Thanks for this Bear.  I did some number crunching based on data provided by RVX.  Here is what RVX said, based on various press releases and MD&A from annual reports:

Nov 15 2015.  First patient enrolled

Aug 16 2016. Crossed 1,000 patients.

March 17 2017. Crossed 1,200 patients

July 10 2017. Crossed 1,600 patients

Nov 1 2017.  Crossed 1,950 patients

March 18 2018. Peaked at 2,425 patients.  Question: there has been chatter re Taiwan enrollment but I could not find it in the press releases.

April 18 2019. Closed at 1,925 patients.

The difference between April 18 2019 and March 18 2018 is startling.  Because the trial stopped enrolling patients, they started losing patients at ONE patient a day!  What a sick patient population this is.  But this also suggests that they must have been losing that many patients even earlier.

Anyway, based on the loss of 1 patient a day for almost the entire trial, I get to 3,400 patient years as of June 2018.  But this is 1,000 more than the 2,400 patient years that Tundup talked about, based on his June 21 2018 write up.  So something does not add up.  But then Tundup says 3,300 patients by the end of 2018.  I have it at 4,500 patient years by 2018 end.  So the increments that I get (1,000 patient years) is what Tundup says. There is a level issue, not a gradient issue.

So if I do the calculations based on my estimates, I get to 5,200 patient years by the end of the trial in April end 2019.  If I deduct the 1,100 patient year from this, we get to 4,100 patient years. 

That is awesome.  At 4,100 patient years, a 7.2 MACE rate for placebo, we would get RRR of 28%, if I assume 260 total MACE.  Because RVX would get 80% power at 3600 patient years, the confidence level goes up with patient years.  At 4,200 patient years, we are 16% more in time elapsed.  

Sensitivity analysis suggests that the more the patient years, the better for RVX.  Every 100 patient years is about 4% RRR increment. 

Also, the higher the MACE rate for placebo, the better for RVX.  For 1,950 patients (say 50% on placebo), would suggest 70 MACE a year..but given the drop out rate of 1 a day, a 7 MACE rate for placebo is too low IMHO.  A MACE rate going from 7 to 8 with 260 MACE events and 4,100 patient years pushes RRR from 19% to 44%.

My take away is that the key metric that will make the most difference is the MACE rate for the placebo.  The higher, the better for RVX.  And 7 is just too low, IMHO.