Amarin Corp Plc (AMRN)

[marjac]

Here is this judicial abortion. Her legal "reasoning" as to why Amarin loses on obviousness, starts on page 54 of 70

UNITED STATES DISTRICT COURT DISTRICT OF NEVADA
* * *
Case No. 2:16-cv-02525-MMD-NJK BENCH ORDER


13 This is a consolidated patent infringement case brought under the Hatch-Waxman
14 Act where Plaintiffs Amarin Pharma, Inc. and Amarin Pharmaceuticals Ireland Limited
15 (collectively, “Amarin”) seek to prevent Defendants West-Ward Pharmaceuticals
16 International Limited and Hikma Pharmaceuticals USA Inc. (collectively, “Hikma”), and Dr.
17 Reddy’s Laboratories, Inc. and Dr. Reddy’s Laboratories, Ltd. (collectively, “DRL”) from
18 launching generic competitor drugs to Plaintiffs’ drug Vascepa®. This order follows a
19 bench trial the Court held in January 2020 (the “Trial”). As further explained below in the
20 Court’s findings of fact and conclusions of law, the Court finds that Defendants infringe the
21 asserted claims under Plaintiffs’ inducement theory, but the asserted patent claims are all
22 invalid as obvious.
23 II. CLAIMS
24 Plaintiffs sued Defendants under the patent laws of the United States, 35 U.S.C. §
25 100, et seq., including 35 U.S.C. § 271(e)(2), and the Declaratory Judgment Act, 28 U.S.C.
26 §§ 2201 and 2202, arising from Defendants’ filing of Abbreviated New Drug Applications
27 (“ANDAs”) under Section 505(j) of the Federal Food, Drug, and Cosmetic Act (“FDCA”),

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 2 of 70



1 21 U.S.C. § 355(j), seeking approval from the United States Food and Drug Administration
2 (“FDA”) to market generic versions of Plaintiffs’ Vascepa product. (ECF No. 324 at 2.)
3 Plaintiffs specifically assert infringement of U.S. Patent No. 8,293,728 (“the ’728 4 patent”), U.S. Patent No. 8,318,715 (“the ’715 patent”), U.S. Patent No. 8,357,677 (“the 5 ’677 patent”), U.S. Patent No. 8,367,652 (“the ’652 patent”), U.S. Patent No. 8,431,560
6 (“the ’560 patent”), and U.S. Patent No. 8,518,929 (“the ’929 patent”).1 (ECF No. 331 at
7 9.) Each of the Asserted Patents is entitled “METHODS OF TREATING
8 HYPERTRIGLYCERIDEMIA.” (Id.) The U.S. applications that ultimately issued as the
9 Asserted Patents are continuations of U.S. Application No. 12/702,889, filed on February
10 9, 2010, which ultimately issued as the U.S. Patent No. 8,293,727 (“the ’727 patent”). (Id.)
11 The Asserted Patents further claim priority to U.S. Provisional Application No. 61/151,291,
12 filed on February 10, 2009, and U.S. Provisional Application No. 61/173,755, filed on April 13 29, 2009. (Id.)
14 Plaintiffs more specifically assert that Defendants infringe the following ten claims
15 of the Asserted Patents: Claims 1 and 16 of the ’728 patent, Claim 14 of the ’715 patent,
16 Claims 1 and 8 of the ’677 patent, Claim 1 of the ’652 patent, Claims 4 and 17 of the ’560
17 patent, and Claims 1 and 5 of the ’929 patent.2 (ECF Nos. 331 at 9-10, 333 at 13 n.1.)
18 Defendants asserted counterclaims of noninfringement and invalidity. (ECF Nos. 27 at 28- 19 34, 33 at 33-56.)
20 III. FINDINGS OF FACT
21 The Court makes the following findings of fact based on the testimony and other
22 evidence admitted during the course of the Trial, along with the pre-trial and post-trial
23 briefing the parties filed in this case. 24 ///
25 ///
26
27 1Collectively, the “Asserted Patents.”
28 2Collectively, the “Asserted Claims.”

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 3 of 70



1 A. Factual Background
2 The Asserted Patents are directed to methods of beneficially lowering the levels of
3 certain fats in the bloodstream using drugs made of purified omega-3 fatty acids from fish
4 oil. Fats are natural biological molecules that scientists call “lipids.” Triglycerides (“TGs”)
5 and cholesterol are two types of lipids that are of major importance in human physiology.
6 TGs are high in calories and are a major source of energy in the diet of humans. (ECF No.
7 370 at 1561:21-1562:21.) After they are absorbed from the intestine, triglycerides are
8 broken down into their component molecules, resynthesized, and reassembled by the
9 intestine into lipoproteins. Lipoproteins are spherical particles that travel through the
10 bloodstream and contain lipids (such as triglycerides and cholesterol) as well as proteins.
11 (ECF Nos. 366 at 324:5-9, 370 at 1562:12-17.) The major proteins that are in lipoproteins
12 are called apolipoproteins. One type of apolipoprotein is Apo B.
13 Cholesterol levels measured in a clinical laboratory generally include levels of both
14 free cholesterol and cholesteryl ester. (ECF No. 333 at 8.) The level of cholesterol
15 measured in the blood is generally an indicator for the amount of low-density lipoprotein
16 cholesterol (“LDL-C”) in the blood. (Id.) LDL-C is the “bad” cholesterol that physicians try
17 to reduce in their patients with drugs such as statins. (Id.) In many patients, there is a
18 strong linear relationship between levels of LDL-C and Apo B. (Id.) In other words,
19 changes in LDL-C levels occur in parallel with changes in Apo B, reflecting the fact that
20 there is one molecule of Apo B per LDL particle. (Id.)
21 The Asserted Claims are directed to methods of treating severe
22 hypertriglyceridemia, a condition in which a patient’s fasting TG levels rise to very high
23 levels of 500 mg/dL or above. (ECF No. 377 at 33.) The term “hypertriglyceridemia”
24 (“HTG”) refers to having elevated TGs, which are the most abundant type of fat in the
25 blood. (ECF No. 373 at 27.) The clinical guidelines that both sides rely on in this case,
26 called “ATP III,” define “normal triglycerides” as less than 150 mg/dL, with levels above 27
28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 4 of 70



1 that considered elevated to various degrees. (Ex. 15263 (National Institutes of Health,
2 National Heart, Lung, and Blood Institute, “Detection, Evaluation, and Treatment of High
3 Blood Cholesterol in Adults (Adult Treatment Panel III), Executive Summary,” May 2001
4 (“ATP-III Executive Summary”)) at 27.) These numbers are referring to the “concentrations
5 of triglycerides in the blood, and [] are always taken in the fasting state.” (ECF No. 366 at 6 329:4-17.)
7 Severe hypertriglyceridemia “has a well-known meaning to doctors who treat the
8 condition.” (Id. at 454:6-8.) It “means that a patient has had triglycerides levels greater
9 than or equal to 500 milligrams per deciliter.” (ECF No. 365 at 52:24-3; see also ECF No.
10 366 at 454:9-12.) In other words, “as long as the patients have [TG] levels above 500,
11 regardless of why, they have severe hypertriglyceridemia.” (ECF No. 366 at 455:8-11.)
12 This definition is consistent with the ATP-III guidelines as well as the Vascepa indication. 13 (Ex. 1526 at 27; Ex. 2248 at 1.)
14 For most patients with elevated TGs, “the primary aim of therapy is to achieve the
15 target goal for LDL[-C levels].” (Ex. 1526 at 27.) This is because research has long shown
16 that “elevated LDL cholesterol is a major cause of CHD”—i.e., coronary heart disease. (Id. 17 at 11.)
18 The primary aim of therapy is different in patients with severe HTG because they
19 have an elevated risk of acute pancreatitis. Pancreatitis, which involves the inflammation
20 of the pancreas, is an excruciatingly painful and potentially life-threatening condition. (ECF
21 No. 370 at 1567:2-22 (“In the setting of severe hypertriglyceridemia, inflammatory changes
22 [c]an occur within the pancreas that can lead to sudden devastating injury to the pancreas
23 leading to dissolution of pancreatic tissue, resulting in severe pain, inability to eat, to drink,
24 and it constitutes a medical emergency. But even more importantly[,] in some cases[,] it
25 [can] even result in death.”); see also ECF Nos. 366 at 331:3-20, 365 at 72:4-13.) In 26
27 3The designation “Ex.” refers to exhibits published by the parties during Trial and admitted by the Court. They are not filed on the docket but are available for public review
28 in the Clerk of Court’s office at 400 S. Virginia St. in Reno, Nevada, upon request, by referencing the case number of this case.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 5 of 70



1 patients with severe hypertriglyceridemia, the primary “aim of therapy is to prevent acute
2 pancreatitis through triglyceride lowering.” (ECF No. 366 at 457:11-15; see also Ex. 1526
3 at 19.) This is the “primary treatment aim [in patients with severe hypertriglyceridemia]
4 regardless of why the patient has triglycerides above 500.” (ECF No. 366 at 457:16-18.)
5 This is because “pancreatitis can be a life-threatening condition.” (Id. at 473:18-20; see
6 also id. at 568:10-16.)
7 As noted, the Asserted Claims are directed to methods of treating severe HTG
8 specifically by administering 4 grams (“4 g”) per day of purified EPA. Treating patients with
9 severe hypertriglyceridemia with purified EPA reduced TGs in those patients without
10 increasing LDL-C, the bad-cholesterol. (ECF Nos. 367 at 851:15-852:1, 370 at 1574:3-
11 1575:1, 1598:14-17.) Other treatments for severe hypertriglyceridemia dramatically
12 increase LDL-C levels, which then often requires the administration of a separate
13 concurrent cholesterol-lowering drug, such as a statin, just to address that LDL-C 14 increase. (ECF Nos. 367 at 813:8-814:2, 370 at 1598:18-1599:18.) Additionally, purified
15 EPA has now been shown to actually reduce cardiovascular risk in severely
16 hypertriglyceridemic patients on top of a statin, the only TG-lowering treatment shown to 17 confer such a benefit. (ECF Nos. 367 at 849:21-24, 368 at 1122:6-13, 370 at 1622:5-16,
18 1625:2-21.) Treating severe HTG with purified EPA therefore offers several benefits over
19 other possible treatments.
20 B. Plaintiff’s Drug
21 Vascepa is a highly purified preparation of EPA (eicosapentaenoic acid), also
22 known as icosapent ethyl. (ECF No. 324 at 24.) FDA first approved Vascepa in July 2012
23 as “an adjunct to diet to reduce triglyceride (“TG”) levels in adult patients with severe (≥
24 500 mg/dL) hypertriglyceridemia.” (Id.) Amarin currently markets Vascepa in both 1 g and
25 500 mg capsules. (Ex. 1186 at 2.) The daily dose of Vascepa is 4 grams per day, taken
26 as two 1-gram (or four 500 mg) capsules twice daily with food. (ECF No. 324 at 24.)
27 Vascepa embodies the Asserted Claims. Vascepa contains a “pharmaceutical
28 composition,” as required by Claims 1 and 16 of the ’728 patent, Claim 14 of the ’715

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 6 of 70



1 patent, Claims 1 and 8 of the ’677 patent, Claim 1 of the ’652 patent, and Claims 1 and 5
2 of the ’929 patent. The “pharmaceutical composition” in Vascepa comprises “at least about
3 96%, by weight of all fatty acids present, ethyl eicosapentaenoate[,] and substantially no
4 docosahexaenoic acid or its esters,” as required by Claims 1 and 16 of the ’728 patent,
5 Claims 1 and 8 of the ’677 patent, and Claims 1 and 8 of the ’652 patent. Vascepa further
6 contains a “pharmaceutical composition” “wherein no fatty acid of the pharmaceutical
7 composition, except for ethyl-EPA, comprises more than about 0.6% by weight of all fatty
8 acids combined,” as required by Claim 16 of the ’728 patent. (Id. at 25.) The
9 “pharmaceutical composition” in Vascepa also comprises “at least about 96% by weight,
10 ethyl eicosapentaenoate (ethyl-EPA) and substantially no docosahexaenoic acid
11 ([“]DHA[”]) or its esters,” as required by Claim 14 of the ’715 patent. (Id.) Vascepa
12 comprises a “capsule comprising about 900 mg to about 1 g of ethyl eicosapentaenoate
13 and not more than about 3% docosahexaenoic acid or its esters, by weight of total fatty
14 acids present,” as required by Claims 4 and 17 of the ’560 patent. (Id.) Finally, the
15 “pharmaceutical composition” in a daily dose of Vascepa comprises “about 4 g of ethyl
16 eicosapentaenoate and not more than about 4% docosahexaenoic acid or its esters, by
17 weight of all fatty acids,” as required by Claims 1 and 5 of the ’929 patent. (Id.)
18 C. Defendants’ ANDA Applications and Products
19 In 2016, after Vascepa’s initial period of exclusivity against generic competition
20 expired, Defendants filed ANDAs seeking FDA approval to market generic versions of
21 Vascepa. As required by law, Defendants’ ANDAs adopted the “same” labelling as
22 Vascepa, which at the time was only approved for severe hypertriglyceridemia. See 21
23 U.S.C. §§ 355(j)(2)(A)(v), (j)(4)(G). However, Plaintiffs have since won FDA approval of a
24 second indication for Vascepa—reducing the risk of adverse cardiovascular events. Now
25 that Vascepa has two indications, the law “permits [Defendants] to file ANDAs directed to
26 a subset of FDA-approved indications and even provides a mechanism for [Defendants]
27 to affirmatively carve out” the new indication from their labels. AstraZeneca Pharm. LP v.
28 Apotex Corp., 669 F.3d 1370, 1381 (Fed. Cir. 2012). Thus, Defendants’ current labels do

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 7 of 70



1 not include Vascepa’s new indication, and are materially the same as the labels the Court
2 previously considered in ruling on the parties’ summary judgment motions.
3 1. Hikma’s ANDA
4 On or about July 26, 2016, Hikma Pharmaceuticals PLC and Roxane Laboratories,
5 Inc., through Roxane Laboratories, Inc. (incorporated in Nevada), submitted to FDA an
6 ANDA (ANDA No. 209457) with paragraph IV certifications under Section
7 505(j)(2)(A)(vii)(IV) of the FDCA, 21 U.S.C. § 355(j)(2)(A)(vii)(IV), seeking approval to
8 market a generic version of Vascepa® (icosapent ethyl) 1 g capsules as Icosapent Ethyl
9 Capsules, 1 gram (“Hikma’s ANDA Product”). (ECF No. 24 at 22.)
10 Pursuant to 21 U.S.C. § 355(j)(2)(B), in a letter dated September 21, 2016, Hikma
11 Pharmaceuticals PLC and Roxane Laboratories, Inc. notified Amarin that they had
12 submitted to FDA ANDA No. 209457, with paragraph IV certifications for the Asserted
13 Patents. (Id.)
14 On or about December 8, 2016, Roxane Laboratories, Inc. transferred ANDA No.
15 209457 to West-Ward Pharmaceuticals International Limited. (Id.)
16 On or about December 8, 2016, West-Ward Pharmaceuticals International Limited
17 appointed West-Ward Pharmaceuticals Corp. as its agent for purposes of communication
18 with FDA regarding ANDA No. 209457. (Id. at 23.)
19 West-Ward Pharmaceuticals International Limited has changed its name to Hikma
20 Pharmaceuticals International Limited. (Id.)
21 On or about July 8, 2019, Hikma Pharmaceuticals International Limited transferred
22 ANDA No. 209457 to Hikma Pharmaceuticals USA Inc. Hikma Pharmaceuticals USA Inc.
23 is now the owner of ANDA No. 209457. (Id.)
24 Vascepa is the Reference Listed Drug (“RLD”) for ANDA No. 209457. (ECF No.
25 324 at 25.) Hikma’s ANDA Product, if approved, will be bioequivalent to Vascepa. (Id.)
26 The indication set forth in the proposed labeling for Hikma’s ANDA Product, submitted in
27 connection with ANDA No. 209457, is “as an adjunct to diet to reduce triglyceride (TG)
28 levels in adult patients with severe (≥500 mg/dL) hypertriglyceridemia.” (Id. at 26.) The

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 8 of 70



1 dosage form of Hikma’s ANDA Product, if approved, will be a 1- gram soft-gelatin capsule.
2 (Id.) The daily dose of Hikma’s ANDA Product, if approved, will be 4 grams per day taken
3 as two 1-gram capsules twice daily with food. (Id.) Hikma’s ANDA Product, if approved,
4 will contain icosapent ethyl. (Id.)
5 Hikma’s ANDA Product, if approved, will contain a “pharmaceutical composition,”
6 as required by Claims 1 and 16 of the ’728 patent, Claim 14 of the ’715 patent, Claims 1
7 and 8 of the ’677 patent, Claim 1 of the ’652 patent, and Claims 1 and 5 of the ’929 patent.
8 (Id.) The “pharmaceutical composition” in Hikma’s ANDA Product, if approved, will
9 comprise “at least about 96%, by weight of all fatty acids present, ethyl
10 eicosapentaenoate[,] and substantially no docosahexaenoic acid or its esters,” as required
11 by Claims 1 and 16 of the ’728 patent, Claims 1 and 8 of the ’677 patent, and Claim 1 of
12 the ’652 patent. (Id.) Hikma’s ANDA Product, if approved, will contain a “pharmaceutical
13 composition” “wherein no fatty acid of the pharmaceutical composition, except for ethyl-
14 EPA, comprises more than about 0.6% by weight of all fatty acids combined,” as required
15 by Claim 16 of the ’728 patent. (Id.) Hikma’s ANDA Product, if approved, will comprise a
16 “capsule comprising about 900 mg to about 1 g of ethyl eicosapentaenoate and not more
17 than about 3% docosahexaenoic acid or its esters, by weight of total fatty acids present,”
18 as required by Claims 4 and 17 of the ’560 patent. (Id. at 26-27.) The “pharmaceutical
19 composition” in a daily dose of Hikma’s ANDA Product, if approved, will comprise “about
20 4 g of ethyl eicosapentaenoate and not more than about 4% docosahexaenoic acid or its
21 esters, by weight of all fatty acids,” as required by Claims 1 and 5 of the ’929 patent. (Id. 22 at 27.)
23 2. DRL’s ANDA
24 On or about July 26, 2016, DRL, through Dr. Reddy’s Laboratories, Inc., submitted
25 to FDA an ANDA (ANDA No. 209499) with paragraph IV certifications under Section
26 505(j)(2)(A)(vii)(IV) of the FDCA, 21 U.S.C. § 355(j)(2)(A)(vii)(IV), seeking approval to
27 market a generic version of Vascepa (icosapent ethyl) 1 g capsules as Icosapent Ethyl
28 Capsules, 1 gram (“DRL’s ANDA Product”). (Id. at 23)

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 9 of 70



1 Pursuant to 21 U.S.C. § 355(j)(2)(B), in a letter dated September 22, 2016, DRL
2 notified Amarin that it had submitted to FDA ANDA No. 209499, with paragraph IV
3 certifications for the Asserted Patents. (Id. at 24.)
4 On or about July 11, 2018, DRL, through Dr. Reddy’s Laboratories, Inc., submitted
5 to FDA a supplement to ANDA No. 209499 with paragraph IV certifications under Section
6 505(j)(2)(A)(vii)(IV) of the FDCA, 21 U.S.C. § 355(j)(2)(A)(vii)(IV), for 500 mg icosapent
7 ethyl capsules purportedly bioequivalent to Vascepa. (Id.)
8 Pursuant to 21 U.S.C. § 355(j)(2)(B), in a letter dated July 11, 2018, DRL notified
9 Amarin that it had submitted to FDA a supplement to ANDA No. 20499, with paragraph IV
10 certifications for the ’728, ’715, ’677, ’652, and ’929 patents. (Id. at 24.)
11 Vascepa is the RLD for ANDA No. 209499. DRL’s ANDA Product, if approved, will
12 be bioequivalent to Vascepa. (Id. at 27.) The indication set forth in the proposed labeling
13 for DRL’s ANDA Product, submitted in connection with ANDA No. 209499, is “as an
14 adjunct to diet to reduce triglyceride (TG) levels in adult patients with severe (≥ 500 mg/dL)
15 hypertriglyceridemia.” (Id.) The dosage form of DRL’s ANDA Product, if approved, will be
16 a 1-gram soft-gelatin capsule. (Id.) The daily dose of DRL’s ANDA Product, if approved,
17 will be 4 grams per day taken as two 1-gram capsules twice daily with food. DRL’s ANDA
18 Product, if approved, will contain icosapent ethyl. (Id.)
19 DRL’s ANDA Product, if approved, will contain a “pharmaceutical composition,” as
20 required by Claims 1 and 16 of the ’728 patent, Claim 14 of the ’715 patent, Claims 1 and
21 8 of the ’677 patent, Claim 1 of the ’652 patent, and Claims 1 and 5 of the ’929 patent.
22 (Id.) The “pharmaceutical composition” in DRL’s ANDA Product, if approved, will comprise
23 “at least about 96%, by weight of all fatty acids present, ethyl eicosapentaenoate[,] and
24 substantially no docosahexaenoic acid or its esters,” as required by Claims 1 and 16 of
25 the ’728 patent, Claims 1 and 8 of the ’677 patent, and Claim 1 of the ’652 patent. (Id.)
26 DRL’s ANDA Product, if approved, will contain a “pharmaceutical composition”
27 “wherein no fatty acid of the pharmaceutical composition, except for ethyl-EPA, comprises
28 more than about 0.6% by weight of all fatty acids combined,” as required by Claim 16 of

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 10 of 70



1 the ’728 patent. (Id. at 27-28.) The “pharmaceutical composition” in DRL’s ANDA Product,
2 if approved, will comprise “at least about 96% by weight, ethyl eicosapentaenoate (ethyl-
3 EPA) and substantially no docosahexaenoic acid (DHA) or its esters,” as required by Claim
4 14 of the ’715 patent. (Id. at 28.) DRL’s ANDA Product, if approved, will comprise a capsule
5 comprising 950 mg to 1050 mg of ethyl eicosapentaenoate. DRL did not assert the claim
6 limitation from Claims 4 and 17 of the ’560 patent that recites a “capsule comprising about
7 900 mg to about 1 g of ethyl eicosapentaenoate” as a basis for noninfringement of Claims
8 4 and 17 of the ’560 patent. (Id.) DRL’s ANDA Product, if approved, will comprise “a
9 capsule comprising . . . not more than about 3% docosahexaenoic acid or its esters, by
10 weight of total fatty acids present,” as required by Claims 4 and 17 of the ’560 patent. (Id.)
11 The “pharmaceutical composition” in a daily dose of DRL’s ANDA Product, if approved,
12 will comprise “about 4 g of ethyl eicosapentaenoate and not more than about 4%
13 docosahexaenoic acid or its esters, by weight of all fatty acids,” as required by Claims 1
14 and 5 of the ’929 patent. (Id.)
15 D. The Asserted Patents
16 1. The ’728 Patent
17 The ’728 patent issued on October 23, 2012 to Mehar Manku, Ian Osterloh, Pierre
18 Wicker, Rene Braeekman, and Paresh Soni (collectively, “Inventors”). The patent issued
19 from Application No. 13/349,153 (“the ’153 application”). (ECF No. 324 at 4.)
20 Claims 1 and 16 of the ’728 patent are asserted. The asserted claims of the ’728
21 patent, and any claims from which they depend, are reproduced below.
22 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
23 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
24 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
25 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
26 has not received the pharmaceutical composition and a concurrent lipid altering therapy.
27
28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 11 of 70



1 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
2 all fatty acids combined.
3 2. The ’715 Patent
4 The ’715 patent issued on November 27, 2012 to the Inventors. The patent issued
5 from Application No. 13/282,145 (“the ’145 application”). (ECF No. 324 at 4.) Claim 14 of
6 the ’715 patent is asserted. The asserted claims of the ’715 patent, and any claims from
7 which they depend, are reproduced below.
8 13. 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
9 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, ethyl
10 eicosapentaenoate (ethyl-EPA) and substantially no docosahexaenoic acid (DHA) or its esters for a period of at least 12 weeks to effect a statistically significant
11 reduction in triglycerides without effecting a statistically significant increase in LDLC or apolipoprotein B in the subject.
12
14. The method of claim 13 comprising administering to the subject about 4 g per
13 day of the pharmaceutical composition to effect a statistically significant reduction in triglycerides and apolipoprotein B without effecting a statistically significant
14 increase of LDL-C in the subject.
15 3. The ’677 Patent
16 The ’677 patent issued on January 22, 2013, to the Inventors. The patent issued
17 from Application No. 13/608,775 (“the ’775 application”). (ECF No. 324 at 4.) Claims 1 and
18 8 of the ’677 patent are asserted. The asserted claims of the ’677 patent, and any claims
19 from which they depend, are reproduced below.
20 1. A method of reducing triglycerides in a subject having a fasting baseline triglyceride level of 500 mg/dl to about 1500 mg/dl comprising: administering orally
21 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
22 substantially no docosahexaenoic acid or its esters for a period of at least about 12 weeks to effect a reduction in triglycerides without substantially increasing LDL-C
23 compared to placebo control.
24 8. The method of claim 1, comprising administering to the subject about 4 g of the pharmaceutical composition daily for the period of at least about 12 weeks to effect
25 a reduction in apolipoprotein B compared to placebo control.
26
27
28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 12 of 70



1 4. The ’652 Patent
2 The ’652 patent issued on February 5, 2013 to the Inventors. The patent issued
3 from Application No. 13/610,247 (“the ’247 application”). (ECF No. 324 at 5.) Claim 1 of
4 the ’652 patent is asserted. The asserted claim of the ’652 patent is reproduced below.
5 1. A method of reducing triglycerides in a subject having a fasting baseline triglyceride level of 500 mg/dl to about 1500 mg/dl comprising: administering orally
6 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
7 substantially no docosahexaenoic acid or its esters for a period of about 12 weeks to effect a reduction in triglycerides without substantially increasing LDL-C
8 compared to baseline.
9 5. The ’560 Patent
10 The ’560 patent issued on October 23, 2012 to the Inventors. The patent issued
11 from Application No. 13/711,329 (“the ’329 application”). (ECF No. 324 at 5.) Claims 4 and
12 17 of the ’560 patent are asserted. The asserted claims of the ’560 patent, and any claims
13 from which they depend, are reproduced below.
14 1. A method of reducing triglycerides in a subject having a fasting baseline triglyceride level of 500 mg/dl to about 1500 mg/dl comprising, administering orally
15 to the subject 4 capsules per day, each capsule comprising about 900 mg to about 1 g of ethyl eicosapentaenoate and not more than about 3% docosahexaenoic acid
16 or its esters, by weight of total fatty acids present, for a period of 12 weeks to effect a reduction in triglycerides in the subject.
17
4. The method of claim 1, wherein said administering effects a reduction in fasting
18 triglycerides of at least about 10% without increasing the LDL-C by more than 5% in the subject.
19
11. A method of reducing triglycerides in a subject having a fasting baseline
20 triglyceride level of 500 mg/dl to about 1500 mg/dl comprising, administering orally to the subject 4 capsules per day, each capsule comprising about 900 mg to about
21 1 g of ethyl eicosapentaenoate and not more than about 3% docosahexaenoic acid or its esters, by weight of total fatty acids present, for a period of 12 weeks to effect
22 a reduction in triglycerides in the subject compared to placebo control.
23 17. The method of claim 11, wherein said administering effects reduction in fasting triglycerides of at least about 20% without increasing LDL-C in the subject
24 compared to placebo control.
25 6. The ’929 Patent
26 The ’929 patent issued on August 27, 2013 to the Inventors. The patent issued from
27 Application No. 13/776,242 (“the ’242 application”). (ECF No. 324 at 5.) Claims 1 and 5 of 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 13 of 70



1 the ’929 patent are asserted. The asserted claims of the ’929 patent, and any claims from
2 which they depend, are reproduced below.
3 1. A method of reducing triglycerides in a subject having fasting triglycerides of at least 500 mg/dl comprising, orally administering to the subject daily for at least
4 about 12 weeks a pharmaceutical composition comprising about 4 g ethyl eicosapentaenoate and not more than about 4% docosahexaenoic acid or its
5 esters, by weight of all fatty acids.
6 5. The method of claim 1, wherein 12 weeks of said daily administration is effective to reduce apolipoprotein B in subjects who have fasting triglycerides levels of at
7 least 500 mg/dl.
8 Pursuant to 21 U.S.C. § 355(b)(1), the Asserted Patents are listed in the Orange
9 Book—published by FDA and formally known as Approved Drug Products with
10 Therapeutic Equivalence Evaluations—in connection with New Drug Application (“NDA”)
11 No. 202057. (ECF No. 324 at 4.) Because the Asserted Patents are related, their
12 disclosures—the information contained within their respective specifications—are
13 essentially the same. (ECF No. 377 at 65.) All of the Asserted Patents were initially
14 rejected as obvious, but the patent examiner responsible for reviewing them later issued
15 materially identical statements of allowance permitting the Asserted Patents to issue
16 because he found that certain secondary considerations of nonobviousness made the
17 Asserted Claims patentable. (Id. at 61-65.) He specifically found the pending claims
18 patentable because “Applicant was able to overcome the above 103 obviousness rejection
19 by showing: 1 - Unexpected results, and 2 - Long felt unmet medical need.” (See, e.g., Ex. 20 38 at 1831.)
21 E. Witnesses
22 Both Plaintiffs and Defendants had witnesses, mostly experts, who testified at the
23 Trial. The parties also stipulated to the admission of the deposition testimony of other
24 expert witnesses, and the Court admitted that testimony. The Court briefly describes the
25 witnesses below.
26 1. Live Testimony
27 The following witnesses testified on Plaintiffs’ behalf during the Trial. Matthew
28 Budoff M.D. was admitted as an expert in the clinical treatment of patients with lipid

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 14 of 70



1 disorders, including severe hypertriglyceridemia, and as an expert in cardiology. (ECF No.
2 366 at 323:11-14.)4 Dr. Budoff’s testimony focused on the infringement portion of the case.
3 Plaintiffs also had a fact witness testify—Steven Ketchum, Ph.D. Dr. Ketchum is the
4 President of Research & Development, a Senior Vice President, and the Chief Scientific
5 Officer at Amarin Pharma, Inc. (ECF No. 365 at 49:18-19.) Dr. Ketchum’s testimony
6 focused on the history of Amarin and the development of Vascepa. Plaintiffs also offered
7 the expert testimony of Sean Nicholson, Ph.D. Dr. Nicholson was admitted as an expert
8 in the economics of the pharmaceutical industry. (ECF No. 369 at 1421:6-11.) He testified
9 about the commercial success of Vascepa and its nexus to the Asserted Claims. (Id. at
10 1417:13-1538:6.) Plaintiffs also offered Carl Peck M.D. as an expert in FDA regulation of
11 new and generic drugs including prescription drug labeling. (Id. at 1323:16-23.) In addition,
12 Peter Toth, M.D., Ph.D. was admitted as an expert in lipidology, the treatment of severe
13 hypertriglyceridemia, including severe hypertriglyceridemia, and the prevention and
14 treatment of cardiovascular disease. (ECF No. 370 at 1560:11-17.) Dr. Toth testified
15 regarding the non-obviousness of the Asserted Patents, and about the clinical attributes 16 of Vascepa. (Id. at 1546:9-1783:13.)
17 Defendants called expert witnesses Jonathan Sheinberg (non-infringement), Jay
18 Heinecke (invalidity), Edward Fisher (invalidity), and Ivan Hofmann (rebutting commercial
19 success). (ECF No. 373 at 19.) Dr. Sheinberg, a board-certified cardiologist, testified as
20 Defendants’ non-infringement expert. (Id. at 19-21.) Dr. Heinecke, an endocrinologist and
21 expert in lipoprotein metabolism and lipid disorders, testified as one of Defendants’
22 invalidity experts. (Id.) Dr. Fisher, a biochemist and expert in cardiovascular medicine, also
23 testified as one of Defendants’ invalidity experts. (Id.) Mr. Hofmann, an economist, testified
24 as Defendants’ commercial success expert. (Id.) 25 ///
26 ///
27
28 4References to the Trial transcripts (ECF Nos. 365-371) are to the transcript page numbers, not the page numbers of that particular document in the CM/ECF system.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 15 of 70



1 2. Deposition Testimony
2 As mentioned, the parties also stipulated to the admission of the following
3 deposition testimony.
4 Jerald Andry, Pharm.D. (Defendant Hikma’s Witness). Andry is the Senior Director
5 of Drug Regulatory Affairs and Medical Affairs at Hikma Pharmaceuticals USA Inc. (Andry 6 Dep. Tr. 8:15-23, 29:3-9.)5
7 Jaya Ayyagari (Defendant DRL’s Witness). Ayyagari is the Director of Regulatory
8 Affairs at Dr. Reddy’s Laboratories, Inc. (Ayyagari Dep. Tr. 5:9-21, 27:25-28:5.)
9 Harold E. Bays, M.D. (Third-Party Witness). Dr. Bays is the Medical Director and
10 President of Louisville Metabolic and Atherosclerosis Research Center. Dr. Bays
11 submitted two declarations to the Patent and Trademark Office during prosecution of the
12 Asserted Patents.
13 Andrea Cady, Ph.D. (Defendant Hikma’s Witness). Cady is the Senior Director of
14 Product Development at Hikma Pharmaceuticals USA Inc. (Cady Dep. Tr. 9:5-16.)
15 Philip Lavin, Ph.D. (Third-Party Witness). Dr. Lavin has a Ph.D. in Applied
16 Mathematics from Brown University. Dr. Lavin is self-employed through Lavin Consulting
17 LLC as a biostatistics consultant. Dr. Lavin submitted two declarations to the Patent and
18 Trademark Office during prosecution of the Asserted Patents.
19 Mehar Manku, Ph.D. (Third-Party Witness). Dr. Manku is one of the named
20 inventors of the Asserted Patents. While he no longer works there, throughout his career
21 at Amarin, Dr. Manku played a central role in the development of Vascepa. (Manku Dep. 22 Tr. 8:22-9:17, 10:5-12:11, 14:19-16:6, 31:10-32:12, 48:19-50:11.)
23 Peter R. Mathers (Defendants’ Expert). Mathers is a partner in the Washington,
24 D.C. law firm of Kleinfeld, Kaplan and Becker LLP, where he practices food and drug law. 25
26
27 5The designation “Dep. Tr.” refers to deposition transcripts admitted as evidence by the Court on the parties’ stipulation in lieu of reading them into the record at Trial. They
28 are also available for public review in the Clerk of Court’s office at 400 S. Virginia St. in Reno, Nevada.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 16 of 70



1 Mathers was retained by Defendants to provide opinions regarding issues relating to
2 patent infringement. (Mathers Dep. Tr. 11:13-24.)
3 Michael Miller, M.D. (Plaintiffs’ Claim Construction Declarant). Dr. Miller is
4 Professor of Cardiovascular Medicine, Epidemiology and Public Health at the University
5 of Maryland School of Medicine. Plaintiffs asked Dr. Miller to offer his expert opinion during
6 claim construction regarding how a person of ordinary skill in the art (“POSA”) would
7 understand certain terms in the Asserted Claims.
8 Ian Osterloh, M.D. (Third-Party Witness). Dr. Osterloh is one of the named
9 inventors of the Asserted Patents. In 2007, Dr. Osterloh joined Amarin as a consultant on
10 the severe hypertriglyceridemia clinical research and development program. (Osterloh 11 Dep. Tr. 8:22-9:18, 22:24-23:24, 49:1-3.)
12 Anuj Srivastava, Ph.D. (Defendant DRL’s Witness). At the time of his deposition,
13 Dr. Srivastava was the Senior Director of Strategic Portfolio & Business Development at
14 Dr. Reddy’s Laboratories, Inc. (Srivastava Dep. Tr. 6:5-8, 17:15-18:15.)
15 Howard S. Weintraub, M.D. (Third-Party Witness). Dr. Weintraub submitted two
16 declarations to the Patent and Trademark Office during prosecution of the Asserted 17 Patents. (Weintraub Dep. Tr. 8:19-9:7, 10:2-16, 114:20-115:19, 185:9-11.)
18 F. Infringement
19 In general, prescription drug labels are referred to alternatively as the label,
20 labeling, prescribing information, and/or package insert. (ECF No. 369 at 1324:13-18.) As
21 discussed below, the Court finds that the Vascepa label supports Plaintiffs’ view that
22 clinicians generally prescribe Vascepa for long-term use of at least 12 weeks.
23 The Indications and Usage section of the Vascepa label states that “Vascepa
24 (icosapent ethyl) is indicated as an adjunct to diet to reduce triglyceride (TG) levels in adult
25 patients with severe (≥ 500 mg/dL) hypertriglyceridemia.” (Ex. 1186 at 2.)6 The Indications 26
27 6The Indications and Usage section of Vascepa’s current labeling adds a second approved indication: “as an adjunct to maximally tolerated statin therapy to reduce the risk
28 of myocardial infarction, stroke, coronary revascularization, and unstable angina requiring hospitalization in adult patients with elevated triglyceride (TG) levels (≥ 150 mg/dL) and

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 17 of 70



1 and Usage section thus instructs clinicians that Vascepa is approved (i.e., safe and
2 effective) for use in combination with diet to reduce TGs in adult patients with severe
3 hypertriglyceridemia—without concurrent administration of any other medication. (ECF 4 No. 369 at 1352:12-20, 1375:16-19.)
5 The Indications and Usage section of the Vascepa label does not specify a duration
6 of use. (Ex. 1186 at 2.) The absence of a limitation on duration tells clinicians that FDA
7 has determined that there are no safety or efficacy concerns that require limiting the
8 duration of use of Vascepa. (ECF No. 369 at 1373:1-11.) Given the lack of any duration
9 of use combined with the indication to treat a chronic condition,7 the Indications and Usage
10 section instructs clinicians to prescribe VASCEPA long-term. (Id. at 1338:8-1339:6, 11 1373:19-1374:1.)
12 Prior to December 2019, Vascepa’s labeling also included a “Limitation of Use”
13 advising clinicians that Vascepa’s effect on cardiovascular mortality and morbidity in
14 patients with severe hypertriglyceridemia had not been determined. (See Ex. 940 at 2.)
15 That “Limitation of Use” was dropped when FDA approved Vascepa’s new indication for
16 cardiovascular risk-reduction.8 (See Ex. 1186 at 2.)
17 The Dosage and Administration section of the Vascepa label includes two sub-
18 headings. The first reads, “2.1 Prior to Initiation of Vascepa.” (Id.) Under this heading, the
19 label advises clinicians to “[a]ssess lipid levels before initiating therapy. Identify other
20 causes (e.g., diabetes mellitus, hypothyroidism, or medications) of high triglyceride levels
21 and manage as appropriate.” (Id.) This subheading also advises clinicians that “[p]atients
22 should engage in appropriate nutritional intake and physical activity before receiving
23 Vascepa, which should continue during treatment with Vascepa.” (Id.) 24
25 established cardiovascular disease or diabetes mellitus and 2 or more additional risk
26 factors for cardiovascular disease.” (Ex. 1186 at 2.) This indication, referred to during the Trial as the “REDUCE-IT Indication” is carved out of Defendants’ labels.
27 7In many cases, as discussed in more detail infra in the Court’s conclusions of law.
28 8Again, the “REDUCE-IT Indication.”

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 18 of 70



1 The second sub-heading is “2.2 Dosage and Administration.” Here, the label states
2 that “[t]he daily dose of Vascepa is 4 grams per day taken as either: four 0.5 gram capsules
3 twice daily with food; or as two 1 gram capsules twice daily with food.” (Id.) The label also
4 instructs clinicians to “[a]dvise patients to swallow Vascepa capsules whole. Do not break
5 open, crush, dissolve, or chew Vascepa.” (Id.; see also ECF No. 365 at 68:24-69:16.)
6 The Dosage and Administration section in Vascepa’s labeling does not specify a
7 duration of use. (Ex. 1186 at 2.) The absence of a duration limitation in this section
8 conveys that Vascepa’s benefit does not stop after a particular duration of treatment. (ECF
9 No. 369 at 1343:5-9.) This means that Vascepa was approved for long-term use to reduce
10 TGs and maintain that reduction. (Id. at 1344:3-14.)
11 The Dosage and Administration section in Vascepa’s labeling does not recommend
12 use of any concomitant medication. (Ex. 1186 at 2.) This conveys that FDA approved
13 Vascepa as a monotherapy to reduce TGs in adult patients with severe
14 hypertriglyceridemia (ECF No. 369 at 1355:7-10), and that FDA does not believe that the
15 safety or effectiveness of Vascepa depends on concurrent administration of another 16 medication (Id. at 1354:20-25, 365 at 67:7-12).
17 The Dosage Forms and Strength section of the VASCEPA label informs clinicians
18 that Vascepa is available as a 1-gram or 0.5-gram soft-gelatin capsule. (Ex. 1186 at 2; 19 see also ECF No. 365 at 67:13-68:6.)
20 The Contraindications section of the Vascepa label states that Vascepa is
21 contraindicated only in patients with known hypersensitivity to Vascepa or any of its
22 components. (Ex. 1186 at 2.)
23 The Warnings and Precautions section of a drug label is intended to describe serious
24 or otherwise clinically significant adverse reactions and safety hazards of which clinicians
25 need to be aware before prescribing the drug. (ECF No. 366 at 358:10-15.) See also 21
26 C.F.R. § 201.57(c)(6). The Warnings and Precautions section of the Vascepa label states
27 that Vascepa was associated with an increased risk of atrial fibrillation or atrial flutter and 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 19 of 70



1 an increased risk of bleeding. (Ex. 1186 at 2-3.) It also cautions against the use of Vascepa
2 in patients with known hypersensitivity to fish and/or shellfish. (Id.)
3 Unlike Lovaza’s9 labeling, the Warnings and Precautions section of the Vascepa
4 labeling does not warn of a potential increase in LDL-C levels. (ECF No. 366 at 407:7-25;
5 compare Ex. 566 at 1 with Ex. 1186 at 2-3.)
6 The Description section of the Vascepa label informs clinicians that the active
7 ingredient in Vascepa is “cosapent ethyl,” which “is an ethyl ester of the omega-3 fatty
8 acid eicosapentaenoic acid (EPA),” and that “[e]ach VASCEPA capsule contains . . . 1
9 gram of icosapent ethyl (in a 1 gram capsule).” (Ex. 1186 at 6; see also ECF No. 365 at
10 68:7-23.) This section also states that Vascepa is for “oral use.” (Ex. 1186 at 6; see also
11 ECF No. 366 at 418:2-5.)
12 The Nonclinical Toxicology section of a prescription drug label discloses the results
13 of studies conducted on rodents, or other non-human subjects. “It’s generally expected
14 that a carcinogenicity study be conducted in two rodent species to support marketing
15 approval of a new chemical entity for a chronic use indication.” (ECF No. 365 at 110:14-
16 17.) Plaintiffs performed two such studies, and their results are reflected in the Nonclinical
17 Toxicology section of the Vascepa label. (Id. at 111:11-20; see also Ex. 1186 at 8.) Both
18 rodent studies, the rat study described in the first paragraph and the mouse study
19 described in the second paragraph of the section, “supported there was no carcinogenic
20 potential of icosapent ethyl.” (ECF No. 365 at 112:11-7.)
21 The Clinical Studies section of the Vascepa label, sub-heading 14.2, describes the
22 design and results of the MARINE study, the primary study that established Vascepa’s
23 effectiveness at reducing triglycerides in adult patients with severe (≥ 500 mg/dL)
24 hypertriglyceridemia. (Ex. 1186 at 10-11.)10 25
26 9A competing, older drug whose guide for use is prior art to the Asserted Claims.
27 Lovaza is described in more detail infra in Section III.G.1(b).
28 10The 2019 label added to the Clinical Studies section the design and results of the REDUCE-IT study, under sub-heading 14.1. (Ex. 1186 at 8-10.) Like the rest of the

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 20 of 70



1 The Clinical Studies section, "14.2 Severe Hypertriglyceridemia," begins by
2 summarizing the major design characteristics of the MARINE study. Section 14.2 states:
3 The effects of Vascepa 4 grams per day were assessed in a randomized, placebo-controlled, double-blind, parallel-group
4 study of adult patients (76 on Vascepa, 75 on placebo) with severe hypertriglyceridemia. Patients whose baseline TG
5 levels were between 500 and 2,000 mg/dl were enrolled in this study for 12 weeks.. The median baseline TG and LDL-C
6 levels in these patients were 684 mg/dl and 86 mg/dl , respectively. Median baseline HDL-C level was 27 mg/dl . The
7 randomized population in this study was mostly Caucasian (88%) and male (76%). The mean age was 53 years and the
8 mean body mass index was 31 kg/mA2 _ Twenty-five percent of patients were on concomitant statin therapy, 28% were
9 diabetics, and 39% of the patients had TG levels >750 mg/dl.
10 (Ex. 1186 at 10-11.)
11 Next, Section 14.2 of the Clinical Studies Section includes a table summarizing the
12 "major lipoprotein lipid parameters for the groups receiving Vascepa or placebo" and
13 beneath the table is a brief summary of the conclusions. (Id. at 11, Tbl. 2.) 14
15
16
17
18
19
20 ,
21
22
23
24
25
26 Beneath Table 2, there is a paragraph highlighting key results of the MARINE trial. (Id. )
27

28 REDUCE-IT Indication, this portion of the Clinical Studies section is carved out of Defendants' labels.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 21 of 70



1 Amarin included the statements below Table 2 because it wanted to “apprise[]” “healthcare
2 professionals” and “draw the healthcare professional’s attention” to the “key information
3 from that pivotal trial.” (ECF No. 365 at 98:8-99:14.)
4 The Patient Counseling Information section of the Vascepa label instructs clinicians
5 to “[a]dvise the patient to read the FDA-approved patient labeling before starting Vascepa
6 (Patient Information),” and then lists five topics for discussion with patients: (1) the
7 potential increased risk for atrial fibrillation or atrial flutter; (2) the potential for allergic
8 reactions in patients with hypersensitivity to fish and/or shellfish; (3) the increased risk of
9 bleeding, particularly in patients receiving other antithrombotic agents; (4) the need to
10 swallow Vascepa capsules whole, and (5) and the need to take Vascepa as prescribed. 11 (See Ex. 1186 at 11-12.)
12 The Patient Information page at the end of the label is a handout that patients may
13 take with them. It reiterates much of the same information included in the label itself, but
14 in lay language. (ECF No. 366 at 359:11-24; see also Mathers Dep. Tr. 126:2-5, 7-20
15 (explaining how the Patient Information page distills information into user-friendly
16 language).)
17 Among other things, the Vascepa Patient Information sheet instructs patients to
18 “[t]ake Vascepa exactly as your doctor tells you to take it” and to “not change your dose
19 or stop taking Vascepa without talking to your doctor.” (Ex. 1186 at 13-14.) The Patient
20 Information sheet also instructs patients to “[t]ake VASCEPA capsules whole” and to “not
21 break, crush, dissolve, or chew VASCEPA capsules before swallowing.” (Id.) The Patient
22 Information sheet also advises that “your doctor may do blood tests to check your
23 triglyceride and other lipid levels while you take VASCEPA.” (Id.)
24 G. Obviousness
25 “Obviousness is a question of law based on underlying factual findings.” Power
26 Integrations, Inc. v. Fairchild Semiconductor Int’l, Inc., 711 F.3d 1348, 1355 (Fed. Cir.
27 2013). The Court now discusses below its factual findings relevant to the question of 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 22 of 70



1 whether the Asserted Claims are obvious in view of the combinations of prior art advanced
2 by Defendants.
3 1. Scope and Content of the Prior Art
4 The parties agree that the relevant prior art includes certain pieces of prior art. (ECF
5 No. 324 at 6-16.) “[T]he scope of the relevant prior art … includ[es] that reasonably
6 pertinent to the particular problem with which the inventor was involved A reference
7 is reasonably pertinent if, even though it may be in a different field of endeavor, it is one
8 which, because of the matter with which it deals, logically would have commended itself
9 to an inventor’s attention in considering his problem.” In re GPAC Inc., 57 F.3d 1573, 1577-
10 78 (Fed. Cir. 1995) (quotation omitted). Amongst those references that the parties agree
11 are prior art, the Court only discusses below the references that are relevant to its findings
12 of law.
13 a) Priority Date
14 Plaintiffs proposed a priority date for all Asserted Patents of March 2008, based on
15 emails sent by one of the Inventors (Manku (ECF Nos. 331 at 10, 377 at 174-76)) while
16 Defendants proposed a priority date of February 2009, the filing date of the patents (ECF
17 Nos. 333 at 55, 373 at 58-64). But the disputed priority date is not material, because
18 Defendants argue all Asserted Claims would have been obvious as of Plaintiffs’ alleged
19 conception date in March 2008. (ECF No. 373 at 167 n. 14.) Further, both sides’ experts
20 assessed obviousness as of March 2008, and made clear that their opinions would not
21 change if the priority date was February 2009. (ECF Nos. 367 at 827:8-10; 370 at 1638:5-
22 10.) Thus, the Court assumes without deciding that the Asserted Patents are entitled to a
23 priority date of March 2008, and its conclusions of law also address obviousness as of
24 March 2008.
25 b) Lovaza PDR (2007)
26 The Lovaza PDR (Physician’s Desk Reference) was published in 2007 and is prior
27 art to the patents-in-suit. 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 23 of 70



1 Lovaza PDR discloses a commercially-available preparation of EPA and DHA
2 administered at 4 grams/day, a pharmaceutical known as Lovaza. (Ex. 1535 at 2.) While
3 the Lovaza PDR published in the 2008 version of the Physician’s Desk Reference, Lovaza
4 was first commercially launched in 2004. (ECF No. 367 at 745:10-21.) Lovaza PDR
5 discloses that “Lovaza is indicated as an adjunct to diet to reduce triglyceride (TG) levels
6 in adult patients with very high (> 500 mg/dl) triglyceride levels.” (Ex. 1535 at 3.) As of the
7 alleged priority date, Lovaza was “widely used” and “a very successful drug.” (ECF No. 8 371 at 1891:7-12.)
9 Lovaza PDR discloses clinical trials in which Lovaza was administered as either
10 “add-on therapy” to a statin or as “monotherapy.” (Ex. 1535 at 2.) Under “High
11 Triglycerides: Add-on to HMG-CoA reductase inhibitor therapy,” the label explains:
12 The effects of Lovaza 4 g per day as add-on therapy to treatment with simvastatin were evaluated in a randomized, placebo-controlled, double-blind, parallel-group
13 study of 254 adult patients (122 on Lovaza and 132 on placebo) with persistent high triglycerides (200-499 mg/dL) despite simvastatin therapy (Table 1).
14 (Id.)
15 In this study, Lovaza PDR explains that all patients were treated with “simvastatin

16 40 mg per day for 8 weeks prior to randomization to control their LDL-C.” (Id.) After the
17 addition of Lovaza 4 g per day to simvastatin 40 mg per day, the median change in LDL-
18 C was an increase of 0.7% compared to baseline. (Id.) Relative to placebo, Lovaza 4 g
19 per day further “significantly reduced” TG and Apo B levels. (Id.) A POSA reading Lovaza
20 PDR would understand that “when Lovaza is used with simvastatin, Apo B is decreased
21 by 4.2 percent” and “there’s barely any LDL-C increase.” (ECF No. 371 at 1872:19-24.) In
22 fact, the combination of Lovaza and simvastatin essentially caused “zero” increase in LDL- 23 C. (Id. at 1872:22-1873:2.)
24 Lovaza PDR also discloses data under “Very High Triglycerides: Monotherapy” in
25 which “[t]he effects of Lovaza 4 g per day were assessed in two randomized, placebo-
26 controlled, double-blind, parallel group studies of 84 adult patients (42 on Lovaza, 42 on
27 placebo) with very high triglyceride levels (Table 2).” (Ex. 1535 at 2.) Table 2 summarizes
28 data from “two studies of 6 and 16 weeks duration.” (Id.) In the monotherapy study in

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 24 of 70



1 patients with very high triglycerides, treatment with Lovaza 4 g/day significantly reduced
2 triglycerides but also caused a significant increase in LDL-C (an increase of 44.5%
3 compared to baseline and 49.3% compared to placebo). (Id. at 3.)
4 Lovaza PDR therefore discloses “Lovaza treatment may result in elevations in LDL-
5 C and non-HDL-C in some individuals.” (Id.) However, as of March 2008, a skilled artisan
6 “would understand that if a patient experiences LDL-C increases from Lovaza, [a] statin
7 could be added to address that side effect.” (ECF No. 371 at 1891:22-25.) A skilled artisan
8 likewise knew that “Lovaza could be safely administered with statins” and was “typically 9 well-tolerated.” (Id. at 1874:22-24, 1893:9-11; see also ECF No. 367 810:11-14.) In fact,
10 Lovaza’s “rise in LDL-C was often offset by concurrent treatment with statins. The safety
11 and efficacy of using prescription Omega-3 in combination with a statin has been well- 12 established.” (Ex. 1953 at 233; see also ECF Nos. 371 at 1875:2-16, 367 at 809:21- 13 810:10.)
14 c) Mori (2000)
15 Mori, et al., Purified Eicosapentaenoic and Docosahexaenoic Acids Have
16 Differential Effects on Serum Lipids and Lipoproteins, LDL Particle Size, Glucose, and
17 Insulin in Mildly Hyperlipidemic Men, 71 Am. J. Clinical Nutrition 1085- 94 (2000) (“Mori”)
18 was published in 2000 and is prior art to the patents-in-suit.
19 Mori discloses “a double-blind, placebo-controlled trial of parallel design, [where]
20 59 overweight, nonsmoking, mildly hyperlipidemic men were randomly assigned to receive
21 4 g purified EPA, DHA, or olive oil (placebo) daily while continuing their usual diets for 6
22 wk.” (Ex. 1538 at 1-2.) The objective of Mori was “to determine whether eicosapentaenoic
23 (EPA) and docosahexaenic (DHA) acids have differential effects on serum lipids and
24 lipoporoteins.” (Id. at 1.)
25 Mori discloses that among the three treatment arms, “[c]apsules contained either
26 purified preparations of EPA ethyl ester (~96%), DHA ethyl ester (~92%), or olive oil (~75%
27 oleic acid ethyl ester).” (Id. at 2.) Further, “[n]one of the subjects were regularly taking
28 nonsteroidal antiinflammatory, antihypertensive, or lipid-lowering drugs or other drugs

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 25 of 70



1 known to affect lipid metabolism.” (Id. at 3.) Therefore, none of the patients in Mori were
2 on concurrent lipid-altering therapy. (ECF No. 367 at 739:22-25.)
3 Mori reports that triacylglycerols (TGs) “decreased significantly by 18.4% with EPA
4 (P = 0.012) and by 20% with DHA (P = 0.003).” (Ex. 1538 at 3.) A POSA would consider
5 this difference in triglyceride reduction “indistinguishable and of no clinical significance.”
6 (ECF No. 367 at 740:1-13.) A POSA would likewise recognize that Mori teaches that “4
7 grams pure EPA could reduce triglycerides by about 20 percent.” (ECF No. 371 at 8 1826:24-1827:5.)
9 Mori also reports that “[s]erum LDL cholesterol increased significantly with DHA (by
10 8%; P = 0.019), but not with EPA (by 3.5%; NS),” (Ex. 1538 at 3), “strongly suggesting
11 that these two Omega-3 fatty acids could have distinct effects on LDL cholesterol levels”
12 (ECF No. 367 at 740:1-17). In the Abstract, Mori summarizes these results as showing
13 that while “LDL, HDL, and HDL2 cholesterol were not affected significantly by EPA, . . .
14 DHA increased LDL cholesterol by 8% (P = 0.019).” (Ex. 1538 at 1; see also ECF No. 371
15 at 1827:8-11.) Mori concludes that “EPA and DHA had differential effects on lipids.” (Ex.
16 1538 at 1; see also ECF No. 371 at 1827:8-19.) Therefore, “a skilled artisan would
17 understand from Mori that DHA and EPA work differently.” (ECF No. 371 at 1829:6-8.)
18 d) Hayashi (1995)
19 Hayashi, et al., Decreases in Plasma Lipid Content and Thrombotic Activity by Ethyl
20 Icosapentate Purified from Fish Oils, 56(1) Curr. Therap. Res. 24-31 (1995) (“Hayashi”)
21 was published in 1995, and is prior art to the patents-in-suit.
22 Hayashi reports the daily administration of 1.8 grams per day of purified EPA over
23 a period of eight weeks to patients with a serum triglyceride level above 150 mg/dl. (Ex. 24 1532 at 4.)
25 Hayashi investigated the effects of EPA in patients with “familial combined
26 hyperlipidemia ([“]FCH[“]) showing phenotype Ila, Ilb, or IV.” (Id.) While Hayashi defined
27 all three phenotypes as “FCH,” (id.), a POSA would have understood that phenotype IV
28 refers to the Fredrickson system of classifying lipid disorders. (ECF No. 371 at 1866:10-

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 26 of 70



1 12.) Fredrickson Type IV is not limited to patients with triglycerides > 500 mg/dL. (See,
2 e.g., Ex. 2005 at 6 (reporting a Zocor study in which patients with Fredrickson Type IV had
3 a median triglyceride level of 404 mg/dL).) However, this phenotype includes patients with
4 severe hypertriglyceridemia. (See, e.g., Ex. 1986 at 21 (reporting a Lipitor study with a
5 median baseline triglyceride level of 565 mg/dL in patients with Fredrickson Type IV); Ex.
6 3007 at 11-12; Ex. 939 at 5 (reporting a Lovaza study “in patients with severe
7 hypertriglyceridemia, type IV, with 500 < TG < 2000 mg/dl”).)
8 A POSA would have understood that Hayashi includes at least one patient with
9 triglyceride levels > 500 mg/dL in light of Hayashi’s data. (ECF No. 367 at 725:21-727:1.)
10 Table I reports that at baseline, the patients in the study had a triglyceride level of 300 ±
11 233 mg/dl. (Ex. 1532 at 5.) Dr. Heinecke11 explained that while “there is some ambiguity
12 in this paper about what the meaning is of the plus minus 233[,] . . . overwhelmingly, in the
13 medical literature, that would be a standard deviation.” (ECF No. 367 at 725:21-727:1.)
14 The standard deviation is the average spread of the data around the mean value
15 of 300 mg/dl (for a normal distribution of data, two-thirds of the data points are within one
16 standard deviation of the mean). (Id.) Accordingly, as Dr. Heinecke explained, “ecause
17 there’s a value of plus or minus 233, there was at least one patient in that study who had
18 a value of greater than 300, and because that’s only encompassing two-thirds of the data,
19 one-sixth of the patients would likely have been above 533.” (Id.) Although Dr. Lavin
20 initially told the PTO12 that not even one patient in Hayashi would have had triglyceride
21 levels > 500 mg/dL, Dr. Lavin later testified that he would “rewrite” his declaration on this
22 point, explaining that in Hayashi “you know that there must be at least one subject” with
23 triglyceride levels > 500 mg/dL, and that it is “likely that you have one or two observations 24
25
26 11Defendants’ invalidity expert.
27 12Plaintiffs submitted a declaration from Dr. Lavin to overcome an initial rejection for obviousness of the ’889 Application. (See ECF No. 324 at 16-18 (stipulating to facts
28 providing more details about these interactions).)

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 27 of 70



1 above 533.” (Lavin Dep. Tr. at 102:24-103:21.) Dr. Toth13 did not “offer any type of
2 statistical opinion to corroborate what Dr. Lavin told the patent office.” (ECF No. 371 at 3 1868:13-16.)
4 Dr. Heinecke explained that there is an alternative theory that Hayashi’s reference
5 to 300 ± 233 mg/dl instead refers to the range of triglyceride values, rather than the
6 standard deviation. (ECF No. 367 at 725:21-727:1.) But “this would be very unusual,” and
7 in any case, under that interpretation there would still be “at least one patient in the study
8 that had a value of 533.” (Id.) Therefore, under either interpretation of Hayashi, at least
9 one patient had triglyceride levels > 500 mg/dL. (Id. at 727:2-6.)
10 Hayashi discloses that “[a]fter 8 weeks, patients treated with ethyl icosapentate
11 showed significant reductions in . . . triglyceride (41%),” and reports reductions in LDL-C
12 (7%) and apolipoprotein B (7%), which was not statistically significant. (Ex. 1532 at 5.)
13 Hayashi therefore concludes that “[p]urified icosapentate (1800 mg/d for 8 weeks)
14 decreased total cholesterol and triglyceride in patients with FCH (Table I),” and that “[n]o
15 overt effects of icosapentate on plasma LDL-C and HDL-C were seen, although a
16 decrease in LDL-C was noted (Table I).” (Id. at 7.)
17 Hayashi does not report the LDL-C data of patients with triglycerides > 400 mg/dL
18 because Hayashi used the Friedewald equation to calculate LDL-C levels. (Id. at 5; see
19 also ECF No. 367 at 798:23-800:7.) The Friedewald equation is commonly used in clinical
20 studies to calculate LDL-C levels and operates by using triglyceride levels to estimate
21 LDL-C levels, but “is not accurate for triglycerides above 400 milligrams per deciliter.”
22 (ECF No. 367 at 798:23-800:7.) But while Hayashi does not report LDL-C data in patients
23 with triglycerides > 400 mg/dL, Hayashi does not limit its conclusion regarding EPA’s
24 effects on LDL-C levels to patients with lower triglyceride levels. Hayashi concludes that
25 “[a]lthough the effects of fish oils on plasma LDL-C and HDL-C are complex, judging from
26 the present study, purified icosapentate apparently has no deleterious effect on plasma 27
28 13Plaintiffs’ invalidity expert.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 28 of 70



1 LDL-C or HDL-C in patients with FCH.” (Ex. 1532 at 7.) Again, some patients with FCH—
2 including at least one patient in the Hayashi study—have triglyceride levels above 500 3 mg/dL. (Id.; see also ECF No. 367 at 725:21-727:1; Lavin Dep. Tr. at 102:24-103:21.)
4 e) Kurabayashi (2000)
5 Kurabayashi, et al., Eicosapentaenoic Acid Effect on Hyperlipidemia in Menopausal
6 Japanese Women. Obstet. Gynecol. 96:521-8 (2000) (“Kurabayashi”) was published in
7 2000 and is prior art to the patents-in-suit.
8 Kurabayashi investigated the effects of administering purified EPA (96.5% EPA) at
9 a dose of 1.8 g/day in combination with estriol (the “EPA group”) as compared to estriol
10 therapy alone (the “control group”) for forty-eight weeks to hyperlipidemic, menopausal
11 women. (Ex. 1534 at 1.) Estriol is a form of estrogen that is commonly used in menopausal
12 women to alleviate the symptoms of menopause. (ECF No. 367 at 735:2-20.) As an
13 estrogen, estriol is known to elevate triglyceride levels. (Id.)
14 Despite coadministration with estriol, Kurabayashi reports a statistically significant
15 27% reduction in triglyceride levels in the EPA group. (Ex. 1534 at 3.) As compared to the 16
17
18
19
20
21
22
23
24
25
26
27
28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 29 of 70



1 control group, the EPA group experienced a statistically significant reduction in triglyceride
2 levels at the 12, 24, and 48-week checkpoints:
3
4
5
6
7
8
9
10
11
12
13
14
15
16 (Id. at 4.) Kurabayashi further reports that “[l]ow-density lipoprotein cholesterol levels in
17 both groups were significantly lower.” (Id. at 3.)
18 Kurabayashi further reports a statistically significant reduction in Apo B levels in the
19 EPA group of 6.9%. (Id. at 4-5.) With a p-value of < .001, EPA’s effects on Apo B were
20 highly significant. (Id.; see also ECF No. 367 at 737:1-23.) In contrast, Kurabayashi reports
21 a non-statistically significant 1.5% reduction in Apo B levels in the control group: 22
23
24
25
26
27
28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 30 of 70



1
2
3
4
5
6
7
8
9 (Ex. 1534 at 5; see also ECF No. 367 at 737:1-23.)
10 The results reported in Kurabayashi do not suggest any interaction or synergy
11 between EPA and estriol. (ECF No. 367 at 735:21-736:9.) Instead, synergy is usually only
12 seen between drugs that have similar effects, such as two drugs that reduce blood
13 pressure. (Id.)
14 In light of the statistically-significant differential effects reported between the EPA
15 and control groups, a POSA would have attributed the reduction in Apo B to EPA. (Id. at 16 737:24-738:8.)
17 f) Rambjør 1996
18 Plaintiffs rely on Rambjør to argue that a POSA would have understood that EPA
19 increased, not decreased, LCL-C levels. (ECF No. 377 at 224-26.) Rambjør reports that
20 EPA “produced significant decreases in both TG and very low density lipoprotein (VLDL)
21 cholesterol,” but was also associated with a statistically significant “increase[] in low
22 density lipoprotein cholesterol levels.” (Ex. 1961 (Rambjør, et al., Eicosapentaenoic Acid
23 Is Primarily Responsible for Hypotriglyceridemic Effect of Fish Oil in Humans, 31 Lipids S-
24 45 (1996) (“Rambjor”)) at 3.) But Rambjor used only 3 g/day of EPA that was only 91%
25 pure. (Id.) Because “omega-3s are complex,” Dr. Toth testified that a skilled artisan “would
26 have no idea” what fatty acids are in the other 9%, which could have included a substantial
27 amount of DHA. (ECF No. 371 at 1814:17-22.) 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 31 of 70



1 Rambjor does not appear authoritative for other reasons as well. Rambjor
2 consolidated data from three separate studies, and only included 9 patients in the DHA
3 group. (Ex. 1961 at 4.) Rambjor further only included a 2-week washout period, and
4 patients were only given EPA or DHA for a period of 3 weeks. (Id. at 3.) The Rambjor
5 study was therefore underpowered, and its design of comparing the effects of two drugs
6 with a significantly different number of subjects in each group was unusual. (ECF No. 367
7 at 782:4-783:1.) Rambjor itself concluded that “[f]urther studies are needed to clearly
8 define individual effects of EPA and DHA on human lipid metabolism.” (Ex. 1961 at 6.)
9 Mori is “one of those further studies” that clearly defined the individual effects of
10 EPA and DHA on human lipid metabolism. (ECF No. 371 at 1842:10-17.) Mori, which
11 published after Rambjor, criticized Rambjor’s design as studying “only a small number of
12 subjects in the DHA group,” for being of “short duration,” and for including “only a 2-wk
13 washout period between treatments.” (Ex. 1538 at 5, 9.) In contrast to Mori—which studied
14 the claimed EPA dose and purity (4/g day at 96% purity), (Ex. 1538 at 2)—the EPA studied
15 in Rambjor was only 91% pure and administered at only 3 g/day (Ex. 1961 at 3; see also
16 ECF No. 371 at 1841:7-1842:1). A POSA as of March 2008 thus would have relied on the
17 teachings of Mori over those in the earlier Rambjor reference—particularly if the skilled
18 artisan were focusing on a dose of 4 g/day and at least 96% purity, as used in Mori but
19 not in Rambjor. (ECF No. 367 at 784:22-785:2.) This is evidenced by the fact that Mori
20 has been repeatedly cited in the literature, including Plaintiffs’ internal documents and
21 submissions to the FDA, but Plaintiffs have not identified any trial exhibit that cites Rambjor
22 other than von Schacky, discussed below. (See, e.g., Ex. 1816 at 68 (summarizing over a
23 dozen prior-art EPA studies to FDA, including Mori but not Rambjor); Ex. 1800 at 12-13
24 (summarizing DHA and EPA’s effects on LDL-C in an investor presentation and citing Mori
25 but not Rambjor).)
26 g) Von Schacky (2006)
27 Another reference relied on by Plaintiffs (see, e.g., ECF No. 377 at 226-229), von
28 Schacky, did not report any primary data on EPA or DHA’s effects, but reported in a table

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 32 of 70



1 that studies suggested that both EPA and DHA increase LDL-C. (Ex. 1605 (von Schacky,
2 A review of omega-3 ethyl esters for cardiovascular prevention and treatment of increased
3 blood triglyceride levels, Vascular Health and Risk Management 2(3):251-262 (2006)
4 (“von Schacky”)) at 9; see also ECF No. 371 at 1844:9-14.) The table, however, merely
5 included arrows pointing in different directions and did not attribute any significance to any
6 of the variables reported. (Ex. 1605 at 9; see also ECF No. 367 at 785:23-786:22.)
7 Von Schacky further reported inconsistent information, citing Mori and claiming that
8 “n more recent comparative studies, no effects of either EPA or DHA . . . were seen on
9 LDL levels.” (Ex. 1605 at 5.) But as Dr. Toth conceded, “[t]hat’s not what Mori said.” (ECF
10 No. 371 at 1847:8-17.) Mori expressly reports that “[s]erum LDL cholesterol increased
11 significantly with DHA (by 8%; P = 0.019).” (Ex. 1538 at 1.) Because von Schacky is a
12 review article, a skilled artisan also would have looked at the underlying clinical studies
13 cited by von Schacky, including Mori. (ECF No. 371 at 1848:4-8.)
14 In any event, as Dr. Heinecke explained, because EPA is LDL-neutral, one would
15 expect to see small increases or decreases across studies due to chance alone. (ECF No.
16 367 at 740:18-25.) Therefore, if among the available literature on EPA’s effects on LDL-C
17 one saw “one-third of the studies showing an increase, one-third of the stud[ies] showing
18 a decrease, and one third of the stud[ies] showing no effect, that would be very strong
19 evidence that there was no overall effect on the intervention.” (Id. 781:21-782:3.)
20 2. Level of Ordinary Skill in the Art
21 The determination of obviousness must be done based on the knowledge
22 possessed by one of ordinary skill in the art at the time the invention was made. The
23 Asserted Claims and the prior art are evaluated at the time of the invention from the
24 standpoint of a POSA. A POSA is a hypothetical person who is presumed to have access
25 to, and be aware of, all of the relevant prior art at the time of the invention. See, e.g.,
26 Rothman v. Target Corp., 556 F.3d 1310, 1318 (Fed. Cir. 2009). Factors that may be
27 considered in determining the level of ordinary skill in the art may include: (1) type of
28 problems encountered in the art; (2) prior art solutions to those problems; (3) rapidity with

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 33 of 70



1 which innovations are made; (4) sophistication of the technology; and (5) educational level
2 of active workers in the field. See Daiichi Sankyo Co. v. Apotex, Inc., 501 F.3d 1254, 1256
3 (Fed. Cir. 2007). Thus, it is not permissible to use hindsight after viewing the claimed
4 invention to determine questions of obviousness or to rely at all on the teachings of the
5 claimed invention in determining whether one of ordinary skill in the art would find the
6 invention obvious. See, e.g., Millennium Pharm., Inc. v. Sandoz Inc., 862 F.3d 1356, 1367
7 (Fed. Cir. 2017) (“The inventor’s own path itself never leads to a conclusion of
8 obviousness; that is hindsight. What matters is the path that the person of ordinary skill in
9 the art would have followed, as evidenced by the pertinent prior art.”) (quoting Otsuka
10 Pharm. Co. v. Sandoz, Inc., 678 F.3d 1280, 1296 (Fed. Cir. 2012)).
11 Plaintiffs and Defendants proposed different definitions of the POSA, but those
12 differences are not material because both sides made clear their arguments apply with
13 equal force regardless of the definition the Court adopts. (ECF Nos. 373 at 64-65, 377 at
14 173-174.) The Court therefore assumes without deciding that one of the two definitions
15 that follow below applies to its conclusions of law.
16 Plaintiffs proposed the following definition. (ECF No. 377 at 173-174.) The POSA
17 in this case world be (1) a clinician with an M.D., or D.O. and at least 2 to 3 years of
18 experience in the diagnosis, evaluation, and treatment of lipid blood disorders, including
19 severe hypertriglyceridemia (i.e., TG levels of at least 500 mg/dl), or (2), alternatively, a
20 clinician, such as a nurse practitioner or physician’s assistant, with 3 to 5 years of
21 experience in the diagnosis, evaluation, and treatment of lipid blood disorders, including
22 severe hypertriglyceridemia. (See id.)
23 Defendants proposed the following definition. (ECF No. 373 at 64-65.) “[T]he POSA
24 to whom the patents in-suit are directed would have had (a) at least a medical degree or
25 an advanced degree in the field of lipid biochemistry; (b) several years of experience in
26 the development and/or clinical use of fatty acids to treat blood lipid disorders, including
27 fish oil based fatty acids, i.e., EPA and DHA, and their dosage forms; and (c) access to a 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 34 of 70



1 team including one or more of a medical doctor, an analytical chemist, or a pharmaceutical
2 chemist.”14 (Id. at 64.)
3 3. Differences between the Prior Art and the Claims at Issue
4 The primary difference between the prior art and the Asserted Claims is that the
5 Lovaza PDR, Defendants’ principal prior-art reference, used a mixture of DHA and EPA,
6 while the Asserted Claims involve a pharmaceutical composition containing purified EPA,
7 but substantially no DHA. Defendants additionally point to other pieces of prior art to
8 explain why the Other Health Benefit Claims were obvious.
9 Here, all 10 Asserted Claims recite the same method of treatment—namely, a
10 method of reducing triglycerides in a patient with triglycerides of at least 500 mg/dL by
11 administering, for at least 12 weeks, about 4 g/day of at least 96% purified EPA. (Ex. 1500
12 (’728 patent claims 1 and 16); Ex. 1502 (’715 patent claim 14); Ex. 1504 (’677 patent
13 claims 1 and 8); Ex. 1506 (’562 patent claim 1); Ex. 1514 (’560 patent claims 4 and 17);
14 Ex. 1516 (’929 patent claims 1 and 5).) The Lovaza PDR taught a method of treating
15 patients with triglycerides of at least 500 mg/dL by administering, for at least 12 weeks, 4
16 g/day of a mixture of EPA and DHA. (Ex. 1535 at 2-3.)
17 The Lovaza PDR warned, however, that this method of treatment could
18 substantially increase patients’ LDL-C levels (at least at a median triglyceride level of 816
19 mg/dL), which was undesirable. (Id. at 3.) Mori taught that DHA increased LDL-C, whereas
20 4 g/day of 96% purified EPA reduced triglycerides without increasing LDL-C. (Ex. 1538 at
21 2-3.) Other prior art (e.g., Kurabayashi and Hayashi) similarly taught that EPA did not
22 increase LDL-C in patients with triglyceride levels up to 400 mg/dL. (ECF No. 367 at 23 715:10-716:4, 759:10-760:1.)
24

25 14Though, as stated, the Court does not choose between the two definitions of the
26 POSA proposed by the parties, Defendants’ proposed definition strikes the Court as more reasonable because it appears calculated to include a person who develops drugs, rather
27 than merely people who would be able to treat a blood lipid disorder like Plaintiff’s definition does. The key obviousness disputes in this case focus on drug development, not merely
28 treatment, of blood lipid disorders.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 35 of 70



1 4. Secondary Considerations
2 The Court’s obviousness inquiry must also consider whether objective indicia of
3 non-obviousness support the Asserted Claims. “Such secondary considerations as
4 commercial success, long felt but unsolved needs, failure of others, etc., might be utilized
5 to give light to the circumstances surrounding the origin of the subject matter sought to be
6 patented.” Graham v. John Deere Co. of Kansas City, 383 U.S. 1, 17-18 (1966); see also
7 In re Rouffet, 149 F.3d 1350, 1355 (Fed. Cir. 1998) (explaining that objective evidence of
8 nonobviousness may include copying, long-felt but unsolved need, failure of others,
9 commercial success, unexpected results created by the claimed invention, unexpected
10 properties of the claimed invention, licenses showing industry respect for the invention,
11 and skepticism of skilled artisans). The Court discusses below its factual findings relevant
12 to its analysis of secondary considerations included in its conclusions of law further below.
13 a) REDUCE-IT
14 Plaintiffs point to the results of the REDUCE-IT study as objective evidence of
15 nonobviousness. (ECF No. 379 at 35-37.) The REDUCE-IT study was “a multicenter,
16 randomized, double-blind, placebo-controlled trial involving patients with established
17 cardiovascular disease or with diabetes and other risk factors, who had been receiving
18 statin therapy and who had a fasting baseline triglyceride level of 135 to 499” mg/dl and a
19 fasting baseline LDL-C level of 41 to 100 mg/dl. (Ex. 1641 at 1 (the “Bhatt Article”).)
20 Each subject in REDUCE-IT had a fasting baseline triglyceride level of 135 to 499
21 mg/dl. (Id. at 2.) “[B]ecause of the intraindividual variability of triglyceride levels, the initial
22 protocol allowed for a 10% lower triglyceride level from the target lower limit, which
23 permitted patients to be enrolled if they had a triglyceride level of at least 135 mg per
24 deciliter.” (Id.) In May 2013, the first protocol amendment “changed the lower limit of the
25 acceptable triglyceride level from 150 mg per deciliter to 200 mg per deciliter, with no
26 allowance for variability.” (Id.)
27 Nevertheless, there was a substantial fraction of patients in the REDUCE-IT Study
28 with median triglyceride values <150 mg/dL during the study, given that the inclusion

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 36 of 70



1 criteria for triglycerides was limited to the screening exam for entry into the study and
2 because triglyceride levels can vary over a wide range. More specifically, about 10% of
3 subjects had triglyceride levels below 150 mg/dl, about 30% had triglyceride levels
4 between 150 and 200 mg/dl, and the remaining subjects had triglyceride levels about 200
5 mg/dl. (Id. at 4, Table 1.)
6 While a small subset of patients had triglyceride levels that rose above 500mg/dl at
7 some point in time during the REDUCE-IT study due to intraindividual variability,
8 “REDUCE-IT focused on patients with triglycerides below 500.” (ECF No. 371 at 1894:12-
9 14.) Again, “eligible patients . . . had to have a fasting triglyceride level of 150 to 499
10 milligrams per deciliter. This is less than 500 milligrams per deciliter.” (ECF No. 367 at
11 818:18-21.) Thus, REDUCE-IT was not “designed to evaluate patients [with] triglycerides
12 above 500” and did not include any patients with a baseline triglyceride level of 500 mg/dl
13 or above. (Id. at 819:14-16.) Dr. Budoff agreed that “REDUCE-IT focused on a different
14 patient population than the patient population” for Defendants’ labels. (ECF No. 366 at
15 530:16-19.) In fact, the MARINE study and REDUCE-IT study, and thus the related
16 indications, involved “completely different patient populations.” (Id. at 589:21-1.)
17 Additionally, “[a]ll the patients in REDUCE-IT were taking statins.” (ECF No. 371 at
18 1896:15-17.) More specifically, “[e]ligible patients . . . had been receiving a stable dose of
19 a statin for at least 4 weeks.” (Ex. 1641 at 2; see also ECF No. 367 at 821:9-22.) Thus, “in
20 REDUCE-IT, we’re talking about patients who are already on a statin for controlling their
21 bad cholesterol.” (ECF No. 365 at 271:10-13.) “REDUCE-IT did not have a monotherapy
22 arm,” i.e. an arm with patients not taking a statin. (ECF No. 371 at 1897:5-7.) In fact, “it
23 would have been unethical to have just a Vascepa monotherapy arm. The FDA would
24 never allow it because statin therapy is the standard of care for patients in secondary
25 prevention for high risk diabetic patients.” (Id. at 1897:7-10.) And approximately 58.6% of
26 the patients enrolled in the treatment arm of the REDUCE-IT Study were diabetics. (Ex. 27 1641 at 4, Table 1.)
28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 37 of 70



1 Patients in REDUCE-IT were randomly assigned to receive either 4 g/day of
2 Vascepa or placebo (mineral oil). (Id. at 1-2.) “The primary efficacy end point was a
3 composite of cardiovascular death, nonfatal myocardial infraction (including silent
4 myocardial infarction), nonfatal stroke, coronary revascularization, or unstable angina in a
5 time-to-event analysis.” (Id. at 3.) “The key secondary end point [was] a composite of
6 cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke in a time-to-event
7 analysis.” (Id. at 3.) A total of 8179 patients were enrolled and were followed for a median 8 of 4.9 years. (Id. at 1, 5.)
9 “The median change in triglyceride level from baseline to 1 year was a decrease of
10 18.3% . . . in the icosapent ethyl group and an increase of 2.2% . . . in the placebo group.”
11 (Id. at 5.) The median reduction [in triglyceride level] from baseline . . . was 19.7% greater
12 in the icosapent ethyl group than in the placebo group.” (Id.) “Baseline triglyceride levels
13 (≥ 150 vs. <150mg per deciliter or ≥200 or < 200 mg per deciliter) had no influence on the
14 primary or key secondary efficacy end points.” (Id. at 7.) “The attainment of triglyceride
15 levels of 150 mg per deciliter or higher or below 150 mg per deciliter at 1 year after
16 randomization also had no influence on the efficacy of icosapent ethyl as compared with
17 placebo with respect to the primary or key secondary efficacy end point.” (Id.)
18 Thus, the REDUCE-IT benefits “occur[ed] irrespective of the attained triglyceride
19 level,” and “the cardiovascular risk reduction was not associated with attainment of a more
20 normal triglyceride level.” (Id. at 10; see also ECF No. 367 at 817:2-5.) As Dr. Toth pointed
21 out, “even if [a subject] didn’t normalize [their] triglycerides in [the] trial, [they would] still
22 derive a benefit.” (ECF No. 370 at 1624:18-20.) With respect to LDL-C levels, “[t]he median
23 change in LDL cholesterol level from baseline was an increase of 3.1% . . . in the icosapent
24 ethyl group and an increase of l0.2% . . . in the placebo group.” (Ex. 1641 at 5.) REDUCE-
25 IT “found no substantial difference in the benefit” of EPA based on whether patients “had
26 an increase in LDL cholesterol levels at 1 year or had no change or a decrease in LDL
27 cholesterol levels.” (Id. at 7.) Thus, “[t]here was no relationship to the change in LDL 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 38 of 70



1 cholesterol levels to the benefit in terms of cardiovascular risk reduction.” (ECF No. 367 2 at 820:22-24.)
3 In November 2018, Plaintiffs announced that REDUCE-IT identified a cardiac
4 benefit in patients receiving Vascepa as compared to placebo. The results show that “[a]
5 primary end-point event occurred in 17.2% of the patients in the icosapent ethyl group, as
6 compared with 22.0% of the patients in the placebo group.” (Ex. 1641 at 1.) “A key
7 secondary efficiency end-point event . . . occurred in 11.2% of the patients in the icosapent
8 ethyl group, as compared with 14.8% of the patients in the placebo group.” (Id. at 5.) The
9 rate of cardiovascular death was 4.4% in the icosapent ethyl group and 5.2% in the
10 placebo group. (Id. at 7.) According to the Kaplan-Meier plots—which demonstrate results
11 for certain time intervals—in the Bhatt Article, the cardiac benefits were not observed until
12 patients had been taking 4 g/day of Vascepa for a year or more. (Id. at 5.)
13 In other words, there is no “evidence that the cardiovascular risk reduction in
14 REDUCE-IT occurs within 12 weeks . . . Instead there is no divergence [between the
15 treated group and placebo group] in terms of cardiovascular risk until year one, and that
16 difference did not become statistically significant until year two.” (ECF No. 367 at 819:22-
17 24.) Thus, “it takes time to accrue the [cardiovascular benefit], and if you stop it at four
18 months . . . then you’re going to lose that benefit.” (ECF No. 371 at 1896:10-14.)
19 Based on these REDUCE-IT results, FDA approved Vascepa to reduce the risk of
20 “myocardial infraction, stroke, coronary revascularization, and unstable angina requiring
21 hospitalization” in patients that had “elevated triglyceride (TG) levels (≥ 150 mg/dL),” and
22 either an “established cardiovascular disease or diabetes mellitus and 2 or more additional
23 risk factors for cardiovascular disease.” (Ex. 2248 at 1.)
24 “Amarin has separate patents covering the method used in the REDUCE-IT study
25 . . . [and] those patents are not being asserted in this case.” (ECF No. 371 at 1895:4-10.)
26 Amarin submitted a Form 3542a for the REDUCE-IT sNDA. (Ex. 2250.) Through this form,
27 Plaintiffs represented to FDA that only the patents listed relate to Vascepa’s REDUCE-IT
28 indication. (Ex. 2299.) None of the asserted patents were listed. If Plaintiffs believed that

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 39 of 70



1 the asserted patents claimed “a method of using [Vascepa] that is the subject of” the
2 REDUCE-IT indication, they would have had to list those patents on the Form 3542a
3 included with their sNDA. (Ex. 2250.) See also 21 C.F.R. § 314.53(b). As discussed above,
4 there is no overlap between the patents listed for the REDUCE-IT indication and the
5 asserted patents. (Ex. 2299.)
6 b) Commercial Success
7 The parties dispute whether Vascepa, which embodies the Asserted Claims, is a
8 commercial success. Predictably, Plaintiffs argue it is (ECF No. 379 at 37-38), Defendants
9 argue it is not (ECF No. 378 at 32). The parties also presented competing expert testimony
10 on this topic at Trial. (ECF No. 369.) Having considered the expert testimony and other
11 evidence presented by both sides, the Court finds Plaintiffs’ argument—that Vascepa is a
12 commercial success—more persuasive.
13 More specifically, substantial and sustained increases in Vascepa prescriptions,
14 net sales, and market share, as well as Vascepa’s positive net present value (“NPV”),
15 demonstrate that Vascepa is a commercial success. (ECF No. 369 at 1423:3-15.)
16 Prescriptions for Vascepa have grown substantially since the product’s launch in
17 January 2013. 174,000 prescriptions for Vascepa were filled in 2013, and the number
18 increased every year, reaching 1.3 million prescriptions in 2018, an average annual
19 increase of about 50%. (Id. at 1427:9-17.) This increase indicates that patients and health
20 insurers are willing to pay a premium for the features of Vascepa, given that a relatively
21 inexpensive generic version of Lovaza has been available since 2014. (Id. at 1427:18- 22 1428:3.)
23 Vascepa’s net sales have also grown substantially since the product’s launch.
24 Vascepa’s net sales were $26 million in 2013 and have increased every year, reaching
25 $228 million in 2018, an average annual increase of 54%. (Id. at 1429:2-9.) The increase
26 indicates that the product is providing value and that patients and health insurers are
27 willing to pay a premium for the features of Vascepa. (Id. at 1429:10-15.) Moreover, the
28 Court finds Defendants’ contention that Vascepa’s sales are driven by rebates and

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 40 of 70



1 discounts unpersuasive. (ECF No. 373 at 113.) The net sales metric relied upon by Dr.
2 Nicholson already accounts for all rebates and discounts. (ECF No. 369 at 1304:17-23,
3 1429:22-1430:5, 1431:3-14.) In any case, the level of rebates and discounts provided for
4 Vascepa is in line with the industry norm. (Id. at 1431:3-14, 1433:12; see also Ex. 746 at 5 5, 10.)
6 Vascepa’s share of the market for omega-3 fatty acid drugs has also grown every
7 year since its launch. Vascepa’s share of omega-3 fatty acid prescriptions was 4% in 2013,
8 increasing to 32% in 2018. (ECF No. 369 at 1435:3-16.) In contrast, branded Lovaza’s
9 share of the same market decreased from approximately 96% in 2013 to under 5% in
10 2018. (Id. at 1436:19-1437:7.) Vascepa’s share of the broader market for TG-reducing
11 drug prescriptions also increased from 1% in 2013 to 6% in 2018. Vascepa’s increasing
12 market share is a strong indicator of its increasing value over time. (ECF No. 369 at
13 1434:8-24, 1435:17-1436:3.) In fact, every other TG-reducing drug’s prescriptions were
14 decreasing from 2013 to 2018, whereas Vascepa’s prescriptions increased in the same
15 period. That Vascepa has bucked the trend speaks highly of its performance in the market. 16 (Id. at 1438:7-18.)
17 Vascepa’s NPV also demonstrates its commercial success. NPV is the most
18 common method that pharmaceutical companies use to determine whether to launch a
19 new product and to track whether the product is successful. (Id. at 1440:1-15, 1444:22- 20 1445:1, 1469:20-1470:7; see also Ex. 600 at 2, 5; Ex. 602 at 5.) A positive NPV means
21 that the product is more profitable than the average for similar products in the industry. 22 (ECF No. 369 at 1440:16-1441:14, 1443:18-21; Ex. 602 at 10 (“Any time you find and
23 launch a positive NPV project, a project with present value exceeding its required cash
24 outlay, you have made your company’s stockholders better off.”). Vascepa’s NPV is
25 expected to be zero in 2024, which means that its investors will have recouped their
26 investment and received the industry average return in Vascepa’s twelfth year in the
27 market. (ECF No. 369 at 1458:5-20.) Over its entire lifecycle, Vascepa is expected to have 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 41 of 70



1 a positive NPV of $1.9 billion, which means that it will deliver a return that exceeds the
2 industry average by $1.9 billion. (Id. at 1458:21-1459:4.)
3 Defendants’ contention that Vascepa is not a commercial success is largely based
4 on the theory that Vascepa did not make a profit in its first six years on the market. But
5 Defendants ignore the reality that drugs have long lifecycles, the beginning of which
6 involves spending vast amounts of money on R&D. (Id. at 1441:15-1442:7; see also Ex.
7 612 at 2.) Here, Plaintiffs spent $465 million in research and development between 2008
8 and 2018. (ECF No. 369 at 1426:17-24.) Moreover, marketing spending tends to be higher
9 at the beginning of a pharmaceutical product’s lifecycle, given the need to educate
10 physicians about the clinical profile of the new drug in question. (Id. at 1306:11-1307:2,
11 1471:7-1472:1.) At the same time, it can take as long as 12 years for new drugs in the top
12 ten percent of sales to achieve peak sales. (Id. at 1468:11-1469:4; see also Ex. 607 at
13 20.) Indeed, a study has shown that it took drugs 16 years on average to reach NPV of
14 zero. (ECF No. 369 at 1469:20-1470:7; see also Ex. 612 at 6.) Therefore, the
15 pharmaceutical industry considers the entire lifecycle of a drug in analyzing commercial
16 success rather than just the first six years after the drug’s launch. (ECF No. 369 at 17 1445:23-1446:19, 1468:11-1469:4, 1512:17-24; see also Ex. 600 at 2.) Defendants’
18 alternative approach, which relies on taking a snapshot of Vascepa’s performance after
19 Plaintiffs have incurred the vast majority of the R&D spending, but before they have
20 enjoyed the fruits of that spending, is less persuasive in light of the testimony at Trial
21 regarding industry practice.
22 Defendants also contend that Dr. Nicholson’s NPV analysis is unreliable because
23 it was excessively influenced by the one of the five forecasts upon which he relied.
24 Defendants’ contention is unpersuasive. The forecast in question is from a firm called H.C.
25 Wainwright, which (as the evidence showed) does not have a history of systematically
26 overestimating Amarin’s revenue or profit. (ECF No. 369 at 1460:22-1463:18; see also Ex.
27 752 at 2; Ex. 637 at 63; Ex. 658 at 3; Ex. 724 at 4.) In any event, Vascepa’s NPV is
28 expected to be positive whether or not H.C. Wainwright’s forecast is included. (ECF No.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 42 of 70



1 369 at 1465:3-10, 1504:1-16, 1521:6-18.) This shows that Dr. Nicholson’s NPV analysis
2 is robust and reliable. Dr. Nicholson’s NPV analysis is also consistent with Defendant
3 Hikma’s own January 2020 presentation to investors, which ranks Vascepa as having the
4 fourth highest U.S. market size among all the drugs in Hikma’s generic pipeline. (Ex. 1218
5 at 12.) In sum, the Court finds that Vascepa is a commercial success.
6 c) Praise
7 Plaintiffs also argue that praise for Vascepa weighs in favor of finding the Asserted
8 Claims nonobvious. (ECF No. 377 at 269-271.) However, the Court finds that the evidence
9 Plaintiffs proffer to show praise is more qualified and equivocal than Plaintiffs represent in
10 their briefing. Thus, the Court finds Plaintiffs’ proffered evidence of praise does not weigh
11 in favor of finding the Asserted Claims nonobvious.
12 Plaintiffs’ expert Dr. Toth cited several articles as purported evidence of such praise
13 at Trial, but none of them support his opinion. (ECF Nos. 370 at 1722:15-5, 371 at
14 1848:11-20.) First, Dr. Toth cited the O’Riordan article, which quoted several doctors on
15 the results of MARINE. (Ex. 1581.) Specifically, Dr. Toth cited a statement by Dr. McGuire
16 that “if you can have favorable cardiovascular effects without raising LDL cholesterol,
17 that’s going to be an advantage,” and a statement by Dr. Nissen that this “gives you all the
18 benefit without the downside.” (Id. at 1-2; see also ECF No. 370 at 1606:24-1612:24.) But
19 as the article reveals, neither doctor gave unmitigated praise; both expressed caveats
20 about those statements. Dr. McGuire “was cautious in interpreting the results” of MARINE,
21 “insert[ed] a dose of caution,” and made clear that his focus was on “cardiovascular
22 effects,” not just triglyceride reduction. (Ex. 1581 at 1.) If anything, Dr. McGuire saved his
23 praise for “trials such as Japan EPA Lipid Intervention Study ([“]JELIS[”]),” which actually
24 “showed a favorable signal of reduced cardiovascular events.” (Id.) Similarly, Dr. Nissen
25 “expressed the same caveats” about MARINE, and noted that he “would like to eventually
26 see a head-to-head comparison between Lovaza” and Vascepa, which to date has never
27 been done. (Id. at 2.) Even apart from these caveats, Dr. Toth ignored the statement by
28 Dr. Blumenthal, which O’Riordan also reported. As discussed above, Dr. Blumenthal did

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 43 of 70



1 not praise Vascepa or MARINE, but instead dismissed MARINE’s significance because
2 typical increases in LDL-C with Lovaza were “‘modest’ and ‘not that big an issue,’”
3 especially since Lovaza “works well with statins.” (Id. at 2.) Given these conflicting
4 statements, O’Riordan as a whole does not suggest that Vascepa’s ability to avoid
5 increases in LDL-C has been praised by the industry.
6 Second, Dr. Toth relied on articles by Fialkow (Ex. 852) and Castaldo (Ex. 866).
7 (ECF No. 370 at 1612:25-1615:13.) But those articles merely state the fact that Vascepa
8 does not increase LDL-C—they do not praise Vascepa for that reason (or indeed, for any
9 reason). The statement that Dr. Toth quoted from Fialkow states that “treatment with the
10 EPA-only product, icosapent ethyl [i.e., Vascepa] has no LDL-C monitoring requirement.”
11 (Ex. 852 at 5.) Similarly, the statement that Dr. Toth quoted from Castaldo states that
12 Vascepa “does not increase LDL-C levels, as supported by clinical studies and the
13 icosapent ethyl product label.” (Ex. 866 at 6.) These matter-of-fact observations, which
14 merely repeat information from the Vascepa product label and the MARINE trial, do not
15 praise Vascepa or the claimed invention. As the Federal Circuit has made clear, such
16 “journal citations that reference the findings stated in [the patentee’s] published efficacy
17 studies . . . fall well short of demonstrating true industry praise.” Bayer Healthcare Pharm.,
18 Inc. v. Watson Pharm., Inc., 713 F.3d 1369, 1377 (Fed. Cir. 2013).
19 Third, Dr. Toth relied on an Amarin-sponsored article in which Dr. Bays said that
20 MARINE’s results were “surprising.” (ECF No. 371 at 1848:11-20 (referring to Ex. 833 at
21 6).) The Federal Circuit has made clear, however, that such “self-referential
22 commendation [also] fall[s] well short of demonstrating true industry praise.” Bayer, 713
23 F.3d at 1377; see also In re Cree, Inc., 818 F.3d 694, 702 (Fed. Cir. 2016) (rejecting
24 patentee’s reliance on “self-serving statements from researchers about their own work” as
25 alleged evidence of praise).
26 In sum, Plaintiffs have not produced evidence that the industry “praised” the
27 claimed invention for avoiding an increase in LDL-C. Thus, the Court finds as a factual 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 44 of 70



1 matter that Plaintiffs’ proffered evidence of praise does not support its nonobviousness
2 arguments discussed in more detail in the Court’s conclusions of law below.
3 IV. CONCLUSIONS OF LAW
4 The Trial focused on induced infringement15 and whether the Asserted Patents are
5 invalid as obvious in light of the prior art. The Court first addresses infringement below,
6 and then obviousness.
7 A. Infringement
8 1. Legal Standard
9 “Infringement is a two-step inquiry, in which a court must first construe disputed
10 claim terms, and then compare the properly construed claims to the accused device.”
11 Nazomi Commc’ns, Inc. v. Arm Holdings, PLC, 403 F.3d 1364, 1367-68 (Fed. Cir. 2005)
12 (citation omitted). The first step as to Plaintiffs’ allegations that Defendants’ proposed
13 products as they will be prescribed infringe the Asserted Claims is already complete—the
14 Court has construed the disputed claim terms. (ECF No. 135.) Plaintiffs bear the burden
15 of persuasion as to infringement and must therefore prove all facts necessary to support
16 their infringement claim. See Medtronic, Inc. v. Mirowski Family Ventures, LLC, 571 U.S.
17 191, 198 (2014) (“It is well established that the burden of proving infringement generally
18 rests upon the patentee.”). Further, “nfringement is a question of fact.” Apple Inc. v.
19 Samsung Elecs. Co., 839 F.3d 1034, 1040 (Fed. Cir. 2016) (citation omitted).
20 In this type of Hatch-Waxman Act patent litigation, where Defendants have filed
21 ANDAs, the question of whether Defendants may be held liable for inducing infringement
22 turns on whether Defendants “have the specific intent, based on the contents of their
23 proposed labels, to encourage physicians to use their proposed ANDA products” in a way
24 that infringes the Asserted Claims. Grunenthal GMBH v. Alkem Labs. Ltd., 919 F.3d 1333,
25 1339 (Fed. Cir. 2019) (citation omitted). In other words, the Court must ask “whether the 26
27 15While Plaintiffs initially asserted two indirect infringement theories, the Court
28 granted summary judgment to Defendants on Plaintiffs’ contributory infringement theory. (ECF No. 278 at 11-13.)

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 45 of 70



1 label encourages, recommends, or promotes infringement.” Id. (citation omitted). And
2 because the Asserted Claims are method claims, the “pertinent question is whether the
3 proposed label instructs users to perform the patented method.” Id. (citation omitted).
4 Plaintiffs have argued at various points in this case that they need only show
5 Defendants’ labels will “inevitably lead some physicians to infringe” to establish
6 Defendants’ inducement liability. (See, e.g., ECF No. 327 at 19 (citing Eli Lilly & Co. v.
7 Teva Parenteral Meds., Inc., 845 F.3d 1357, 1369 (Fed. Cir. 2017).) Defendants counter
8 that labels permitting or even describing an infringing use are insufficient for finding
9 inducement unless those labels “specifically encourage” or “require” infringement. (ECF
10 No. 332 at 17-18.) The Court agrees with Defendants on this point. The fact that some
11 physicians will infringe when they read and follow the labels is necessary, but not sufficient
12 to show inducement based on those labels. See Grunenthal, 919 F.3d at 1339 (finding no
13 inducement where the defendants’ proposed ANDA labels did not “specifically encourage”
14 using the patented drug in an infringing way); HZNP Medicines LLC v. Actavis Labs. UT,
15 Inc., 940 F.3d 680, 702 (Fed. Cir. 2019) (“the mere existence of direct infringement is not
16 sufficient for inducement[,] nstead, our inquiry focuses on whether the instructions reflect
17 an affirmative or specific intent to encourage infringement.”) (internal quotation marks,
18 punctuation, and citation omitted).16 Thus, the Court’s inducement inquiry focuses on
19 Defendants’ proposed labels, specifically whether they encourage, recommend, or
20 promote infringement. See Grunenthal, 919 F.3d at 1339. 21 ///
22 ///
23
24
25
26 16Grunenthal distinguished AstraZeneca LP v. Apotex, Inc., 633 F.3d 1042, 1059- 60 (Fed. Cir. 2010), which Plaintiffs also relied on at Trial in support of an effectively lower
27 inducement burden, because there “the defendant proceeded with a plan to distribute the generic drug knowing that its label posed infringement problems.” Grunenthal, 919 F.3d
28 at 1340. Both in Grunenthal and in this case, the parties relied only on the indications of the proposed labels, making AstraZeneca inapposite. See id.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 46 of 70



1 2. Discussion
2 Though the Court agrees with Defendants’ view of the induced infringement legal
3 standard, it disagrees with Defendants’ application of it. (ECF No. 378 at 12-19 (arguing
4 against Plaintiffs’ induced infringement theory).) To the contrary, the Court finds Plaintiffs
5 carried their burden at Trial to show Defendants’ proposed labels17 will induce infringement
6 of the Asserted Claims.
7 The focal point of the Court’s decision is the Clinical Studies section of the labelling
8 because it provides the only explicit text that addresses each and every disputed element
9 of the Asserted Claims. As Defendants point out, the Court found in ruling on the parties’
10 motions for summary judgment that there was nothing in the labelling that explicitly told
11 doctors to prescribe the drugs in an infringing way. (ECF No. 373 at 142.) But the Court
12 finds—after receiving the benefit of the testimony and evidence presented at Trial—that
13 the Clinical Studies section of the labelling recommends or encourages doctors to
14 prescribe the applicable drug in a way that would, on average, infringe the Asserted
15 Claims.18 Finding otherwise would essentially require finding that doctors would not read
16 the Clinical Studies section of Defendants’ proposed labels. Such a finding would be
17 contrary to medical practice, and contrary to the evidence presented at Trial. Moreover,
18 there is explicit textual support for Plaintiffs’ inducement theory in the Clinical Studies
19 section of the labelling for all Asserted Claims—that a doctor would understand to suggest
20 she should prescribe the drugs in an infringing way.
21 Defendants do not dispute that their proposed labelling will induce infringement of
22 many common elements of the Asserted Claims. (ECF No. 324 at 26-28 (listing several
23 undisputed elements of the Asserted Claims).) Instead, Defendants divide their induced
24 infringement arguments into three parts regarding: (1) the limitation present in all Asserted 25
26 17The Court refers interchangeably to Plaintiffs’ Vascepa labels and Defendants’ proposed labels as they are materially the same for purposes of this analysis.
27 18As explained supra, other sections of the labelling also provide support for the
28 Court’s findings. The Court highlights the Clinical Studies section of the label here because it is pertinent to all Asserted Claims.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 47 of 70



1 claims that the drug must be administered for at least 12 weeks; (2) the limitations present
2 in most Asserted Claims that the drug either reduce TG levels by certain percentages, not
3 increase LDL-C levels, or reduce Apo B levels (the “Other Health Benefits” claims); and
4 (3) the limitations that exclude co-administration of the drug with a with another lipid
5 altering drug such as a statin (the “Excluding a Statin” claims). (ECF No. 378 at 12-19, 32-
6 33, 36-37.) The Court addresses each of these arguments in turn.
7 a) 12 Week Limitation
8 First, the evidence at Trial showed that, based on the proposed labelling,
9 Defendants’ ANDA Products will be prescribed for more than 12 weeks a sufficient
10 percentage of the time for the Court to conclude Defendants will induce infringement of
11 this claim limitation common to all Asserted Claims. A number of factors weigh in favor of
12 this finding. To start, both Plaintiffs’ and Defendants’ experts testified that the indication
13 and usage section of the proposed labels is directed to reducing TG levels below 500
14 mg/dL and then maintaining that reduction—suggesting that the applicable drugs will be 15 prescribed long term. (Compare ECF No. 366 at 331:18-20, 364:19-365:18, 367:11-
16 368:20, 536:22-537:15 (Plaintiffs’ expert Dr. Budoff testifying as such) with ECF No. 367
17 at 672:11-675:2 (Defendants’ expert Dr. Sheinberg conceding he would normally try to
18 reduce TG levels and then maintain that reduction); see also ECF No. 368 at 1210:5-8
19 (Defendants’ expert Dr. Fischer agreeing that, in many patients, “the indication is to reduce
20 below 500 and to maintain that reduction below 500[.]”).) Were a treating physician to stop
21 therapy once TG levels had been reduced below 500, “in most cases [the TG levels] will
22 go back up[.]” (ECF No. 366 at 378:21-379:2; see also 536:22-537:5.) That also supports
23 Plaintiffs’ view that the drug will often be prescribed for long-term treatment. So too do the
24 prescribing practices of experts on both sides, who testified that they generally prescribe
25 either four or twelve months of Vascepa at a time. (ECF Nos. 367 at 391:2-8, 393:10-21, 26 367 at 663:2-19.)
27 Trial testimony further established that severe hypertriglyceridemia generally has a
28 genetic component, meaning that it is usually a chronic condition requiring long-term

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 48 of 70



1 treatment. (ECF No. 366 at 367:23-25, 373:12-389:25 (discussing various trial exhibits
2 that support this view, and offering his own testimony to that effect).) And even
3 Defendant’s expert Dr. Sheinberg agreed that “sometimes severe hypertriglyceridemia is
4 a chronic condition that requires indefinite drug treatment,” even if his estimate of the
5 percentage of chronic cases is lower than that of the other witnesses. (ECF No. 367 at
6 696:16-19.) Thus, there is no real dispute that severe hypertriglyceridemia is a chronic
7 condition requiring long-term treatment at least some of the time. Conversely, there is also
8 no real dispute that severe hypertriglyceridemia can be an acute condition some of the
9 time, where a person experiences, for example, a spike in TG levels above 500 after, say,
10 a bout of binge drinking. (ECF No. 366 at 450:12-15 (“severe hypertriglyceridemia can be
11 an acute phenomenon[.]”).) But overall, the Court finds Plaintiffs’ expert Dr. Budoff’s
12 testimony to the effect that it is generally a chronic condition caused by genetics more
13 persuasive. The Court therefore finds that severe hypertriglyceridemia is generally a
14 chronic condition requiring long-term treatment. Prescribing doctors would bring that
15 understanding to bear when they read Defendants’ proposed labelling lacking an explicit
16 duration of treatment—and most of them would prescribe Defendants’ proposed ANDA
17 Products for more than 12 weeks.
18 Moreover, the Clinical Studies section of Defendants’ proposed labelling points
19 towards the Court’s finding that most doctors would prescribe Defendants’ proposed
20 ANDA Products for more than 12 weeks. Specifically, the Clinical Studies section of
21 Defendants’ labels, like Vascepa’s label, reports the results of the MARINE study, which
22 established the effectiveness of EPA 4 g per day in treating patients with severe
23 hypertriglyceridemia. In describing the important details of the study, this section of the
24 labeling expressly states that patients were administered icosapent ethyl 4 g per day “for
25 12 weeks.” (Ex. 1186 at 11.) And as Defendants’ regulatory expert Mr. Mathers conceded,
26 Defendants’ proposed labeling reports the treatment effects only at 12 weeks, not earlier,
27 and thus reflects approval for reducing TGs below 500 mg/dL and maintaining that
28 reduction through 12 weeks. (Mathers Dep. Tr. 97:2-16.) The fact that the Clinical Studies

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 49 of 70



1 section describes a 12 week trial suggests to prescribing doctors that they should “try to
2 follow the prescribing information, and if the prescribing information was done at 12 weeks,
3 then that informs the physician, that instructs the physician that you should wait 12 weeks
4 to reassess lipids to see what the full effect of your treatment is, because [clinicians’] goal
5 when putting [patients] on Vascepa is to achieve the results in Table 2.” (ECF No. 366 at
6 372:3-12.”) The labels therefore encourage, recommend, promote, or suggest that
7 clinicians should administer Defendants’ ANDA Products for at least 12 weeks to achieve
8 the treatment effects reported in the labeling. (See id. at 372:16-374:5 (“[T]he only way I
9 can compare my patient to the label and what’s being encouraged is to follow the
10 instructions that are given, and the instructions here are to treat for 12 weeks.”).)
11 b) Other Health Benefits Claims
12 Defendants’ narrower noninfringement argument is directed at the Other Health
13 Benefits claims that require the claimed methods either reduce TG levels by certain
14 percentages, not increase LDL-C levels, or reduce Apo B levels. (ECF No. 378 at 36-37.)
15 But the Court finds Defendants’ argument unpersuasive. As discussed above, the Court
16 finds that a doctor would read and understand the Clinical Studies section of the labelling
17 before she prescribed Defendants’ ANDA Products because it is vital to understanding the
18 effects of the applicable drug. (See ECF No. 367 at 665:1-13.) The Clinical Studies section
19 of the labelling describes how the average patient enrolled in the MARINE study received
20 the benefits described in the Other Health Benefits claims. A doctor would read these
21 results as reported in the Clinical Studies section of the labelling as specifically
22 encouraging infringement of the Other Health Benefits Claims.
23 Moving on to focus on the specific claim limitations within the Other Health Benefits
24 Claims, Defendants’ proposed ANDA labels specifically suggest to doctors that their
25 ANDA Products will decrease TG levels without raising LDL-C levels. Not only does the
26 Clinical Studies section report that patients experienced a 5% reduction in LDL-C
27 compared to baseline and a 2% reduction in LDL-C compared to placebo, the Clinical
28 Studies section also states that “[t]he reduction in TG [triglycerides] observed with

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 50 of 70



1 icosapent ethyl was not associated with elevations in LDL-C levels relative to placebo.”
2 (Ex. 1186 at 11; see also ECF No. 366 at 405:5-406:7.) Defendants’ proposed labeling
3 will thus inform prescribers that the drug is safe and effective for administration to patients
4 with severe hypertriglyceridemia to reduce TGs without raising LDL-C. Indeed, Vascepa’s
5 ability to reduce TGs without raising LDL-C, as depicted in the Clinical Studies section, is
6 a primary reason clinicians choose to prescribe Vascepa over other available medications.
7 (ECF No. 366 at 406:7-407:6.) The Clinical Studies section of the labelling therefore
8 suggests to doctors that they can prescribe Defendants’ ANDA Products to lower TG
9 levels without also raising LDL-C levels.19 For these reasons, based on the instructions in
10 Defendants’ proposed labeling, Defendants intend their ANDA Products to be used—and
11 in clinical practice they will be used—“without substantially increasing LDL-C” as required,
12 for example, by Claim 1 of the ’728 patent.
13 Defendants’ proposed ANDA labels also suggest to treating clinicians that they can
14 expect a decrease in Apo B levels when they prescribe Defendants’ ANDA Products.
15 Similar to the analysis above concerning LDL-C, Defendants will induce infringement of
16 the limitations concerning Apo B because clinicians will read Defendants’ labeling as
17 encouraging, recommending, promoting, or suggesting administration of Defendants’
18 ANDA Products to reduce TGs in severely hypertriglyceridemic patients and in conjunction
19 with the TG reduction, “effect a statistically significant reduction . . . in apolipoprotein B.”
20 (ECF No. 366 at 427:9-19; see also ECF No. 369 at 1407:11-15.) Here, too, the Clinical
21 Studies section of the labeling reports the statistically significant decrease in Apo B
22 resulting from administration of Vascepa in Table 2 and then calls out in text below that 23
24 19Moreover, the Warnings and Precautions section in Defendants’ labeling, like the same section in Vascepa’s labeling, omits any warning that patients’ LDL-C levels may
25 rise as a result of treatment. (Ex. 1186 at 2-3.) The absence of a warning would be conspicuous to clinicians because the prescribing information for Lovaza and several
26 fibrates contain such a warning. (ECF No. 366 at 407:17-25.) And physicians who treat patients with severe hypertriglyceridemia would be intimately familiar with the effects of
27 other available drugs (niacin, fibrates, and Lovaza). (ECF No. 367 at 659:11-18.) The lack of a warning about LCL-C increases in Defendants’ labeling is thus a further suggestion
28 to doctors that Defendants’ ANDA Products will decrease TG levels without increasing LDL-C levels.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 51 of 70



1 the drug reduced both median TG and Apo B. (Ex. 1186 at 11; see also ECF No. 366 at
2 427:9-22.) The labeling thus conveys to physicians both the clinical significance of the
3 drugs’ effect on Apo B and the fact that such a reduction will generally occur in their
4 patients in clinical practice. (ECF No. 366 at 427:15-428:5; see also ECF No. 369 at
5 1408:19-22 (testifying that FDA “interpreted this information and it called out that
6 decrease. And so FDA approved this label, it approved this drug for the treatment of
7 hypertriglyceridemia while reducing apo B”); Mathers Dep. Tr. 134:10-22 (stating that the
8 Clinical Studies section of the labeling identifies Apo B among the “relevant parameters to
9 measure on a routine basis and to monitor”). By instructing clinicians that 4 g per day of
10 icosapent ethyl has been shown to cause a statistically significant reduction in TGs and
11 Apo B when administered to adult patients with severe hypertriglyceridemia, the Clinical
12 Studies section of Defendants’ labeling encourages, recommends, promotes, or suggests
13 that clinicians administer Defendants’ ANDA Products with the intent to effect a statistically
14 significant reduction in TGs while having the additional beneficial effect of a statistically
15 significant reduction in Apo B. For these reasons, based on the instructions in Defendants’
16 proposed labeling, Defendants’ intend their ANDA Products to be used—and in clinical
17 practice they will be used—“to effect a statistically significant reduction . . . in
18 apolipoprotein B” as required by Claim 14 of the ’715 patent. (Ex. 22 at 22, Claim 14.)
19 Defendants’ proposed ANDA labels also suggest to doctors that they can expect
20 certain reductions in TG levels by prescribing those ANDA Products, as required by certain
21 other Asserted Claims. Defendants will therefore induce infringement of these limitations
22 because clinicians will read the Clinical Studies section of Defendants’ labeling as
23 encouraging, recommending, promoting, or suggesting administration of Defendants’
24 ANDA Products to achieve, on average, the percentage TG reductions described in certain
25 Asserted Claims. Table 2 in the Clinical Studies section of Defendants’ proposed labeling,
26 like the same table in Vascepa’s labeling, reports that, when administered for 12 weeks to
27 patients with severe hypertriglyceridemia, EPA 4 g per day caused a median 27%
28 reduction in triglycerides from baseline and a median 33% reduction in triglycerides

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 52 of 70



1 compared to placebo. (Ex. 1186 at 11; see also ECF No. 366 at 433:23-434:3.) For these
2 reasons, based on the instructions in Defendants’ proposed labeling, Defendants intend
3 their ANDA Products to be used—and in clinical practice they will be used—to reduce TG
4 levels by the percentages required by Claim 4 of the ’560 Patent and Claim 17 of the ’560 5 Patent. (ECF No. 366 at 433:16-435:2, 435:6-436:20.)
6 c) Excluding a Statin Claims
7 Defendants’ narrowest noninfringement argument is directed at the Excluding a
8 Statin claims. (ECF No. 378 at 32-33.) The Court is also unpersuaded by this argument.
9 To the contrary, the labels of Defendants’ proposed ANDA Products suggest to a doctor
10 that the drugs could be used with or without a statin or other lipid-lowering drug.
11 The Excluding a Statin limitation requires administration of the claimed
12 pharmaceutical composition to a patient “who does not receive concurrent lipid altering
13 therapy.” (Ex. 21 at 21-22 Claims 1,16; see also Ex. 22 at 22, Claim 14 (“who does not
14 receive a concurrent lipid altering therapy”).) The Court construed the term “concurrent
15 lipid altering therapy” to mean “a medication to alter lipid levels in a subject whereby the
16 medication is administered concurrently / concomitantly with the administration of a
17 pharmaceutical composition comprising ethyl eicosapentaenoate.” (ECF No. 135 at 5-7.)
18 Statins are an example of a “medication to alter lipid levels.” (ECF No. 366 at 412:1-6,
19 414:1-20 (identifying statins as concurrent lipid altering therapies).) Based on the Court’s
20 construction, a clinician who administers Defendants’ ANDA Products to a patient who is
21 not on another lipid altering medication (e.g., a statin) will directly infringe this limitation.
22 There is text in several places on Defendants’ proposed labelling that would
23 suggest to doctors Defendants’ proposed ANDA Products could be administered without
24 a concurrent lipid altering therapy. First, the Indications and Usage section does not
25 contain any instructions that Defendants’ ANDA Products must be administered with a
26 lipid-altering drug, though FDA regulations would have required instructions to that effect
27 were that the case. (ECF No. 366 at 410:11-25 (testifying that the label does not require
28 concurrent lipid-altering therapy); Ex. 573 at 7, 12 (stating that coadministration should be

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 53 of 70



1 listed were it a requirement).) Second, and similarly, the Dosage and Administration
2 section of the labelling would have had to mention it, but did not. (Ex. 572 at 8 (stating any
3 concomitant medications should be listed in this section); see also Ex. 1186 at 2 (labelling,
4 which does not include such a restriction); ECF No. 369 at 1355:3-6 (explaining that the
5 labelling does not mention such a restriction).) Third, the Clinical Studies section of the
6 labelling indicates that only 25% of the MARINE study participants were on a concomitant
7 lipid-altering therapy. (Ex. 1186 at 11.) Clinicians appreciate from this clinical study
8 description that the remaining 75% of patients in the study described in the Clinical Studies
9 section were not on concurrent lipid altering therapy (e.g., statins). (ECF No. 369 at
10 1413:8-18; see also Mathers Dep. Tr. at 68:1-5, 68:7-15.) For these reasons, based on
11 the instructions in Defendants’ proposed labeling, Defendants intend their ANDA Products
12 to be used—and in clinical practice will be used—by patients who “do[] not receive
13 concurrent lipid altering therapy” as required by certain claims of the Asserted Patents.
14 (ECF No. 366 at 409:7-415:11 (discussing the monotherapy limitation of the ’728 patent).)
15 The Court therefore finds that the labels of Defendants’ proposed ANDA Products
16 encourage, recommend, promote, or suggest that clinicians prescribe those products in a
17 way that infringes all of the Asserted Claims.
18 Defendants’ arguments to the contrary are unavailing. First, as Defendants
19 continue to argue that their proposed ANDA Products’ substantial noninfringing uses
20 should change the Court’s analysis in various ways (ECF No. 378 at 12-13), the Court
21 reiterates that “contributory infringement can turn on whether there are substantial non-
22 infringing uses, while inducement does not.” (ECF No. 278 at 8.) See also Sanofi v.
23 Watson Labs. Inc., 875 F.3d 636, 646 (Fed. Cir. 2017) (“[T]here is no legal or logical basis
24 for the suggested limitation on inducement.”). Second, and relatedly, Defendants argue
25 that induced infringement cannot be inferred under these circumstances—that inducement
26 cannot be found without specific instructions in the label. (ECF No. 378 at 12.) But the
27 Court has done no such thing. The Court is not inferring infringement without looking at
28 the content of the label. Rather, and as explained above, the Court is reading primarily the

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 54 of 70



1 Clinical Studies section of the label as trial testimony established a doctor would read it.
2 For that same reason, the caselaw Defendants rely on, Grunenthal and Horizon, is
3 distinguishable. (ECF No. 378 at 14.) Unlike in those cases, there is support in the text of
4 Defendants’ proposed ANDA labels for the plausible interpretation of those labels,
5 supported by expert testimony, that the Court finds encourages infringement here. Third,
6 to the extent the Court has not made it clear above, the Court finds the evidence presented
7 at Trial shows that severe hypertriglyceridemia is a chronic condition necessitating
8 indefinite treatment most of the time, or at least enough of the time for the Court to properly
9 find inducement here. Thus, the Court rejects Defendants’ argument that they do not
10 infringe the 12 week limitation of the Asserted Claims because severe
11 hypertriglyceridemia is not a chronic condition. (ECF No. 378 at 8.)
12 In sum, the Court finds that Defendants’ labelling will induce infringement of all
13 Asserted Claims. However, as further explained below, the Court also finds that All
14 Asserted claims are invalid as obvious in light of the prior art.
15 B. Obviousness
16 1. Legal Standards
17 Under 35 U.S.C. § 103, a patent is invalid as obvious “if the differences between
18 the claimed invention and the prior art are such that the claimed invention as a whole
19 would have been obvious before the effective filing date of the claimed invention to a
20 person having ordinary skill in the art to which the claimed invention pertains.” Whether a
21 patent claim is obvious is ultimately a question of law based on four underlying factual
22 determinations: (1) “the scope and content of the prior art”; (2) “the level of ordinary skill
23 in the pertinent art”; (3) the “differences between the prior art and the claims at issue”; and
24 (4) “[s]uch secondary considerations as commercial success, long-felt but unsolved
25 needs, [and the] failure of others ” Graham, 383 U.S. at 17.
26 “A party seeking to invalidate a patent based on obviousness must demonstrate ‘by
27 clear and convincing evidence that a skilled artisan would have been motivated to combine
28 the teachings of the prior art references to achieve the claimed invention, and that the

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 55 of 70



1 skilled artisan would have had a reasonable expectation of success in doing so.’” Procter
2 & Gamble Co. v. Teva Pharm. USA, Inc., 566 F.3d 989, 994 (Fed. Cir. 2009) (quoting
3 Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1361 (Fed. Cir. 2007)). Defendants, as the
4 accused infringers, bear the ultimate burden of proving, by clear and convincing evidence,
5 that the Asserted Claims are invalid. See Microsoft Corp. v. i4i Ltd. P’ship, 564 U.S. 91,
6 95 (2011). That said, where “the PTO did not have all material facts before it, its considered
7 judgment may lose significant force,” and courts should “consider that fact when
8 determining whether an invalidity defense has been proved by clear and convincing
9 evidence.” Id. at 111; see also Syntex (U.S.A.) LLC v. Apotex, Inc., 407 F.3d 1371, 1379
10 (Fed. Cir. 2005) (finding reversible error where “district court failed to appreciate that the
11 prosecution history of the relevant patents, while not establishing inequitable conduct,
12 casts some doubt on the final examiner’s conclusion that the claimed [invention] produces
13 unexpected results sufficient to overcome a prima facie case of obviousness.”).
14 a) Motivation to Combine
15 Federal Circuit “case law does not require that a particular combination must be the
16 preferred, or the most desirable, combination described in the prior art in order to provide
17 motivation for the current invention.” In re Fulton, 391 F.3d 1195, 1200 (Fed. Cir. 2004)
18 (internal quotation omitted). “The question is whether there is something in the prior art as
19 a whole to suggest the desirability, and thus the obviousness, of making the combination,
20 not whether there is something in the prior art as a whole to suggest that the combination
21 is the most desirable combination available.” Id. (citation omitted). “[T]here is no
22 requirement that the prior art contain an express suggestion to combine known elements
23 to achieve the claimed invention.” Motorola, Inc. v. Interdigital Tech. Corp., 121 F.3d 1461, 24 1472 (Fed. Cir. 1997).
25 b) Reasonable Expectation of Success
26 For the reasonable expectation of success component, although the definition is
27 “somewhat vague, [Federal Circuit] case law makes clear that it does not require a
28 certainty of success.” Medichem, SA v. Rolabo, SL, 437 F.3d 1157, 1165 (Fed. Cir. 2006).

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 56 of 70



1 “Conclusive proof of efficacy is not necessary to show obviousness. All that is required is
2 a reasonable expectation of success.” Hoffmann-La Roche Inc. v. Apotex Inc., 748 F.3d
3 1326, 1331 (Fed. Cir. 2014) (citation omitted). Difficulties in receiving FDA approval “are
4 not particularly probative with respect to obviousness” because “[t]here is no requirement
5 that one of ordinary skill have a reasonable expectation of success in developing” the FDA
6 approved drug. Allergan, Inc. v. Sandoz Inc., 726 F.3d 1286, 1292 (Fed. Cir. 2013).
7 Rather, “the person of ordinary skill need only have a reasonable expectation of success
8 of developing the claimed invention.” Id.
9 c) Secondary Considerations
10 Part of the obviousness inquiry also considers whether objective indicia of non-
11 obviousness support the Asserted Claims. “Such secondary considerations as commercial
12 success, long felt but unsolved needs, failure of others, etc., might be utilized to give light
13 to the circumstances surrounding the origin of the subject matter sought to be patented.”
14 Graham, 383 U.S. at 17-18; see also In re Rouffet, 149 F.3d 1350, 1355 (Fed. Cir. 1998)
15 (explaining that objective evidence of nonobviousness may include copying, long felt but
16 unsolved need, failure of others, commercial success, unexpected results created by the
17 claimed invention, unexpected properties of the claimed invention, licenses showing
18 industry respect for the invention, and skepticism of skilled artisans). “Secondary
19 considerations help inoculate the obviousness analysis against hindsight.” ZUP, LLC v.
20 Nash Mfg., Inc., 896 F.3d 1365, 1373 (Fed. Cir. 2018) (quotation omitted). However, “a
21 strong showing of obviousness may stand even in the face of considerable evidence of
22 secondary considerations.” Id. at 1374 (quotation omitted).
23 2. Discussion
24 The Court first discusses Defendants’ prima facie obviousness case, which the
25 Court finds Defendants supported with clear and convincing evidence of obviousness at
26 Trial, and then discusses each of Plaintiffs’ proffered objective indicia of nonobviousness.
27 The Court will go on to explain why the Court does not find that Plaintiffs’ proffered
28 evidence of secondary considerations saves the Asserted Claims.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 57 of 70



1 a) Prima Facie Obviousness
2 As an initial matter, the Court is persuaded that Defendants presented clear and
3 convincing evidence at Trial that all Asserted Claims are invalid as obvious. The heart of
4 Defendants’ persuasive obviousness argument is that the Lovaza PDR covers many of
5 the limitations of the Asserted Claims, and making the obvious substitution of only EPA
6 instead of a mixture of EPA and DHA renders most limitations of the Asserted Claims
7 obvious. The result of this obvious substitution, obtained by combining the Lovaza PDR
8 and Mori, is the method recited in all Asserted Claims.
9 Although Plaintiffs dispute that the claimed method was obvious, they concede a
10 number of Defendants’ key premises. For instance, there is no dispute that the only
11 difference between the method in the Lovaza PDR and the method in the asserted claims
12 is that Lovaza contained a mixture of EPA and DHA, instead of purified EPA. (ECF No.
13 367 at 762:6-14; see also ECF No. 371 at 1821:5-1823:1.) Nor is there any dispute that
14 the increases in LDL-C caused by Lovaza were known, and that “a skilled artisan would
15 have been motivated to avoid LDL-C increases when treating patients with severe
16 hypertriglyceridemia.” (ECF No. 371 at 1822:8-11.) Moreover, while “many patients who
17 took Lovaza were also given a statin to address the LDL-C increases,” Plaintiffs’ expert
18 Dr. Toth agreed that since “those patients would have to take two pills, the Lovaza and a
19 statin,” “a skilled artisan would have been motivated to develop a single pill that treats
20 severe hypertriglyceridemia without LDL-C increases.” (Id. at 1822:12-21; see also ECF 21 No. 367 at 813:8-814:2.)
22 Further, the Court finds that a skilled artisan would have wanted to know which
23 active ingredient in Lovaza—EPA or DHA—was responsible for the LDL-C increase (if not
24 both), and that Mori addressed this exact issue. Indeed, Dr. Toth did not dispute that “a
25 skilled artisan seeing that there’s DHA and EPA in Lovaza, and seeing a side effect, would
26 at least consider whether the side effect could be associated with only DHA or only EPA.”
27 (ECF No. 371 at 1787:6-10.) Nor did he dispute that “Mori found that the increase of LDL-
28 C with DHA was statistically significant and the increase with EPA was not.” (Id. at

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 58 of 70



1 1788:18-25.) While Dr. Toth disputed other aspects of Defendants’ obviousness defense
2 (addressed further below), the key premises that he conceded lead directly to the
3 motivation to combine and reasonable expectation of success that Defendants have
4 asserted.
5 In addition to the claimed method of treatment, and as discussed above as to
6 infringement, all but one asserted claim (claim 1 of the '929 patent) requires certain effects
7 on a patient’s lipids—a minimum reduction in triglycerides (e.g., at least about 20%); no
8 increase in LDL-C; or a reduction in Apo B (again, these are the Other Health Benefits
9 Claims). As discussed in the findings of fact above, the prior art showed that purified EPA
10 produced each of the claimed effects in clinical studies. In particular, Mori and Hayashi
11 disclosed that EPA reduced triglycerides by at least about 20%; Mori, Hayashi, and
12 Kurabayashi disclosed that EPA did not increase LDL-C; and Kurabayashi disclosed that
13 EPA reduced Apo B.
14 One asserted claim (claim 16 of the '728 patent) further requires that the EPA
15 product used to treat the patient contains no more than 0.6% of any other fatty acid. There
16 is no dispute that this level of purity was disclosed and rendered obvious at least by WO
17 '900,20 which taught a process for producing “99.9% EPA” with “less than 0.1% of DHA.” 18 (Ex. 1525 at 17.)
19 Critically, in view of the claim language, obviousness is proven as long as there
20 was a reasonable expectation that 4 g/day of 96% purified EPA would achieve the claimed
21 effects (i.e., not cause an LDL-C increase) in patients with triglycerides of exactly 500
22 mg/dL. “It is a long-established rule that claims which are broad enough to read on obvious
23 subject matter are unpatentable even though they also read on nonobvious subject
24 matter.” In re Cuozzo Speed Techs., LLC, 793 F.3d 1268, 1281 (Fed. Cir. 2015) (quotation
25 omitted). Thus, to prove obviousness, Defendants do not need to prove that a skilled 26
27

28 9.)

20The parties stipulated to the fact that this reference is prior art. (ECF No. 324 at

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 59 of 70



1 artisan would have reasonably expected success in achieving the claimed effects in
2 patients with triglycerides above 500 mg/dL, much less substantially above that level.
3 Also, this case is unlike many other obviousness cases because, when the Patent
4 Office issued the patents-in-suit, it maintained its finding from earlier rejections that the
5 prior art rendered all of the claims prima facie obvious. (Ex. 1521 at 1822-35, see also id.
6 at 1830-31.) As the examiner explained, “it was concluded that it will be obvious to treat
7 patients having triglycerides above 500 mg/dL with 96% pure ethyl-EPA.” (Id. at 1830.)
8 The examiner thus agreed with Defendants’ view that the prior art would have motivated
9 a skilled artisan to practice the asserted claims with a reasonable expectation of success
10 (issuing the patents based solely on secondary considerations). (ECF No. 371 at 1804:22-
11 1806:1; see also ECF No. 331 at 152 (noting in Plaintiffs’ proposed findings of fact that
12 “the Examiner concluded that it would be prima facie obvious to treat patients having TG
13 above 500 mg/dl with 96% pure ethyl-EPA”).)
14 The Court therefore finds that Defendants established by clear and convincing
15 evidence at Trial that all Asserted Claims are prima facie obvious. Plaintiffs arguments to
16 the contrary are unavailing. Many of Plaintiffs’ arguments depend on the premise that
17 POSAs as of March 2008 would not have expected that using a composition of purified
18 EPA would not increase LCL-C levels. (ECF No. 379 at 22-23.) But this premise is not
19 supported by the evidence. To explain, Plaintiffs primarily point to testimony from Dr. Toth
20 to support this premise. But there are at least three issues with Dr. Toth’s testimony. First,
21 he agreed under questioning that, as of “March 2008 [. . .] the prior art reflect[ed] that all
22 these treatments increased LDL-C in patients with very high triglycerides.” (ECF No. 370
23 at 1574:1-1575:1.) But that cannot be correct, because Mori taught that EPA did not
24 increase LDL-C levels like DHA did. (Ex. 1538 at 3.) Second, Dr. Toth testified that von
25 Schacky contributed to his view that all TG-lowering therapies increase LDL-C levels.
26 (ECF No. 370 at 1697:9-1703:7.) But as Defendants point out (ECF No. 378 at 26), von
27 Schacky did not correctly summarize Mori. Specifically, von Schacky, citing Mori, wrote,
28 “In more recent comparative studies, no effects of either EPA or DHA were seen on total

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 60 of 70



1 cholesterol, HDL, or LDL levels.” (Ex. 1605 at 5.) But even Dr. Toth agreed on cross-
2 examination that is not what Mori said. (ECF No. 371 at 1847:8-17.) Mori actually found
3 that LDL-C increased with DHA, but not EPA. (Ex. 1538 at 3.) Third, part of Dr. Toth’s
4 opinion, and Plaintiffs’ argument, is based on the Carlson reference from 1977. (ECF No.
5 377 at 43-44 (citing ECF No. 370 at 1577:22-25 and Ex. 1026.).) The Court is unpersuaded
6 that an article from 1977 reflects the knowledge of a POSA in 2008. Thus, Plaintiffs’
7 argument, in part based on Dr. Toth’s testimony—that a POSA would have thought that
8 both DHA and EPA would cause an increase in LDL-C in March 2008—lacks evidentiary
9 support. The Court accordingly rejects this argument.
10 Moreover, Plaintiffs’ arguments also depend on another factual premise that lacks
11 evidentiary support—that patients with TG levels above 500 mg/dL respond differently to
12 TG-lowering therapy than patients with TG levels below 500 mg/dL. (ECF No. 379 at 23-
13 24.) But even if Mori and other studies on patients with lower TGs did not provide
14 “conclusive proof” of EPA’s effects, they were enough to form “a reasonable expectation
15 of success.” Hoffmann-La Roche, 748 F.3d at 1331. Indeed, Dr. Toth conceded that
16 POSAs could rely on data in patients with triglycerides below 500 mg/dL to make
17 reasonable predictions about how patients above that threshold would respond. As he
18 admitted, “a skilled artisan would know that a drug that reduces triglycerides in a patient
19 at 400, is very likely to also reduce triglycerides in a patient at 600.” (ECF No. 371 at
20 1860:8-11.) Thus, the Court finds that a POSA “would have reasonably expected purified
21 EPA to reduce triglyceride levels above 500,” even without data confirming that result. (Id. 22 at 1860:12-15.)
23 There was no reason to expect differently for LDL-C. Dr. Toth cited no evidence
24 that the 500 mg/dL threshold reflects any difference in how patients metabolize drugs, or
25 any relationship between that specific threshold and LDL-C. As he admitted, “[t]he 500
26 threshold was not set because above 500 you are expected to have a greater increase in
27 LDL-C in response to a drug.” (Id. at 1860:3-7.) Instead, all experts agreed that the
28 threshold simply represents a marker for the risk of pancreatitis, which has nothing to do

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 61 of 70



1 with LDL-C levels. (ECF No. 371 at 1859:3-13; see also Bays Dep. Tr. at 143:9-11, 143:13-
2 19.) In Dr. Heinecke’s words, there is no “magical mechanistic difference” between having
3 triglycerides of 400, 500, or 600 mg/dL. (ECF No. 367 at 796:5-20.) A skilled artisan would
4 understand that, regardless of a patient’s baseline triglycerides, “the qualitative effects of
5 medications . . . tend to be the same.” (Id. at 797:16-18.)
6 Finally, Plaintiffs try to discredit Mori by pointing to von Schacky. (ECF No. 379 at
7 24.) But the Court credits Mori over von Schacky, because, as described above, von
8 Schacky incorrectly summarized Mori, and is therefore not credible. In sum, having found
9 that Defendants met their clear and convincing burden to prove their prima facie
10 obviousness case at trial, the Court turns to consideration of Plaintiffs’ proffered secondary
11 considerations.
12 b) Secondary Considerations
13 “[E]vidence rising out of the so-called ‘secondary considerations’ must always when
14 present be considered en route to a determination of obviousness.” Stratoflex, Inc. v.
15 Aeroquip Corp., 713 F.2d 1530, 1538 (Fed. Cir. 1983). The Court therefore addresses
16 each of the secondary considerations proffered by Plaintiffs. Plaintiffs specifically point to
17 unexpected benefits, satisfaction of long-felt but unmet need, skepticism, praise, and
18 commercial success. (ECF No. 377 at 10.) But before the Court addresses each of these
19 secondary considerations, the Court addresses Defendants’ challenge to the nexus
20 between the REDUCE-IT clinical trial results and the Asserted Claims—which the Court
21 finds persuasive.
22 i. REDUCE-IT
23 Plaintiffs rely on the results of the REDUCE-IT clinical trial to support several of
24 their secondary considerations arguments. (ECF No. 379 at 35-38.) However, Defendants
25 counter that, as a matter of law, the Court should not consider the results of the REDUCE-
26 IT study in analyzing Plaintiffs’ proffered secondary considerations because REDUCE-IT
27 lacks a sufficient nexus to the Asserted Claims. (ECF No. 378 at 30-32.) The Court agrees
28 with Defendants.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 62 of 70



1
2 Regardless of whether a presumption of nexus applies here,21 there is no nexus
3 between REDUCE-IT and the Asserted Claims. “It is the established rule that objective
4 evidence of non-obviousness must be commensurate in scope with the claims which the
5 evidence is offered to support.” Allergan, 754 F.3d at 965 (quotation omitted; reversing
6 judgment of nonobviousness). “Where the offered secondary consideration actually
7 results from something other than what is both claimed and novel in the claim, there is no
8 nexus to the merits of the claimed invention.” In re Huai-Hung Kao, 639 F.3d 1057, 1068
9 (Fed. Cir. 2011) (emphasis omitted). For multiple reasons, Plaintiffs’ evidence regarding
10 REDUCE-IT does not satisfy these requirements.
11 First, REDUCE-IT lacks a nexus to the claimed use of Vascepa without a statin. As
12 Dr. Toth admitted, “none [of] the asserted claims require a statin.” (ECF No. 371 at
13 1896:23-24.) In fact, three claims expressly require treating a patient “who does not
14 receive concurrent lipid altering therapy,” and thus preclude using a statin. (Ex. 1500 (’728
15 patent claims 1 and 16); Ex. 1502 (’715 patent claim 14).) In contrast, “all the patients in
16 REDUCE-IT were taking statins”—“100 percent.” (ECF No. 371 at 1896:15-19; see also
17 Ex. 1641 at 2.) In fact, there is no dispute that a statin must be administered to reduce
18 cardiovascular risk with Vascepa. As Dr. Toth testified, “it would have been unethical to
19 have just a Vascepa monotherapy arm [in REDUCE-IT]. The FDA would never allow it
20 because statin therapy is the standard of care.” (ECF No. 371 at 1897:5-10.) This is
21 reflected in the REDUCE-IT indication, which makes clear that Vascepa reduces
22 cardiovascular risk only “as an adjunct to maximally tolerated statin therapy.” (Ex. 2248 at 23 2.)
24 The REDUCE-IT results are therefore not “commensurate in scope with the claims.”
25 Allergan, 754 F.3d at 965. For the three claims that exclude statins, the benefits of 26
27 21The parties dispute whether a presumption of nexus applies (ECF Nos. 378 at 30-31, 379 at 35), but the Court need not—and does not—resolve that dispute because
28 the Court finds, as explained infra, that there is an insufficient nexus between REDUCE- IT and the Asserted Claims.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 63 of 70



1 REDUCE-IT are entirely outside the scope of the claims. But even for the claims that are
2 silent on statin use, there is no dispute that Vascepa can be, and often is, used without a
3 statin in accordance with the claimed method. As Dr. Toth agreed, only “25 percent of the
4 patients in MARINE were taking statins.” (ECF No. 371 at 1896:20-22.) At most, therefore,
5 the REDUCE-IT results could only be relevant to that subset of patients. But the Asserted
6 Claims are much broader—they include the 75% of patients in MARINE who took Vascepa
7 without a statin. Because the REDUCE-IT results are “not commensurate with the full
8 scope of the patent’s claims,” they “lack[] a nexus with the scope of the [asserted] patent[s’]
9 claimed invention.” Allergan, 754 F.3d at 965.
10 Put differently, the benefits in REDUCE-IT “actually result[ed] from something other
11 than” the claimed invention, which at least allows using Vascepa without a statin. In re
12 Huai-Hung Kao, 639 F.3d at 1068. Instead, the benefits resulted from a different
13 invention—one claimed in Plaintiffs’ unasserted patents—which requires using a statin.
14 (Ex. 2001 at 1, 52-53.) REDUCE-IT thus lacks a nexus to the Asserted Claims. (ECF No. 15 367 at 821:2-18.)
16 Second, REDUCE-IT lacks a nexus to the claimed use of EPA to reduce
17 triglycerides. As Dr. Toth conceded, “none of the patent claims at issue in this case have
18 a limitation with regard to reducing cardiovascular risk.” (ECF No. 371 at 1894:15-18.)
19 Instead, all asserted claims are directed to “[a] method of reducing triglycerides.” The
20 benefits in REDUCE-IT, however, were unrelated to reducing triglycerides. According to
21 the REDUCE-IT publication (the Bhatt Article), “the significantly lower risk of major adverse
22 cardiovascular events with icosapent ethyl than with placebo appeared to occur
23 irrespective of the attained triglyceride level at 1 year (≥150 or suggest that at least some
24 of the effect of icosapent ethyl that resulted in a lower risk of ischemic events than that
25 with placebo may be explained by metabolic effects other than a reduction of triglyceride
26 levels.” (Ex. 1641 at 10.) In other words, the REDUCE-IT benefits “actually result[ed] from
27 something other than” the claimed method of reducing triglycerides, which precludes any
28 finding of nexus. In re Huai-Hung Kao, 639 F.3d at 1068. (See also ECF Nos. 367 at 816:8-

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 64 of 70



1 817:12, 368 at 1035:4-1037:2.) On cross-examination, Plaintiffs argued that “the Bhatt
2 [A]rticle doesn’t rule out TG lowering as responsible for at least part of the CV benefit.”
3 (ECF No. 368 at 1119:11-14.) But on the contrary, the evidence of record, including the
4 Bhatt Article, suggests the opposite. Thus, there is no basis to conclude that the REDUCE-
5 IT results have a nexus to the claimed method of reducing triglycerides.
6 Third, REDUCE-IT lacks a nexus to avoiding an increase in LDL-C, which is a
7 limitation of all but two Asserted Claims, and is the purported discovery that allegedly
8 distinguishes the Asserted Claims from the prior art. According to the Bhatt Article, the
9 REDUCE-IT investigators “found no substantial difference in the benefit of icosapent ethyl
10 as compared with placebo with respect to the primary end point according to whether the
11 patients who received placebo had an increase in LDL cholesterol levels at 1 year or had
12 no change or a decrease in LDL cholesterol levels.” (Ex. 1641 at 7.) Thus, the REDUCE-
13 IT benefits “actually result[ed] from something other than” the claimed method of avoiding
14 an increase in LDL-C, as required by eight of the asserted claims. (ECF No. 367 at 820:13-
15 821:1. See also In re Huai-Hung Kao, 639 F.3d at 1068.
16 Fourth, the REDUCE-IT results are not commensurate in scope with the Asserted
17 Claims because the results were limited to patients with multiple cardiovascular risk factors
18 that the asserted claims do not require. As explained in the Bhatt Article, REDUCE-IT was
19 limited to patients who “were 45 years of age or older and had established cardiovascular
20 disease or were 50 years of age or older and had diabetes mellitus and at least one
21 additional risk factor.” (Ex. 1641 at 2.) Likewise, the REDUCE-IT indication is limited to
22 patients with “established cardiovascular disease or diabetes mellitus and 2 or more
23 additional risk factors for cardiovascular disease.” (Ex. 2248 at 2.) By contrast, the
24 Asserted Claims do not contain any of these limitations. As Dr. Toth admitted, “aside from
25 severe high triglycerides, there’s no other risk factor[] required by the patents related to
26 cardiovascular issues.” (ECF No. 371 at 1894:22-25.) For example, none of the Asserted
27 Claims are limited to patients with diabetes. (ECF Nos. 367 at 826:10-12, 368 at 1093:21-
28 22.) Moreover, there is no dispute that many patients with severe hypertriglyceridemia do

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 65 of 70



1 not have risk factors such as diabetes. For example, in MARINE, only 28% of patients
2 were diabetic. (Ex.1741 at 2; see also ECF No. 367 at 825:22-826:9.) The Asserted Claims
3 cover the treatment of the remaining patients who were not diabetic, as well as patients
4 who more generally do not have two or more cardiovascular risk factors. Because the
5 REDUCE-IT results are limited to patients with such risk factors, they are “not
6 commensurate with the full scope of the patent’s claims.” Allergan, 754 F.3d at 965.
7 Fifth, REDUCE-IT lacks a nexus to the limitation in all Asserted Claims that patients
8 must have TG levels of at least 500 mg/dL. As Dr. Toth admitted, “REDUCE-IT focused
9 on patients with triglycerides below 500.” (ECF No. 371 at 1894:12-14.) According to the
10 Bhatt Article, “[e]ligible patients had a fasting triglyceride level of 150 to 499 mg per
11 deciliter,” which means that patients with triglyceride of at least 500 mg/dL were not eligible
12 to participate. (Ex. 1641 at 2.) The benefits in REDUCE-IT thus “actually result[ed] from
13 something other than” the claimed invention, which is limited to treating patients with
14 triglycerides of at least 500 mg/dL, so “there is no nexus[.]” (ECF No. 367 at 818:12-
15 819:16.) See also In re Huai-Hung Kao, 639 F.3d at 1068. Indeed, because REDUCE-IT
16 focused on patients with triglycerides below 500 mg/dL, conducting REDUCE-IT did not
17 even infringe the Asserted Claims. Moreover, in analogous circumstances, the Federal
18 Circuit has held that evidence regarding products that are not covered by the asserted
19 claims cannot be relevant to secondary considerations. The same principle applies to the
20 method claims here—because the Asserted Claims do not cover the REDUCE-IT study,
21 evidence regarding REDUCE-IT is irrelevant. See Ashland Oil, Inc. v. Delta Resins &
22 Refractories, Inc., 776 F.2d 281, 306 n.42 (Fed. Cir. 1985) (stating if “products were not
23 covered by the [asserted] patents, [ ] then the secondary considerations [based on those
24 products] would not have had any relevance to the obviousness/nonobviousness
25 determination”); Amazon.com, Inc. v. Barnesandnoble.com, Inc., 239 F.3d 1343, 1366
26 (Fed. Cir. 2001) (holding that secondary considerations based on “copying Amazon’s ‘1-
27 Click®’ feature is legally irrelevant unless the ‘1-Click®’ feature is shown to be an
28 embodiment of the claims”).

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 66 of 70



1 Plaintiffs argue that some patients in REDUCE-IT developed higher triglyceride
2 levels after they became eligible for the study, and thus the study did include a handful of
3 patients with triglycerides of at least 500 mg/dL. (ECF No. 379 at 35 n.10.) But Plaintiffs’
4 argument contradicts their position that Defendants’ prior-art references are not relevant
5 unless all patients in the study had triglycerides of at least 500 mg/dL. Plaintiffs cannot
6 have it both ways. If studies in which no patients, or only a handful of patients, had
7 triglycerides of at least 500 mg/dL are irrelevant, then so is REDUCE-IT.
8 In sum, for multiple independent reasons, the REDUCE-IT results are not
9 commensurate in scope with, and did not actually result from practicing, any of the
10 Asserted Claims. Thus, there is an insufficient nexus between REDUCE-IT and the
11 Asserted Claims. As a result, evidence concerning REDUCE-IT is not relevant to
12 determining whether the Asserted Claims are invalid as obvious.
13 ii. Unexpected Benefits
14 Plaintiffs also argue that the positive lipid effects recited in the Other Health Benefit
15 claims are unexpected benefits that constitute another secondary consideration weighing
16 in favor of nonobviousness. (ECF No. 377 at 252-257.) Defendants counter that these
17 benefits were not unexpected because they were predicted by the relevant prior art. (ECF
18 No. 378 at 29.) The Court agrees with Defendants.
19 As explained above as to Defendants’ prima facie obviousness case, Mori found
20 that EPA did not raise LDL-C levels, and Kurabayashi suggested that EPA reduced Apo
21 B levels. (ECF No. 373 at 76-80, 246-47.) Further, while the Patent Office found that a
22 decrease in Apo B was an unexpected benefit constituting a valid secondary
23 consideration, the Patent Office’s examiner did not consider Kurabayashi. (Id. at 246-47.)
24 Where “the PTO did not have all material facts before it, its considered judgment may lose
25 significant force[.]” See i4i, 564 U.S. at 95. Thus, the Court finds that the unexpected
26 benefits secondary consideration does not weigh in favor of finding the Asserted Claims
27 nonobvious. 28 ///

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 67 of 70



1 iii. Satisfaction of Long-Felt Need
2 Plaintiffs also argue that the Asserted Claims are not obvious because Vascepa
3 satisfied long-felt needs—“as it is the first approved treatment that reduces TGs without
4 raising LDL-C in patients with severe hypertriglyceridemia, and the first treatment for
5 reducing TGs in severely hypertriglyceridemic patients that reduces cardiovascular risk on
6 top of statin.” (ECF No. 377 at 261.) Defendants counter that there was no long-felt need
7 to reduce TGs without raising LDL-C because a patient could also be put on a statin to
8 avoid the LDL-C increase. (ECF No. 378 at 29-30.) The Court agrees with Plaintiffs.
9 The Court is persuaded that there was a long-felt need for a drug like Vascepa that
10 could reduce TG levels without raising LDL-C levels, primarily because both sides’ experts
11 testified that patients are more likely to comply with a prescribed treatment regime when
12 they only have to take one pill, rather than two—and the Court relied on this evidence in
13 finding a POSA would be motivated to combine the Lovaza PDR with the finding from Mori
14 that EPA did not raise LDL-C levels.22 (See supra Section IV.B.2(a).) It is better to take
15 one pill than two if taking that one pill will give you all the same benefit. Moreover, there is
16 no real dispute that some patients may not be able to tolerate statins. (ECF No. 367 at
17 660-61.) Thus, the Asserted Claims represent an improvement—albeit a prima facie
18 obvious one—over the prior art. And this secondary consideration therefore weighs slightly
19 in favor of finding the Asserted Claims nonobvious.
20 iv. Skepticism
21 Skepticism about an invention is evidence that an invention was not obvious. See
22 In re Rouffet, 149 F.3d 1350, 1355 (Fed. Cir. 1998). Plaintiffs argue that this secondary
23 consideration weighs in their favor because experts were skeptical that Vascepa could 24
25

26 22However, the Court notes that the Court does not credit the REDUCE-IT
27 Indication as weighing in Plaintiffs’ favor as to this factor because the Court has already found REDUCE-IT lacks the required nexus to the Asserted Claims supra in Section
28 IV.B.2(b).i.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 68 of 70



1 lower TG levels without also raising LDL-C levels.23 (ECF No. 377 at 268.) Defendants
2 counter that Plaintiffs did not present any expert testimony at Trial regarding skepticism,
3 and only cite to the opinions of two experts retained by Plaintiffs to serve on an expert
4 panel during Vascepa’s development—and their opinions are irrelevant because Plaintiffs
5 did not present any evidence these experts were aware of the prior art Defendants relied
6 on in this case. (ECF No. 378 at 30.) The Court agrees with Defendants.
7 Plaintiffs’ proffered evidence of skepticism is not inconsistent with Defendants’
8 argument. Specifically, Plaintiffs point to notes taken by Ian Osterloh at Plaintiffs’ expert
9 meeting earlier on in the development of Vascepa and related deposition testimony, and
10 specifically point to this note: “LDL-C is likely to go up as it does with virtually all tg-lowering
11 therapies in this group of patients.” (ECF No. 377 at 268 (citing Ex. 754 at 2).) But of
12 course, the phrase ‘virtually all’ does not mean ‘all,’ and the Court agrees with Defendants
13 that this view does not appear to account for Mori. And a skeptical statement is entitled to
14 less weight if, as appears to be the case here, the person who made the statement was
15 unaware of relevant prior art that would likely have made them less skeptical. See
16 PharmaStem Therapeutics, Inc. v. ViaCell, Inc., 491 F.3d 1342, 1365 (Fed. Cir. 2007)
17 (discounting testimony expressing surprise where “there was no indication that either [the
18 declarant] or members of his research group were previously aware of the prior art
19 references that laid the groundwork for the inventors' experiments.”). In sum, the Court
20 finds that the skepticism secondary consideration does not weigh in favor of finding the
21 Asserted Claims nonobvious.
22 v. Praise
23 The Court found, as a factual matter supra in Section III.G.4(c), that Plaintiffs’
24 proffered evidence of praise for Vascepa was more qualified and equivocal than Plaintiffs 25
26
27 23Plaintiffs also make skepticism arguments based on the REDUCE-IT Indication (ECF No. 377 at 268-69), but the Court does not consider those arguments because
28 REDUCE-IT lacks the required nexus to the Asserted Claims, as explained supra in Section IV.B.2(b).i.

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 69 of 70



1 argued, and thus finds that the praise secondary consideration does not weigh in favor of
2 finding the Asserted Claims nonobvious.
3 vi. Commercial Success
4 But the Court also found, as a factual matter supra in Section III.G.4(b), that
5 Vascepa is a commercial success. This secondary consideration therefore weighs in favor
6 of finding the Asserted Claims nonobvious.
7 vii. Weighing These Secondary Considerations
8 The Court thus finds that the satisfaction of long-felt need and commercial success
9 secondary considerations weigh in Plaintiffs’ favor, and the remaining secondary
10 considerations weigh in Defendants’ favor. More specifically, the Court finds that Vascepa
11 is a commercial success even though it has not yet turned a profit, and that there was long
12 felt need for a single pill that reduced TG levels without increasing LDL-C levels. However,
13 these secondary considerations are outweighed by the fact that the Court found Plaintiffs’
14 other proffered secondary considerations favor Defendants. Thus, at best, Plaintiffs have
15 presented weak evidence of the existence of secondary considerations, which do not
16 overcome the Court’s finding that all Asserted Claims are prima facie obvious. See, e.g.,
17 ZUP, 896 F.3d at 1373 (holding that “a strong showing of obviousness may stand even in
18 the face of considerable evidence of secondary considerations”).
19 For the reasons discussed above, in view of all four Graham factors (including
20 alleged secondary considerations), Defendants have proven by clear and convincing
21 evidence that all Asserted Claims are invalid as obvious under 35 U.S.C. § 103.
22 C. Remedies
23 Plaintiffs seek a permanent injunction that Defendants be prohibited from marketing
24 their proposed ANDA Products until Plaintiffs’ Asserted Patents expire, and that their
25 ANDA applications similarly should not be made effective until Plaintiffs Asserted Patents
26 expire. (ECF No. 377 at 300-01.) However, Plaintiffs are not entitled to these remedies
27 because, while the Court found that Defendants’ proposed ANDA Products will induce 28

Case 2:16-cv-02525-MMD-NJK Document 381 Filed 03/30/20 Page 70 of 70



1 infringement of the Asserted Claims, all of the Asserted Claims are invalid as obvious 2 under 35 U.S.C. § 103.
3 V. CONCLUSION
4 The Court notes that the parties made arguments and cited to cases not discussed
5 above. The Court has reviewed these arguments and cases, and has determined they do
6 not materially affect the outcome of this case.
7 The Court finds that Defendants’ proposed ANDA Products will induce infringement
8 of the Asserted Claims, but all the Asserted Claims are invalid as obvious under 35 U.S.C.
9 § 103. Thus, the Court finds in favor of Defendants on Plaintiff’s remaining infringement
10 claim, and in their favor on their counterclaims asserting the invalidity of the Asserted
11 Claims under 35 U.S.C. § 103.
12 The Clerk of Court is ordered to enter judgment in favor of Defendants on Plaintiffs’
13 claim and on Defendants’ counterclaims, and close this case.
14 DATED THIS 30th day of March 2020. 15
16
MIRANDA M. DU
17 CHIEF UNITED STATES DISTRICT JUDGE
18
19
20
21
22
23
24
25
26
27
28