Amarin Corp Plc (AMRN)

[iwfal]

Either show any proof (not a "DISCUSSION") or quit FUD'ing.

Tsk, tsk. Even without the below, there is plenty of reason to think there is likely an issue - but I found the numbers in the Jelis paper:

Haemorrhage (cerebral, fundal, epistaxis,subcutaneous)
60 (0·6%) 105 (1·1%) (p=)0·0006

Note that, interestingly, both the Vascepa data and the Jelis data line up at about HR=2. Again, this expands the credibility of the data - even if Jelis hadn't been stat sig there would be significant reason to believe there is some risk this shows up in REDUCE IT. Risk is not B/W.

On some more optimistic notes:

a) given the rates of bleeds in Jelis per man-year and the fact that I'd expect REDUCE IT to have significantly fewer man-years (about 1/3), I'd expect bleeding to be only somewhere near borderline stat sig (not the very stat sig value it was in Jelis)

b) Since REDUCE-IT has a significantly sicker population, the number that benefit vs the number hurt by bleeding should be better than Jelis (unless many of the sicker patients also come with Atrial Fib).