Amarin Corp Plc (AMRN)

[MedResCollab]

chas123,

We never said K/M curves were tea leaves, Charles, only this unscientific interpretation of them you are ascribing to. But, since you are focused on their shape so much, we’ll allow that there are a handful of normalities associated with these curves we could discuss. For example, very large trials will have a much more steady-looking curve than those from smaller studies. This is because the event rate does not fluctuate nearly as much. And there won’t be much of a “tail,” as is often seen with smaller studies, and in particular from immunotherapy trials. But, these are all very general observations and do not at all allow one to make sweeping statements about a given arm of a trial, as you are attempting to. You even go so far as to say:

If MO had the immediate and large deadly effect you claim, in the initial period before most of the V benefit had had time to influence event rates, the P curve would have to start out high above the V curve. It didn't. -chas123

No, it simply takes time for events to occur. When subjects enter a study they are all screened as healthy and stable enough to be in it. It’s fully in line with the theory that MO inhibited cardiac medications, such as statins, where the full benefits takes years to realize. The rest of your post continues on the same faulty premise.

You cannot read into the K/M curves what you want to be there. Only significant hazard ratios are reliable.