Amarin Corp Plc (AMRN)

[jomama9231]

IR/Covington- Help from the Board before I send this to them. Any edits just reply to this chain. Thanks in advance to everyone.

As a physician who has read all of the data and all of the briefs in this case, please review the below fatal errors in the patent case that may be of assistance to Amarin.

In Judge Du’s Decision page 30 Lines 14-16, she states “In light of the statistically-significant differential effects reported between the EPA and control groups, a POSA would have attributed the reduction in Apo-B to EPA. (Id. at 737:24-738:8.)
Judge Du makes a fact error because the above statement is entirely incorrect. There was no statistically significant difference between the EPA group (defined as EPA +estriol) and the control group (estriol alone). In fact, if you read the Kurabayashi study, APo-B reduction is mentioned no where in the discussion section or the conclusion of the article. It is merely mentioned in one line in the results section (3rd paragraph from the Results section) and states, “The apolipoprotein B level in the eicosapentaenoic acid group was significantly lower at week 48 compared with the baseline level, but there was no significant difference between the groups.” Let me emphasize the second part of that statement, “but there was no significant difference between the groups.” This carries tremendous weight. Furthermore, if you look at Table 3 which Judge Du also has in her decision, there was no statistically significant difference in Apo-B levels between the “EPA group” and the “control group” at every time interval measured. Please see “NS” (Not significant) underneath week 12, week 24, and week 48 of the Table.
Scientifically, you cannot compare a patients baseline numbers to their numbers at week 48 and conclude anything of significance. Comparisons must be done to a control group and this is well known. One citation of a multitude can be seen here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3286439/. titled, “ Comparisons against baseline within randomised groups are often used and can be highly misleading.” The author concludes ,“ Randomised groups should be compared directly by two-sample methods and separate tests against baseline are highly misleading.”
Another error within the Apo-B discussion occurs. The judge points to a reduction in Apo-B with a p value of <0.001 as being ?“highly significant”. However, the flaw here is the within group comparison is a well known misleading conclusion. Please see another example below with the same exact p value quoted by Judge Du (p<0.001), yet concluding there was not even “good evidence” of a between group difference.
“In a randomized trial of morphine vs. placebo for the anaesthesia of mechanically ventilated pre-term babies, it was reported that morphine-treated babies showed a significant reduction in adrenaline concentrations during the first 24 hours (median change -0.4 nmol/L, P < 0.001), which was not seen in the placebo group (median change 0.2 nmol/L, P < 0.79 (sic)) [15]. There is no way to test whether the between group difference is significant. Even though the median changes in this example are in opposite directions, this does not imply that there is good evidence that the treatments are different ( https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3286439/)

Aside from the fact errors by Judge Du, the Kurabayashi article in and of itself has several limitations which severely impair any results of the article even those accepted as valid by Judge Du. The trial is Japanese only, women only, limited to post menopausal females, small in size, and in a non-peer reviewed nutritional journal. Therefore, any data taken from this article would not make a POSA feel they have a reasonable chance of success in reducing Apo- B, let alone reducing it in a TG>500 population never before studied. The trial was considered by the patent examiner and referenced in the patents (Despite Judge Du incorrectly stating otherwise). It should be clear that Amarin’s own Marine trial failed to reduce Apo-B with its 2 gram dose and only showed this non obvious unexpected benefit with its 4 gram dose. In conclusion, a POSA would conclude exactly what the patent examiner concluded, “
The prior art is either silent or teaches that there is no statistically significant change in Apo-B levels when patients with TG levels less than 150 mg/dl or between 150-499 mg/dl are treated with either 96% pure ethyl-EPA or a mixture of ethyl-EPA and DHA, or when a mixture of ethyl-EPA and DHA was administered to patients with TG levels above 500 mg/dl (see item 25 on page 6 and Table 1 of the Bays’ declaration dated 05/16/2012; see also Table A on page 15 of Applicant’s response dated 01/13/2012). Applicant also presented convincing arguments (see Bays' declaration dated 05/16/2012, items 10 through 24 on pages 2 through 6) against the three references presented by the Examiner (Connor et. al., Fisher et. al. and Park et. al.) regarding the predictability of lowering Apo-B with 96% pure ethyl EPA in patients with TG levels above 500 mg/dl (see non-final rejection dated 03/02/2012, pages 20-23).

This statement is inclusive of Kurabayashi as no pure ethyl-EPA drug showed a reduction in Apo-B. Finally, if the initial patent examiner granted the patent essentially based on “long felt and unmet need” and the “unexpected benefit,” and Amarin now has in its favor those two conditions plus commercial success (granted by Judge Du), there is no clear and convincing evidence of obviousness to invalidate the Amarin patents.