Amarin Corp Plc (AMRN)

[Atom0aks]

Type 1 error was well controlled.

Please take a close look at Fig 4 from the NEJM paper on R-IT. That is what was formally tested.

The Primary Prevention population (T2DM w/risk factor) alone was not formally tested.

Those patients were part of the trial. And the trial was successful.

Now, if you want to say they (the FDA) are seeking guidance on potentially limiting approval to only a subset of those tested (secondary prevention) that may well be correct.

But the justification behind that is weak. (On the primary 5 pt MACE the interaction term was p=0.14 and the protocol says below 0.15 would be considered significant, so yes, there is evidence in 5 pt MACE that there is a difference in efficacy between populations. The counter argument to that would be that the FDA focuses on Hard MACE. And the Hard MACE primary prevention data is compelling (with no SS interaction test).

R-IT's primary prevention cohort is a relatively high-risk patient population with a 10 year risk of developing ASCVD of 20% (Calc'd by doubling the 5 year event rate of 3 pt MACE in placebo in the primary prevention cohort).

They may be asked to balance the benefit with the risk (afib and bleeding). They may be asked to balance the uncertainty that MO introduces to the benefit in this PP population.

The NNT for PP in Hard MACE is 63 with a NNH (number needed to harm) of 71 (afib) and 167 (bleeding).

That's reasonable. The text in bold is indeed what I am predicting to be the question that's presented to the committee.

You guys discussing the IMPROVE-IT Trial and Ezetimibe is not really relevant. IMPROVE-IT was a secondary prevention trial in high risk patients (all had been hospitalized for ACS with 10 days of randomization). It went before the same ADCOM with their 5% RRR on hard MACE and it was voted down; they did not get the label for preventing cardiac events. R-IT saw 19% RRR on Hard MACE in our primary prevention cohort.

I did not use it as a 1:1 comparison. I wanted to illustrate that clinical context is also important, as discussed in the committee's rationale, not just numbers.