Whose Numbers Do You Believe?

One of the more interesting episodes in Pharma is whether Vascepa's brand new trial results would be positive or negative. And whether the share price would rocket up or plummet down respectively. Adam Feuerstein has been something of a skeptic on the fish oil, and now that the results of the trial have been made public over the weekend, great argument has ensued.

You can find all you want to know about Vascepa and Amarin, its maker, with a quick innterTubes search, so I won't go into all of that. But what is of interest here: being a blinded, placebo controlled trial, the efficacy numbers for the fish oil are measured relative to the patients getting the placebo. Almost all the time, there is no dispute over the placebo (the substance itself, not other factors such as administration or detection), but not this time. There's reasoned question whether the placebo used was really inert. When is the data just bad?

Here's a sample of the argument courtesy of Feuerstein. One might not be surprised to see battling forces. It's funny. And demonstrates that something as seemingly simple as testing a compound for efficacy most likely isn't. You would think that totally hiding which compound the patient gets is the first order of business, and that the placebo is truly inert (at least, for the indication under study). Guess not. The capsule used as placebo needed to look like Vascepa, so mineral oil was chosen. Turns out now that this is widely known, it is also widely known that mineral oil has been shown (in other studies) to be a bit of an anti-stain. Oops. Evidently, neither Amarin nor the FDA made the content of the placebo widely public, or all those who should know better ignored the issue. Let the jousting begin. Will FDA now change its mind?? Will the company go bankrupt?? And so on. All because the yardstick might be a foot short. Measure twice, cut once.