Replies to post #104600 on Amarin Corp Plc (AMRN)

[Whalatane]

Chas ..interesting post

Bottom line ...more time is better ?

Kiwi

[jessellivermore]

Chas..

Very nice post...I'm agree with virtually all your points and have expressed my opinion that the trial has good chance of running to completion..

The consideration of the time it takes for EPA to work it magic may not be the real issue. IMO almost all the effects of EPA occur within a relatively short time (within a few months at most) after starting on EPA. What I interpret from the literature and what I have seen from own experience of treating patients with V. The inflammatory system design is pretty much the same in all organs (in all cells)..and is locally based. So the functional or structural damage (loss) in the separate organs is more related to what their function was originally... What EPA does is lower inflammation and speed up recovery in areas which have been damaged..For example relief from DES takes less than three weeks. Resolution of injury due to interruption of blood flow can be seen in cutaneous injuries in two or three weeks...EPA uses the same receptors that aspirin, and steroids use and we know how quickly they respond.

The divergences seen in long term CVOT between events in the active and passive arms are much more likely to result from the progress of the disease (inflammation) on the unprotected organs of those in the placebo arm. So the benefits of EPA are seen very early and remain as long as you are on EPA and provide on-going protection that is absent in the placebo arm... The therapeutic effect begins early in the V arm. What takes time, is the disease process occurs gradually until it reaches a critical stage. This could be the point at which the atheromatous plaque ruptures into a coronary vessel...Being on V could have a number of protective advantages..It could prevent the plaque from developing. It might stabilize an existing plaque and prevent the rupture. It could limit the extent of myocardial damage by suppressing the expression of harmful cytokines. All these effects would offer the members of the active arm a protection that the placebo group did not enjoy...But it takes time for these events to occur...The longer the time the more events.

":>) JL

[jfmcrr]

the RI primary endpoint, it forms an evolving estimate of the true value that becomes more reliable and precise as the trial progresses.

Within limits; you run the study to the last man drops, you get a perfect evaluation and characterisation of the sample. Reduce It is supposed to generalize to a larger population but it is good to keep the basics in mind...

And the rest of the post are a nice look past the top line (N=8000, T= 4yrs +/-).

Whew. if this was concrete rather than concept... Reduce It has been handled, taken apart and put back together, tasted, and smelled, distilled and reconstituted to the point that you could read through it. Ending the study will be an act of mercy.

Mercy, Mercy ,Mercy.

[oneragman]

Chas,
Good point on patient years. A couple of things:

Quote:"One subgroup could be less than stellar, or there could have been too many "soft" events."
-Even though a subgroup might be less than stellar, the DMC could determine that running the trial to completion will not change it's significance. Stroke is 12% at the 60% IA and 13% at 80%. Going to 100%(more time) will not really move the needle, but with 2 readings at 80%, you pretty much know where it will end up. Only if borderline and trending up or down would a particular SE affect a determination of whether to continue. A RRR of greater than 30% should leave few if any to be questionable.
-We don't know the breakdown between hard and soft Mace, but on a composite basis, AMRN does. This must not be a factor or they wouldn't be increasing inventory now.
-When looking at the JELIS graphs, I think of it as comparing 2 grams vs. 4 grams of EPA due to Japanese diet compared to 0 vs 4 in RI. The separation should be greater in RI.

[Biobillionair]

Very good post, and I like to add Amarin moved up the proxy for a reason; perhaps management wanted to get R-I success bonuses votes in prior to an impending 80% stop?
BB

[VuBru]

Chas - Good points.

When an early stop depends on sufficiently high estimates of multiple endpoints, the odds change dramatically. With three endpoints, for example, the odds of the estimate of at least one being significantly low go from 40% for a single endpoint to about 80% (100 x (1 - .6^3) = 78.4). That is, from probably no problem, to very likely a problem.

We know from the design paper that the "key SEs" are the individual MACE components. So, in your calculations above, the multiple endpoints are not independent, which would probably reduce the odds somewhat.