Amarin Corp Plc (AMRN)

[sts66]

That's not the correct post, doesn't contain a link to Bhatt's paper - I missed that thread too, can anyone post a link to what Bhatt wrote?

Nevermind - just googled "Deepak Bhatt twitter", first link showed what appears to be excerpts of his last three tweets, and one of them leads directly to the paper in question:

Federal Judge Invalidates Icosapent Ethyl Patents – But Based on a Common Statistical Mistake @SSRN https://t.co/t3iNf2RTo5 #cardiology #cardiotwitter

— Dr. Deepak L. Bhatt (@DLBHATTMD) June 5, 2020

The paper abstract (at bottom of page you can download the entire thing):

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3618671#.XtqhCv2KLeU.twitter

Abstract

In a recent decision that stunned many in the legal and pharmaceutical communities, the chief judge of the U.S. District Court of the District of Nevada in Las Vegas, Judge Miranda Du, struck down 6 key method-of-use patents for the omega-3 fatty acid preparation, icosapent ethyl (Vascepa®). If this judgment is allowed to stand on appeal in Amarin Pharma, Inc., v. Hikma Pharmaceuticals USA, Inc., et al., it will have major implications not only for the patent protection of this drug but for pharmaceutical patents and biomedical innovation more generally.

Crux of the author's argument:

We conclude that the Judge’s opinion on obviousness was based
on a statistically incorrect interpretation of the results presented in Mori
et al.5, which she relied upon as prior art. Statistically significant
difference within one arm and no difference within the other arm do not
imply a statistically significant difference between the arms. In fact, if
Mori et al.5 had conducted the appropriate statistical test for difference
between arms, most likely the opposite conclusion would have been
reached, and it would not have been possible to claim that it was obvious
that there were differential effects of DHA and EPA on LDL levels. The
correct analysis and null conclusion would have been unlikely to
motivate further research.

The proper statistical analysis that we conducted is not a highly
technical analysis; it would be known not only by expert biostatisticians
but also by persons having ordinary statistical knowledge. A person with
a general familiarity with statistical analysis would have questioned the
use of Mori et al.5 as valid prior art, based both on the invalid statistical
analysis and the hypothesis-generating nature of the study. To provide a
sound legal interpretation of the validity of the icosapent ethyl patents,
we believe that Judge Du’s determination of obviousness should be
reviewed by the U.S. Court of Appeals for the Federal Circuit in the
light of the flawed statistical analysis in the small study by Mori et al.5,
according to the “clear error” standard.

We review the background of this judicial opinion and suggest the decision should be reversed on appeal. The Judge based her opinion on her conclusion that the patents, at the time of the invention, would have been obvious to a person having ordinary skill in the art (35 U.S.C. § 103). As we will show, a key piece of the prior art, which was central to this case, included an incorrect conclusion based on a common statistical error. Our central argument is that prior art that is scientifically incorrect should not be an acceptable basis for invalidating patents as obvious.

Bhatt replied to someone that he "sent it to her", which I assume means Du - why? Nothing she can do about it now unless the CAFC remands back to her.