolled diabetic patients with ACS events w/i 180 days (started enrolling in July 2010) and had a placebo event rate of ~6.4 events per 100 patient years. ELIXA was looking a 4-point MACE, but very few events were of the non-3-point MACE variety, so safe the 3-point and 4-point MACE events rates for ELIXA are essentially the same. That AHA 2018 statement above with the 7 per 100 patient year estimate is kind of right smack in the middle of these two. 8, 7, or 6? We don't know yet. 

Yes, BETonMACE has the additional low-HDL requirement and the median time from ACS occurence to enrollment in BETonMACE is shorter in BETonMACE than either EXAMINE or ELIXA. However, one has to ask oneself how reproducible are these assumed/modeled placebo event rates? Not knowing the status of patient years in BETonMACE and not knowing the true BETonMACE placebo event rate makes interpretation difficult.

BearDownAZ