Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 1 of 19
`
`15 Civ. 3588 (PAE)
`
`ECF Case
`
`UNITED STATES DISTRICT COURT
`SOUTHERN DISTRICT OF NEW YORK
`AMARIN PHARMA, INC., DR.
`JONATHAN HERBST, DR. ERIC RISHE,
`DR. PETER GOTTESFELD, and DR.
`RALPH YOUNG,
`
`Plaintiffs,
`
`v.
`
`UNITED STATES FOOD & DRUG
`ADMINISTRATION, UNITED STATES OF
`AMERICA, STEPHEN OSTROFF, M.D., in
`his official capacity as Acting Commissioner
`of Food and Drugs, and SYLVIA
`MATHEWS BURWELL, in her official
`capacity as Secretary of the Department of
`Health & Human Services,
`
`Defendants.
`
`DECLARATION OF CURTIS ROSEBRAUGH
`
`I, Curtis Rosebraugh, M.D., M.P.H., hereby declare under penalty of perjury, pursuant
`
`to 28 U.S.C. § 1746, that the following is true and correct to the best of my knowledge,
`
`information, and belief:
`
`1. I am the Director of the Office of Drug Evaluation II, Office of New Drugs, Center for
`
`Drug Evaluation and Research (CDER), United States Food and Drug Administration (FDA or
`
`the Agency). I am a board-certified Internist, and I have been with the Agency for
`
`approximately fifteen years. For the last ten years, I have served as Deputy Director, Acting
`
`Director, and Director of the Office of Drug Evaluation II, in which the Division of Metabolism
`
`and Endocrinology Products resides. That Division is responsible for reviewing and approving,
`
`among other things, new drug applications for drugs intended for the prevention and treatment of
`
`conditions relating to hyperlipidemia, which refers to elevated levels of lipids in the blood.
`
`1
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 1 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 2 of 19
`
`Drugs that fall into this class include Trilipix, Niaspan, Lovaza, and Vascepa. I have also held
`
`the positions of Deputy Director of Over-the-Counter Drug Products and Senior Medical
`
`Reviewer in the Division of Pulmonary and Allergy Drug Products in FDA’s CDER. I received
`
`my medical degree from the University of Kansas School of Medicine, my master’s degree in
`
`public health from The Johns Hopkins School of Public Health, and my undergraduate degree in
`
`pharmacy from the University of Kansas School of Pharmacy. I have held teaching
`
`appointments at University of Texas Medical Branch and University of Kansas Medical Center.
`
`2. In these capacities, I am familiar with the steps FDA has taken to ensure that the
`
`labeling for triglyceride-lowering drug products reflects the most accurate and up-to-date
`
`scientific information regarding the relationship between drug-induced lowering of triglyceride
`
`levels and reducing the risk of cardiovascular events in patients on statin therapy. More
`
`specifically, I am familiar with FDA’s decisions to remove indications related to statin co-
`
`administration from the labeling of Trilipix (fenofibric acid) and Niaspan (niacin extended-
`
`release) and to remove data and information about a triglyceride-lowering trial in statin-treated
`
`patients from the Clinical Studies section of the labeling for Lovaza (omega-3 acid ethyl esters).
`
`I am also familiar with the regulatory history for Vascepa (omega-3 acid eicosapentaenoic acid),
`
`which is accurately set forth in the June 5 letter from FDA to Amarin Pharma, Inc. (“Amarin”).
`
`3. In this declaration, I describe the approvals for Trilipix, Niaspan, Lovaza, and
`
`Vascepa, certain cardiovascular outcomes trials that are relevant to the labeling and approvals for
`
`these products, and steps FDA has taken and continues to take to ensure that healthcare
`
`professionals have information reflecting the best and most current scientific data about the lack
`
`of evidence to support the conclusion that decreasing triglyceride levels with a drug further
`
`reduces the risk of cardiovascular events among patients on statin therapy.
`
`2
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 2 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 3 of 19
`
`Approvals for Trilipix, Niaspan, Lovaza, and Vascepa
`
`4. Trilipix is a fenofibrate-based drug. FDA first approved Trilipix on December 15,
`
`2008, for several indications, including the following indication for co-administration with a
`
`statin: “Trilipix is indicated as an adjunct to diet in combination with a statin to reduce TG
`
`[triglycerides] and increase HDL-C [high-density lipoprotein cholesterol] in patients with mixed
`
`dyslipidemia and CHD [coronary heart disease] or a CHD risk equivalent who are on optimal
`
`statin therapy to achieve their LDL-C [low-density lipoprotein cholesterol] goal.” 2013 Trilipix
`
`Labeling, attached hereto as Exhibit 1.
`
`5. Niaspan is an extended-release formulation of niacin. FDA initially approved Niaspan
`
`on July 28, 1997, for five indications. In 2003, FDA approved a supplemental new drug
`
`application (“sNDA”), adding an indication for the use of Niaspan in combination with lovastatin
`
`for the treatment of primary hypercholesterolemia and mixed dyslipidemia. See 2003 Niaspan
`
`Letter, attached hereto as Exhibit 2. In 2009, FDA approved an sNDA revising the indication
`
`related to statin co-administration to include mention of simvastatin as well. As of March 2015,
`
`this indication read as follows: “NIASPAN in combination with simvastatin or lovastatin is
`
`indicated for the treatment of primary hyperlipidemia and mixed dyslipidemia when treatment with
`
`NIASPAN, simvastatin, or lovastatin monotherapy is considered inadequate.” 2013 Niaspan
`
`Labeling, attached hereto as Exhibit 3.
`
`6. Lovaza is composed of omega-3-acid ethyl esters, and contains mostly the ethyl esters
`
`of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). In 2004, FDA approved
`
`Lovaza (originally known as Omacor) as an adjunct to diet to reduce very high (cid:11)(cid:149)(cid:3)(cid:24)(cid:19)(cid:19)(cid:3)(cid:80)(cid:74)(cid:18)(cid:71)(cid:47)(cid:12)(cid:3)
`
`triglyceride levels in adult patients. Subsequently, Lovaza’s sponsor conducted a trial in
`
`simvastatin-treated patients with triglyceride levels between 200 and 499 mg/dL, and
`
`well-controlled low-density lipoprotein cholesterol (“LDL-C”) levels, to investigate the effect of
`3
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 3 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 4 of 19
`
`Lovaza on lipid measurements such as non-HDL-cholesterol and triglycerides, after 8 weeks of
`
`Lovaza treatment. FDA determined that the data on reductions in triglyceride levels from the
`
`trial were not sufficient to support the approval of an indication for the reduction of
`
`non-HDL-cholesterol, triglycerides, and other lipid parameters in this population, but FDA
`
`approved labeling that summarized data and information about the trial in the Clinical Studies
`
`section. See 2007 Lovaza Letter (approving Lovaza’s proposed labeling), attached hereto as
`
`Exhibit 4.
`
`7. Vascepa is a purified ester of EPA derived from fish oil. See June 5 Letter and
`
`citations therein. FDA approved Vascepa in July 2012 as a drug to be used as an adjunct to diet
`
`to reduce triglyceride levels in adult patients with severe hypertriglyceridemia, defined as
`
`trigylceride levels (cid:149)(cid:24)(cid:19)(cid:19)(cid:3)(cid:80)(cid:74)(cid:18)(cid:71)(cid:47)(cid:3)(cid:11)(cid:179)(cid:89)(cid:72)(cid:85)(cid:92)(cid:3)(cid:75)(cid:76)(cid:74)(cid:75)(cid:3)trigylceride levels”). The primary rationale for
`
`treating individuals with very high trigylceride levels is to reduce the risk of pancreatitis.
`
`Pursuant to a Special Protocol Assessment (“SPA”) agreement with FDA, Amarin also
`
`conducted the “ANCHOR trial” to assess the effect of Vascepa on trigylceride levels in statin-
`
`treated patients with well-controlled LDL-C levels whose trigylceride levels remained high. In
`
`this context, changes in trigylceride levels were being used as a surrogate to predict lowering the
`
`risk of cardiovascular events. In February 2013, Amarin then sought FDA approval to market
`
`Vascepa for another use, namely to treat patients with trigylceride levels between 200 mg/dL and
`
`499 mg/dL (“high trigylceride levels”) who are already being treated with statins to lower
`
`cholesterol. The primary rationale for treating statin-treated patients with this range of
`
`trigylceride levels with a second drug is to further reduce the risk of cardiovascular events, such
`
`as cardiovascular morbidity or mortality, resulting from atherosclerotic cardiovascular disease.
`
`4
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 4 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 5 of 19
`
`ACCORD-Lipid Trial
`
`8. In March 2010, the results from the ACCORD-Lipid trial were published online in the
`
`New England Journal of Medicine, attached hereto as Exhibit 5. The ACCORD-Lipid trial
`
`evaluated the effectiveness of fenofibrate. Specifically, the ACCORD-Lipid trial was designed
`
`to answer the following question: In the context of good glycemic control, does a therapeutic
`
`strategy that uses a fibrate to increase HDL-C and lower triglyceride levels together with a statin
`
`to lower LDL-C reduce the rate of cardiovascular disease events compared with a strategy that
`
`uses a statin and a placebo?
`
`9. Although there were favorable changes in lipids, including reductions in triglyceride
`
`levels, the ACCORD-Lipid trial failed to demonstrate a statistically significant reduction in
`
`major adverse cardiovascular events among individuals treated with fenofibrate and simvastatin
`
`compared with those treated with simvastatin alone. The active ingredient in Trilipix is the
`
`active metabolite of fenofibrate.
`
`Trilipix and ACCORD-Lipid Advisory Committee
`
`10. FDA convened an advisory committee on May 19, 2011, to discuss the results of the
`
`ACCORD-Lipid trial and their implications regarding the Trilipix labeling. The committee
`
`considered whether FDA should allow continued marketing of Trilipix’s indication for co-
`
`administration with a statin without revision of the labeling, withdraw approval of the indication,
`
`or allow continued marketing of the indication with revision of the labeling to incorporate the
`
`principal findings from the ACCORD-Lipid trial. Three members voted to allow continued
`
`marketing of the indication without revision to the labeling; four members voted to withdraw
`
`approval of the indication for co-administration with a statin; and six members voted to allow
`
`continued marketing with a statin, but to revise the labeling to incorporate the principal findings
`
`5
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 5 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 6 of 19
`
`from ACCORD-Lipid. See 2011 Summary Minutes, attached hereto as Exhibit 6.
`
`Trilipix Labeling Changes and Postmarketing Clinical Trial
`
`11. In July 2011, based on the results of the ACCORD-Lipid trial, FDA notified the
`
`Trilipix sponsor that another postmarketing clinical trial would be required to evaluate the effect
`
`of Trilipix on the incidence of major adverse cardiovascular events in high-risk men and women
`
`at LDL-C goal on statin therapy, but with residually high triglycerides and low HDL-C. See
`
`2011 AbbVie Letter, attached hereto as Exhibit 7. FDA also required changes to the Trilipix
`
`labeling. Under the heading “Important Limitations of Use,” the following language was added:
`
`“Fenofibrate at a dose equivalent to 135 mg of Trilipix was not shown to reduce coronary heart
`
`disease morbidity and mortality in 2 large, randomized controlled trials of patients with type 2
`
`diabetes mellitus.” The ACCORD-Lipid trial was one of the two trials referenced in this
`
`statement. See 2011 Trilipix Labeling, attached hereto as Exhibit 8. In addition, a description of
`
`the ACCORD-Lipid trial was added to the Warnings and Precautions section of labeling
`
`(“Mortality and Coronary Heart Disease Morbidity” subsection), and the Medication Guide was
`
`updated. See Exhibit 8.
`
`12. FDA’s decisions to require a postmarketing clinical trial and the labeling changes
`
`described above, rather than to remove the indication for statin co-administration, were informed
`
`by the majority opinion of the advisory committee, which did not recommend removal at that
`
`time. FDA’s decisions were also informed by the fact that two other trials briefly described
`
`below (the AIM-HIGH trial and the HPS2-THRIVE trial) were well underway at the time and
`
`were expected to further inform the effect on cardiovascular outcomes of adding a second lipid-
`
`altering drug (specifically, trigylceride-lowering/HDL-C-raising drugs) to statin therapy.
`
`6
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 6 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 7 of 19
`
`AIM-HIGH Trial
`
`13. In December 2011, the results of the AIM-HIGH trial were published in the New
`
`England Journal of Medicine, attached hereto as Exhibit 9. The AIM-HIGH trial tested the
`
`effectiveness of extended-release niacin in patients on simvastatin therapy. Specifically, the trial
`
`was designed to test the hypothesis that niacin added to optimal statin therapy will reduce the
`
`risk of cardiovascular events compared with statins alone in patients with atherosclerotic
`
`cardiovascular disease and atherogenic dyslipidemia.
`
`14. The AIM-HIGH trial was terminated earlier than expected because a formal interim
`
`analysis demonstrated a lack of efficacy. The AIM-HIGH trial failed to demonstrate a
`
`cardiovascular benefit of adding extended-release niacin to simvastatin therapy in patients with
`
`atherosclerotic cardiovascular disease and atherogenic dyslipidemia. FDA did not pursue
`
`removal of the statin co-administration indication for Niaspan at that time because the results of
`
`the HPS2-THRIVE trial were forthcoming. As discussed further below, the HPS2-THRIVE trial
`
`also studied the cardiovascular effects of a niacin-containing product in statin-treated patients,
`
`but in a much larger population than the AIM-HIGH trial. The HPS2-THRIVE trial had the
`
`potential to provide valuable information to inform not only the potential benefits of niacin, but
`
`also the clinical benefits, if any, of modulating lipid/lipoprotein biomarkers other than LDL-C
`
`among statin-treated patients.
`
`Niaspan Labeling Changes
`
`15. Following review of the AIM-HIGH data, FDA updated the labeling for Niaspan.
`
`Results from the AIM-HIGH trial were added to the Warnings and Precautions and Adverse
`
`Reactions sections of labeling, and a new Limitation of Use was added. The letter approving the
`
`prior approval sNDA with the package insert is attached hereto as Exhibit 10. Specifically, the
`
`7
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 7 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 8 of 19
`
`following statement in the Limitations of Use was added to the Indications and Usage section of
`
`the Niaspan labeling:
`
`NIASPAN, at doses of 1,500-2,000 mg/day, in combination with simvastatin, did not
`reduce the incidence of cardiovascular events more than simvastatin in a randomized
`controlled trial of patients with cardiovascular disease and mean baseline LDL-C levels
`of 74 mg per deciliter [see Warnings and Precautions (5.1)].
`
`Exhibit 10. FDA considered the addition of the AIM-HIGH results to Niaspan’s labeling to be
`
`adequate to inform physicians about the trial results, and FDA planned to evaluate whether
`
`additional labeling changes were necessary once the results from the HPS2-THRIVE trial were
`
`made available.
`
`HPS2-THRIVE Trial
`
`16. The results of the HPS2-THRIVE trial were presented publicly on March 9, 2013, at
`
`a meeting of the American College of Cardiology and then published in the New England
`
`Journal of Medicine on July 17, 2014, attached hereto as Exhibit 11. The HPS2-THRIVE trial
`
`was designed to assess the effects of adding an extended-release niacin formulation, in
`
`combination with the anti-flushing drug laropiprant, to an effective statin-based LDL-C-lowering
`
`treatment for high-risk patients with prior vascular disease.
`
`17. The HPS2-THRIVE trial was expected to be a more definitive trial than the AIM-
`
`HIGH trial. The population of the HPS2-THRIVE trial was 25,673 patients, compared with the
`
`3,414 patients in the AIM-HIGH trial. In general, in randomized controlled trials, having more
`
`patients in a trial makes it more likely that a difference between treatment groups (e.g., drug
`
`versus placebo) will be detected, if a difference truly exists. Like the AIM-HIGH trial, the
`
`HPS2-THRIVE trial failed to demonstrate clinical benefit from the addition of an extended-
`
`release niacin-containing product to effective LDL-C-lowering statin therapy among high-risk
`
`patients with prior cardiovascular disease.
`
`8
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 8 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 9 of 19
`
`The Vascepa Advisory Committee Meeting, the Rescission of the ANCHOR SPA
`Agreement, and FDA’s Decision Not to Approve Amarin’s sNDA
`
`18. During the same timeframe in which the data from the cardiovascular outcomes trials
`
`became available, Amarin conducted and completed the ANCHOR trial. The ANCHOR trial
`
`was an adequate and well-controlled clinical trial designed to determine whether Vascepa lowers
`
`triglyceride levels in statin-treated patients with well-controlled LDL-C levels and high
`
`triglyceride levels. The results demonstrated a statistically significant reduction in triglyceride
`
`levels in the Vascepa groups compared with the placebo (mineral oil) groups. Based on the
`
`ANCHOR trial results, and following 50% enrollment of patients into Amarin’s cardiovascular
`
`outcomes trial involving Vascepa (“the REDUCE-IT trial”), Amarin submitted a sNDA seeking
`
`approval for Vascepa as an adjunct to diet and in combination with a statin to reduce TG, non-
`
`HDL-C, Apo B (apolipoprotein B), LDL-C, TC (total cholesterol), and VLDL-C (very low-
`
`density lipoprotein cholesterol) in adult patients with mixed dyslipidemia and coronary heart
`
`disease or a coronary heart disease risk equivalent.
`
`19. On October 16, 2013, FDA consulted with an advisory committee to obtain outside
`
`expert advice regarding the experts’ level of confidence that the ANCHOR results would
`
`translate into a reduction in cardiovascular risk among the target population. FDA also asked the
`
`committee whether Vascepa’s effects in the target population were sufficient to grant approval
`
`for co-administration with statin therapy for treatment of patients similar to the ANCHOR
`
`population, prior to the availability of results from the REDUCE-IT trial. The REDUCE-IT trial
`
`is expected to provide evidence for whether or not adding Vascepa to statin therapy further
`
`reduces the risk for cardiovascular events among patients with high triglyceride levels. Among
`
`other things, the committee discussed the results of the ANCHOR trial and the results of the
`
`three aforementioned cardiovascular outcomes trials and their potential impact on Amarin’s
`
`9
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 9 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 10 of 19
`
`pending sNDA for Vascepa. As noted in the summary minutes of the meeting, “panel members
`
`stated that the available data from recent clinical trials do not strongly support an expected
`
`cardiovascular outcome benefit from lipid changes similar to those observed with Vascepa
`
`treatment.” 2013 Summary Minutes, attached hereto as Exhibit 12. There was substantial
`
`discussion that the rationale for use of Vascepa in this patient population was to impart a
`
`cardiovascular benefit, and the panel expressed their considered opinion that “there is uncertainty
`
`regarding the clinical benefits of the observed lipid changes.” Exhibit 12. The committee voted
`
`9 “no” and 2 “yes” on the question of whether Vascepa should be approved for this indication
`
`prior to the completion of the REDUCE-IT trial.
`
`20. Following the advisory committee meeting, FDA’s review division rescinded the
`
`ANCHOR SPA agreement. FDA’s review division concluded that it no longer had sufficient
`
`confidence that a change in triglyceride levels is sufficient to establish the effectiveness of a drug
`
`intended to reduce cardiovascular risk in statin-treated subjects with high trigylceride levels.
`
`FDA’s review division denied Amarin’s request for reconsideration, and the rescission decision
`
`was upheld during two subsequent levels of formal dispute resolution on the grounds that (1) no
`
`adequate and well-controlled trial has demonstrated a cardiovascular benefit resulting from
`
`drug-induced lowering of triglyceride levels in statin-treated patients, and (2) three recent
`
`clinical trials failed to show additional cardiovascular benefit of adding a non-statin drug to statin
`
`therapy, even though each drug had lowered triglyceride levels significantly in statin-treated
`
`patients.
`
`21. Rescission of a SPA agreement is a rare occurrence. Of the approximately 1,000
`
`SPA agreements entered into by FDA and sponsors of investigational new drug applications over
`
`the last seven years, only ten have been rescinded. The rarity of SPA agreement rescission
`
`10
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 10 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 11 of 19
`
`indicates that FDA does not take the rescission process lightly or without due consideration.
`
`FDA determined that it could not approve Amarin’s sNDA in its current form because there are
`
`insufficient data to support lowering triglyceride levels as a surrogate for reducing cardiovascular
`
`risk in statin-treated patients with well-controlled LDL-C levels and high triglyceride levels.
`
`FDA advised Amarin that to obtain approval, it would need to provide evidence that Vascepa
`
`reduces the risk of major adverse cardiovascular events in patients at high risk for CVD, with
`
`high triglyceride levels and well-controlled LDL-C levels on statin therapy. FDA also told
`
`Amarin that it anticipated that the final results from the REDUCE-IT trial could be submitted to
`
`satisfy this deficiency. Amarin has committed to continuing the REDUCE-IT trial and has stated
`
`that it expects that the trial will be completed in 2017, with results expected to be available in
`
`2018. See Compl. ¶ 67.
`
`FDA’s Removal of the Niaspan and Trilipix Statin Co-Administration Indications
`
`22. The ACCORD-Lipid, AIM-HIGH, and HPS2-THRIVE trials all failed to
`
`demonstrate incremental cardiovascular benefit of a second lipid-altering drug (fenofibrate or
`
`formulations of niacin) when added to statin-treated patients with well-controlled LDL-C,
`
`despite favorable effects on biomarkers of cardiovascular risk (e.g., triglyceride levels and HDL-
`
`C levels). Based on the totality of these new data, FDA determined that the indications related to
`
`co-administration with statins should be removed from the labeling of Niaspan and Trilipix on
`
`the basis that the recent cardiovascular outcomes trials failed to support the previously held
`
`belief that the putatively favorable changes in lipid biomarkers induced by these drugs would
`
`lead to a reduction in cardiovascular disease risk in statin-treated patients. Accordingly, FDA
`
`revised the labeling for Niaspan and Trilipix to remove the indications related to statin co-
`
`administration. Consistent with the removal of this indication from the Trilipix labeling, FDA
`
`11
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 11 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 12 of 19
`
`also released the Trilipix sponsor from the requirement to conduct a post-marketing
`
`cardiovascular outcomes trial to evaluate the risk of major adverse cardiovascular events in
`
`patients treated with Trilipix when co-administered with statin therapy; such a trial had not been
`
`required of other fenofibrate sponsors, because no other fenofibrate product has an indication for
`
`co-administration with a statin. Copies of the current labeling for Niaspan and Trilipix are
`
`attached hereto as Exhibits 13–14.
`
`23. FDA’s revision of the labeling for these products to remove the indications for co-
`
`administration with a statin was based on the evolving nature of the available scientific data. At
`
`the time the statin co-administration indications for Niaspan and Trilipix were approved,
`
`available scientific evidence suggested that favorable changes in lipid parameters would translate
`
`to a reduction in cardiovascular risk. The statin co-administration indications for Niaspan and
`
`Trilipix were approved based on data demonstrating statistically significant changes in lipid
`
`parameters such as TG, HDL-C, and non-HDL-C beyond those achieved by taking a statin alone.
`
`But in all three cardiovascular outcomes trials discussed above, further lowering of triglycerides
`
`(as well as other putatively favorable lipid changes) failed to further reduce cardiovascular risk in
`
`patients already treated with a statin. Although elevated triglyceride levels are associated with
`
`adverse cardiovascular outcomes, a drug-induced change in a risk factor does not always result
`
`in the expected effect on clinical outcomes, and the best scientific data currently available have
`
`failed to demonstrate that drug-induced triglyceride lowering (with either Niaspan, an extended-
`
`release niacin, or fenofibrate) results in a reduction in cardiovascular risk in patients who are
`
`already on optimal statin therapy. FDA thus determined that the statin co-administration
`
`indications should be removed from the labeling for Niaspan and Trilipix.
`
`12
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 12 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 13 of 19
`
`FDA’s Removal of Statin Co-Administration Data from the Lovaza Labeling
`
`24. As mentioned above, FDA permitted inclusion of data and information in the
`
`Clinical Studies section of Lovaza’s labeling about a trial that studied the effects of Lovaza on
`
`lipid measurements (such as triglycerides and non-HDL-C) in simvastatin-treated patients with
`
`triglyceride levels between 200 and 499 mg/dL, and well-controlled LDL-C levels. An
`
`indication was not granted based on this trial, however, because FDA determined that the
`
`reductions in triglyceride and non-HDL-C levels were not sufficient to support the approval of an
`
`indication. FDA later concluded that, in the absence of an approved indication for treatment of
`
`patients with high triglyceride levels despite statin therapy, it was not appropriate to continue to
`
`include data about the co-administration of Lovaza with a statin in this patient population in the
`
`Clinical Studies section of the Lovaza labeling. This conclusion was based on several reasons.
`
`First, the inclusion of non-indication-specific data and information other than for safety related
`
`reasons in the Clinical Studies section of the drug labeling is inconsistent with FDA’s regulation,
`
`guidance, and policies. See, e.g., 21 C.F.R.§ 201.57. Second, in patients on statin therapy, the
`
`published results of ACCORD-LIPID, AIM-HIGH, and HPS2-THRIVE raised substantial doubt
`
`regarding the clinical benefit of reducing triglyceride levels with a second lipid-altering drug
`
`intended for the purpose of reducing cardiovascular disease risk. Consequently, FDA no longer
`
`believes that the totality of the scientific evidence supports a conclusion that a drug-induced
`
`reduction in triglyceride levels in statin-treated patients results in a reduction in the risk of
`
`cardiovascular events. Third, with respect to Lovaza, its NDA sponsor had not carried out any
`
`cardiovascular outcomes trials to establish such a benefit. For all these reasons, FDA asked the
`
`sponsor of the NDA for Lovaza to remove the content related to this trial from the Clinical
`
`Studies section, and revised labeling was approved shortly thereafter. See 2014 GSK Letter and
`
`13
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 13 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 14 of 19
`
`2014 Lovaza Labeling, attached hereto as Exhibits 15 and 16, respectively.
`
`The JELIS Trial and Its Limitations
`
`25. I understand that Plaintiffs allege that the JELIS trial results support a cardiovascular
`
`benefit from EPA therapy in the studied population. It is important to exercise caution in
`
`drawing such a conclusion from the published results of the JELIS trial. The results of the JELIS
`
`trial were published in the Lancet in March 2007, attached hereto as Exhibit 17. To give a brief
`
`summary of the JELIS trial, Japanese adults with elevated cholesterol levels with or without
`
`coronary artery disease were administered either EPA with a statin or statin alone. The primary
`
`endpoint of the trial was a major coronary event, defined as sudden cardiac death, fatal or
`
`nonfatal myocardial infarction, unstable angina, and coronary bypass surgery or angioplasty.
`
`There was a 5 percentage point difference between the groups in the relative changes in
`
`triglyceride levels from baseline to the last clinic visit. The 5-year cumulative rate of major
`
`coronary events was 2.9 percent in the EPA plus statin group and 3.5 percent in the statin-alone
`
`group, resulting in a statistically significant relative risk reduction of 19 percent. The breakdown
`
`of the individual components of the primary endpoint shows the strongest evidence of the
`
`treatment effect with EPA on unstable angina.
`
`26. As part of its presentation to the advisory committee during the October 16, 2013,
`
`meeting, as well as when considering whether to rescind the SPA agreement for the ANCHOR
`
`trial, FDA reviewed the published results of the JELIS trial. FDA identified limitations to the
`
`design of the JELIS trial that affect the interpretation of the trial’s results. First, the subjects in
`
`the JELIS trial were limited to Japanese adults receiving a low dose of statin therapy that may be
`
`considered inadequate in the United States. Second, the JELIS trial was an open-label trial. In
`
`such a trial, both researchers and participants know whether a participant is being administered
`
`14
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 14 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 15 of 19
`
`the drug or placebo. Having this knowledge can influence physician and patient behavior, such
`
`as the reporting of symptoms. Third, the main component of the primary endpoint in the JELIS
`
`trial was unstable angina, which is a more subjective endpoint than, for example, objective major
`
`adverse cardiovascular event endpoints (e.g., heart attack, stroke, or cardiovascular death). A
`
`subjective endpoint such as unstable angina may be particularly unreliable in an open-label trial
`
`where patients and physicians are making decisions regarding hospitalizations.
`
`27. In addition to these design limitations, patients treated with EPA and statin achieved
`
`triglyceride levels that were only 5 percent lower, on average, than those achieved among the
`
`patients treated with statin alone; however, the reduction in cardiovascular risk was 19 percent.
`
`The large difference in magnitude between the triglyceride reduction and risk reduction suggests
`
`that the effects of EPA on triglycerides may not be responsible for, or predict, the observed
`
`differences in cardiovascular events between treatment groups in this trial. For all these reasons,
`
`CDER’s review division determined, when considering whether to rescind the ANCHOR SPA
`
`agreement, that the results from the JELIS trial could not be used as support for or against the
`
`use of triglyceride levels as a surrogate for cardiovascular risk reduction. This conclusion was
`
`upheld during the dispute resolution process regarding the rescission of the ANCHOR SPA
`
`agreement.
`
`PCSK9 Inhibitors Advisory Committee
`
`28. FDA continues to base its approval decisions on the best and most current data
`
`regarding the relationship between lipid-lowering drugs and cardiovascular risk. For example,
`
`for more than the last two decades, FDA has used a reduction in LDL-C as a surrogate for
`
`cardiovascular risk reduction for several lipid-altering drugs to support traditional approval.
`
`Certainly, at least for statins, the validity of a reduction in LDL-C as a surrogate for reduced
`
`15
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1035, p. 15 of 19
`
`

`

`Case 1:15-cv-03588-PAE Document 54 Filed 06/23/15 Page 16 of 19
`
`cardiovascular risk has been confirmed through numerous rando