On September 3rd, the Federal Circuit affirmed under Rule 36 the decision by the District Court of Nevada (Du, J.) in March that the claims asserted by Amarin Pharma against West-Ward Pharmaceuticals International Ltd., Hikma Pharmaceuticals USA Inc., Dr. Reddy's Laboratories, Inc., and Dr. Reddy's Laboratories, Ltd. were invalid for obviousness, in Amarin Pharma, Inc. v. Hikma Pharmaceuticals USA Inc.  Rule 36 decisions are rarely further discussed; this case (while not particularly remarkable) does provide an opportunity to understand the factual and legal circumstances surrounding this increasingly frequent outcome.

The case arose in ANDA litigation over Amarin's Vascepa® drug, which is a highly purified preparation of eicosapentaenoic acid (EPA).  Specifically the formulation used in the patented methods is specified as comprising at least 96% EPA by weight and (relevant to these claims) further comprises substantially no docosahexaenoic acid (DHA) or related esters.  The drug is used to treat hypertriglyceridemia (hyperTG), a condition where blood concentrations of triglycerides (TG) can be as high as 500-1500 mg/dL (normal blood levels are less than 150 mg/dL TG).  Prior art drugs, specifically Lovaza comprised a mixture of EPA and DHA, but suffered from the side effect that patients developed increased amounts of low-density lipoprotein-associated cholesterol (LDL-C) in their blood, increasing the risk of heart attacks.  As a consequence, Lovaza was administered with a statin to counteract this negative side effect.  Amarin's Vascepa® drug did not have this side effect and did not require concomitant statin administration.

Amarin asserted U.S. Patent Nos. 8,293,728; 8,318,715; 8,357,677; 8,367,652; 8,431,560; and 8,518,929; claims 1 and 16 of the '728 patent are representative:

1.  A method of reducing triglycerides in a subject having a fasting baseline triglyceride level of 500 mg/dl to about 1500 mg/dl who does not receive concurrent lipid altering therapy comprising: administering orally to the subject about 4 g per day of a pharmaceutical composition comprising at least about 96% by weight of all fatty acids present, ethyl eicosapentaenoate, and substantially no docosahexaenoic acid or its esters for a period of 12 weeks to effect a reduction in triglycerides without substantially increasing LDL-C compared to a second subject having a fasting baseline triglyceride level of 500 mg/dl to about 1500 mg/dl who has not received the pharmaceutical composition and a concurrent lipid altering therapy.

16.  The method of claim 1, wherein no fatty acid of the pharmaceutical composition, except for ethyl-EPA, comprises more than about 0.6% by weight of all fatty acids combined.

The District Court held that Defendants' label would induce infringement of all asserted claims of all Amarin's patents, a not uncommon outcome in ANDA litigation in view of FDA requirements for a generic drug to conform to the specifications of the corresponding branded version.  However, infringement liability of any kind was precluded by the court's further finding that all of the claims of all of Amarin's asserted patents were invalid for obviousness.

Defendants asserted the following prior art references to support their obviousness arguments:

1.  Lovaza as cited in the Physicians' Desk Reference (PDR), that disclosed the combination of EPA + DHA for hyperTG, administered at same dose as Amarin's patents prescribe; Lovaza was reported to have same effects as Vascepa® on LDL-C when used with a statin but had raised LDL-C when administered by itself

2.  Mori et al., Purified Eicosapentaenoic and Docosahexaenoic Acids Have Differential Effects on Serum Lipids and Lipoproteins, LDL Particle Size, Glucose, and Insulin in Mildly Hyperlipidemic Men, 71 Am. J. Clinical Nutrition 1085- 94 (2000), which disclosed that administration of EPA alone reduced TG levels by 18.4% and DHA alone by 20%, stating these were essentially the same effects; also disclosed that serum LDL-C increased by 8% ("significantly") with DHA administration but by only 3.5% with EPA.

3.  Hayashi et al., Decreases in Plasma Lipid Content and Thrombotic Activity by Ethyl Icosapentate Purified from Fish Oils, 56(1) Curr. Therap. Res. 24-31 (1995) disclosed that "at least one patient with hyperTG that improved with EPA treatment"; no disclosure of effects on LDL-C except that EPA did not increase LDL-C.

4.  Kurabayashi et al., Eicosapentaenoic Acid Effect on Hyperlipidemia in Menopausal Japanese Women. Obstet. Gynecol. 96:521-8 (2000), which disclosed that EPA administration at ½ dosage with estrogen (which was known to increase TG); reported 27% reduction in serum TG levels.

The Lovaza reference was the principle reference relied upon by the District Court in making its obviousness determination, for teaching that this formulation was effective at reducing TG levels in patients with hyperTG, but with the risk of pathological increases in LDL-C levels.  The Court considered the knowledge in the art regarding Lovaza in combination with the Mori reference for its teachings that the LDL-C raising effect was due to DHA in the Lovaza formulation.  The other cited art was relied upon for teaching that EPA alone could reduce TG levels, albeit on hyperTG patients as well as patients with elevated TG but not above 500 mg/dL.  Based on this art, the District Court found the person of skill in the art would have been motivated to combine the teachings of the Lovaza reference and the Mori reference because there would be a benefit in being able to administer one drug (EPA) rather than Lovaza plus a statin (for, inter alia, improved patient compliance reasons).  The District Court also perceived Plaintiff's expert as conceding this point, and that the prior art taken as a whole showed the desired effect known in the art to be associated with EPA administration alone.  The Court found that the skilled worker would have had a reasonable expectation of success in this combination, based on expert testimony and the District Court's understanding of what was disclosed in the prior art.

The District Court then turned to the objective indicia of non-obviousness (termed the "secondary considerations in Graham v. John Deere & Co.).  In rendering its decision, the District Court characterized the procedural status of her deliberations by saying that Defendants had established a prima facie case of obviousness and she would then consider whether Plaintiffs had rebutted that case with evidence of secondary considerations; this analytical framework would be one of Plaintiff's arguments for errors of law on appeal.  Plaintiffs asserted the secondary considerations of unexpected benefits, satisfaction of long-felt but unmet need, skepticism, praise, and commercial success; after considering Plaintiff's evidence the Court found that certain of these secondary considerations (satisfaction of long-felt need, commercial success) provided evidence contrary to an obviousness determination but on balance Plaintiff's evidence of secondary considerations did not rebut the prima facie case.  (In an interesting twist recognized by the District Court, these patents were granted after the Examiner deemed the secondary considerations rebutted the Examiner's prima facie case.)

On appeal, Amarin argued that the District Court erred by establishing the prima facie case before considering the secondary considerations, which Amarin argued was improper, citing In re Cyclobenzaprine, 676 F.3d 1063 (Fed. Cir. 2012).  Amarin also accused the District Court of engaging in hindsight bias, and for shifting the burden from Defendants in how it performed its obviousness analysis (again, but establishing the prima facie case and then placing the burden on Amarin to rebut it).  For their part, Defendants countered Amarin's Cyclobenzaprine citation with the Federal Circuit's later-decided Novo Nordisk A/S v. Caraco Pharm. Labs., Ltd., 719 F.3d 1346 (Fed. Cir. 2013) case, and further opposed Amarin's other grounds of appeal.

The Federal Circuit heard oral argument on September 2nd, before a panel of Judges Dyk, Reyna, and Hughes (a recording of the oral argument can be obtained here).  Judge Dyk raised with Amarin's counsel the effect of the Novo decision regarding the prima facie case approach, stating by question that the panel was bound by earlier cases and Novo had affirmed the legal correctness of the District Court's approach.  Amarin's counsel argued that the District Court further erred by balancing the various species of secondary indicia but Judge Dyk appeared to believe that the one sentence Amarin pointed to in the District Court's opinion was insufficient evidence of analytical impropriety.  Judge Dyk asserted that merely weighing the secondary considerations versus the prima facie case seemed to satisfy the law in his view.  Judge Reyna's less extensive questioning focused on Amarin's argument that the District Court reached its conclusion before looking to secondary considerations, asking where in the record that would be.  Judge Hughes did not pose any questions to Amarin's counsel, and none of the judges questioned Defendants' counsel.  The Court's Rule 36 affirmance followed the next day.

It is clear that the Court was not convinced the District Court had committed any legal error in how it performed its obviousness analysis and that Amarin did not show clear error in any of the District Court's factual findings.  Thus, the Rule 36 affirmance followed.

Amarin Pharma, Inc. v. Hikma Pharmaceuticals USA Inc. (Fed. Cir. 2020)
Nonprecedential disposition
Panel: Circuit Judges Dyk, Reyna, and Hughes
Per curiam