PROPIEDADES FARMACOLÓGICAS

Omacor contiene ésteres etílicos de ácidos Omega-3 90,principalmente en forma de ácido eicosapentaenoico(EPA) y el ácido docosahexaenoico (DHA), código ATC:C10AX06.

Omacor actúa sobre los lípidos plasmáticos reduciendo elnivel de los triglicéridos como resultado del descenso delcolesterol VLDL (lipoproteínas de muy baja densidad) ytambién actúa sobre la homeostasia y la presión arterial.

Omacor reduce la síntesis hepática de triglicéridos puestoque el EPA y el DHA son malos substratos para lasenzimas responsables de la síntesis de triglicéridos einhiben la esterificación de otros ácidos grasos.

El aumento de la beta-oxidación de ácidos grasos en losperoxisomas del hígado también contribuye al descensode los triglicéridos, reduciendo la cantidad de ácidosgrasos libres disponibles para su síntesis. La inhibición deesta síntesis disminuye el VLDL.

COMPOSICIÓN

Cada cápsula blanda contiene: Ésteres etílicos 90 deácidos Omega-3, 1000 mg. Conteniendo 840 mg en formade: (ácido eicosapentaenoico (EPA) etil éster, 460 mg) y

(ácido docosahexaenoico (DHA) etil éster, 380 mg).

Antioxidante: alfa-Tocoferol …… 4 mg

INDICACIONES TERAPÉUTICAS

Hipertrigliceridemia

En la hipertrigliceridemia endógena, como suplemento a ladieta, cuando las medidas dietéticas por sí solas resultaninsuficientes para generar una respuesta adecuada: tipo IVen monoterapia y tipo IIb/III en combinación con estatinas,cuando el control de los triglicéridos es insuficiente.

Tras infarto de miocardio

Tratamiento adyuvante en la prevención secundaria trasun infarto de miocardio, en combinación con lostratamientos de referencia [incluyendo estatinas,medicamentos antiplaquetarios, betabloqueantes,inhibidores de la enzima de conversión de la angiotensina(IECA)].

POSOLOGÍA Y FORMA DE ADMINISTRACIÓN

Hipertrigliceridemia

Tratamiento inicial de dos cápsulas diarias. Si no seobtiene una respuesta adecuada, puede aumentarse ladosis a cuatro cápsulas diarias.

Las cápsulas pueden tomarse con los alimentos, a fin deevitar trastornos gastrointestinales.

Tras infarto de miocardio

Una cápsula diaria.

Los datos clínicos sobre el uso de Omacor en pacientesancianos de más de 70 años de edad y pacientes coninsuficiencia renal son limitados.

No hay información sobre el uso de Omacor en niños yadolescentes o en pacientes con insuficiencia hepática.

CONTRAINDICACIONES

Hipersensibilidad al principio activo, a la soja o a algúnotro de los componentes.

ADVERTENCIAS Y PRECAUCIONES ESPECIALES DEEMPLEO

Debido al aumento moderado del tiempo de hemorragia(con la dosis elevada, es decir, 4 cápsulas), debemonitorizarse a los pacientes que reciban tratamientoanticoagulante y ha de ajustarse la dosis deanticoagulante en caso necesario. El uso de estemedicamento no excluye la necesidad de vigilancia,generalmente necesaria en esta clase de pacientes. Debeconsiderarse el aumento del tiempo de hemorragia enpacientes con un alto riesgo de hemorragia (a causa detraumatismo grave, cirugía, etc.).

Durante el tratamiento con Omacor, disminuye laproducción de tromboxano A2. No se observa un efectosignificativo en los otros factores de coagulación. Algunosestudios con ácidos omega-3 demostraron unaprolongación del tiempo de hemorragia, no obstante, eltiempo de hemorragia declarado en estos estudios noexcedía los límites normales y no produjo episodios dehemorragias clínicamente significativos. A falta de datossobre la eficacia y seguridad, no se recomienda el uso deeste medicamento en niños.

Los datos clínicos sobre el uso de Omacor en pacientesancianos de más de 70 años de edad son limitados. Sedispone únicamente de información limitada con respectoal uso en pacientes con insuficiencia renal. En algunospacientes se observó un pequeño pero significativoaumento (dentro de los valores normales) de AST y ALTpero no hay datos que indiquen un mayor riesgo enpacientes con alteración hepática. Es necesaria unamonitorización de los niveles de AST y ALT en pacientescon signos de daño hepático (en particular en los quereciban la dosis elevada, es decir, 4 cápsulas). Omacor noestá indicado en hipertrigliceridemia exógena (tipo 1hiperquilomicronemia). Solo se dispone de experiencialimitada sobre la hipertrigliceridemia endógena secundaria(especialmente diabetes no controlada). En el caso dehipertrigliceridemia no se dispone de experiencia encuanto a su combinación con fibratos.

Interacción con otros medicamentos y otras formas deinteracción

Omacor se ha administrado conjuntamente con warfarinasin que se hayan producido complicaciones hemorrágicas.No obstante, ha de controlarse el tiempo de protrombina aladministrar Omacor conjuntamente con warfarina o alsuspenderse el tratamiento con Omacor.

Embarazo: No hay datos adecuados sobre el uso deOmacor en mujeres embarazadas. Estudios en animalesno han mostrado toxicidad reproductiva. El riesgopotencial para humanos es desconocido y, enconsecuencia, Omacor no debería ser usado durante elembarazo a menos que sea claramente necesario.Lactancia: No hay datos sobre la excreción de Omacor enla leche animal y humana. Omacor no debería ser usadodurante la lactancia. Efectos sobre la capacidad paraconducir y utilizar máquinas: No se han realizado estudiosde los efectos sobre la capacidad para conducir y utilizarmáquinas. Sin embargo, es de esperar que Omacor notenga influencia sobre la capacidad para conducir y utilizarmáquinas o que ésta sea insignificante.

REACCIONES ADVERSAS

Al igual que todos los medicamentos, Omacor puedeproducir efectos adversos, aunque no todas las personaslos sufran. A continuación, se describen posibles efectosadversos que puede producir el medicamento:

Frecuentes (afectan a entre 1 y 10 de cada 100pacientes): problemas de estómago como distensiónabdominal, dolor, estreñimiento, diarrea, indigestión(dispepsia), gases, eructos, reflujo ácido, sensación demareo (náuseas) y vómitos.

Poco frecuentes (afectan a entre 1 y 10 de cada 1.000pacientes): niveles de azúcar en sangre elevados, gota,mareo, alteración del gusto, dolor de cabeza, presiónarterial baja, sangrado por la nariz, sangre en las heces,erupción.

Raros (afectan a entre 1 y 10 de cada 10.000 pacientes):reacciones alérgicas, erupción de la piel con elevaciones yenrojecimiento (ronchas o urticaria), trastornos del hígadocon posibles cambios en los resultados de algunos análisisde sangre.

PRESENTACIÓN

Envases con 20 y con 28 cápsulas.

OMACOR®

Cápsulas blandas EPA/DHA 1

g

275

J Atheroscler Thromb, 2017; 24: 275-289.doi: 10.5551/jat.34231

Original Article

A Randomized, Double-Blind, Placebo-Controlled Clinical Trial to Assess the Efficacy and Safety of Ethyl-Ester Omega-3 Fatty Acid in Taiwanese Hypertriglyceridemic Patients

Ta-Chen Su1, 2, Juey-Jen Hwang1, Kuo-Chin Huang3, Fu-Tien Chiang1, Kuo-Liong Chien1, 4, Kuo-Yang Wang5,

Min-Ji Charng6, Wei-Chuan Tsai7, Lian-Yu Lin1, Runar Vige8, José Emilio Ruiz Olivar9 and Chuen-Den Tseng1, 10

1Division of Cardiology, Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan2Institute of Occupational Medicine and Industrial Hygiene, National Taiwan University College of Public Health, Taipei, Taiwan3Department of Family Medicine, National Taiwan University Hospital, Taipei, Taiwan4Institute of Epidemiology and Preventive Medicine, College of Public Health, National Taiwan University, Taipei, Taiwan 5Cardiovascular Center and Department of Anesthesiology, Taichung Veterans General Hospital, Taichung, Taiwan 6Division of Cardiology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan 7Division of Cardiology, Internal Medicine, National Cheng Kung University Hospital, Tainan, Taiwan 8Pronova Biopharma Norge AS, Norway 9Ferrer Group, Spain

10Division of Cardiology, Department of Internal Medicine, Shin Kong Wu Ho-Su Memorial Hospital, Taipei, Taiwan

Aim: Information regarding the effects of omega-3 fatty acid on hypertriglyceridemic patients in Chinese is still limited. This study aimed to investigate the efficacy and safety of Omacor®, a pre-scription ethyl-ester omega-3 fatty acid for the treatment of hypertriglyceridemia, administered at doses of 2 g/day and 4 g/day to Taiwanese hypertriglyceridemic patients.Methods: A multicenter, randomized, double-blind, placebo-controlled, parallel study in adults with hypertriglyceridemia was conducted. After a five-week diet lead in period patients with triglycerides =200–1000 mg/dL were randomized to receive Omacor®, a concentrated preparation of omega-3 eicosapentaenoic acid (EPA) plus docosahexaenoic acid (DHA) in a dose of 1 g twice daily (2 g Oma-cor®), 2 g twice daily (4 g Omacor®) or placebo, for eight weeks. The primary endpoint was the per-centage change in triglyceride serum levels from baseline to the end of treatment.Results: A total of 253 Taiwanese patients were randomized, of which 65.6% (166) were men. At the end of the treatment, the percentage change in triglyceride serum levels in both the Omacor® 4 g/day (－32.1%) and 2 g/day (－29.7%) groups was larger than in the placebo group (－5.4%) (p＜0.001). The incidence of drug-related adverse events was as follows: 0.0%, 1.2%, and 0.0% in Oma-cor® 4 g/day, Omacor® 2 g/day, and placebo groups, respectively. No drug-related serious adverse events were reported during the study.Conclusions: Omacor® may be a feasible option to treat hypertriglyceridemia in Taiwanese patients.

See editorial vol. 24: 256-257

Key words: Hypertriglyceridemia, Omega-3 fatty acids, Omacor, Docosahexaenoic acid, Eicosapentaenoic acid, Ethnic chinese

Address for correspondence: Ta-Chen Su, Department of Internal Medicine and Cardiovascular Center, National Tai-wan University Hospital and National Taiwan University Col-lege of Medicine, No. 7, Chung-Shan South Rd, Taipei, 10020, TaiwanE-mail: tachensu@ntu.edu.twReceived: February 13, 2016Accepted for publication: June 27, 2016

Introduction

Hypertriglyceridemia, together with low levels of high-density lipoprotein cholesterol (HDL-C), is an independent risk factor for coronary heart disease, according to the guidelines for the management of tri-glycerides (TG) and cardiovascular disease1, 2). High TG levels are often concomitant with insulin resis-

Copyright©2017 Japan Atherosclerosis SocietyThis article is distributed under the terms of the latest version of CC BY-NC-SA defined by the Creative Commons Attribution License.

Su et al.

276

visit (leading period), one randomization visit, and two biweekly visits with one monthly visit during the double-blind/treatment period. The study was sup-ported and coordinated by Excelsior Pharmatech Labs. The registration number in ClinicalTrial.gov is NCT01725646. This study was conducted in accor-dance with the Investigational New Drug (IND), Informed Consent and IRB Regulations from the Tai-wan Food and Drug Administration (TFDA), and Good Clinical Practice as outlined in the International Conference on Harmonization (ICH), E6 Good Clin-ical Practice (GCP). The study was approved by all appropriate national regulatory authorities and ethics committees of the participating hospitals. All patients participated voluntarily in the study after signing the informed consent.

EthicsThe study was approved by National Taiwan

University Hospital (NTUH) Research Ethics Com-mittee on May 13, 2011, by Institutional Review Board of the Taichung Veterans General Hospital (TCVGH) on July 27, 2011, by Institutional Review Board, Taipei Veterans General Hospital (TPVGH) on Sep 16, 2011, and by National Cheng Kung University Hospital (NCKUH) Institutional Review Board on July 26, 2011. The first patient was recruited on July 8, 2011 in NTUH, on Aug 25, 2011 in TCVGH, on Nov 24, 2011 in TPVGH, and on Sep 29, 2011 in NCKUH, and the last patient was recruited on Feb 20, 2013 and completed the follow-up on May 24, 2013.

The study was registered in the Center of Drug Evaluation (CDE) Taiwan and assigned a registration number, 1001401529, before recruiting the first patient, according to Taiwan’s regulations. It was then assigned a delayed registration number of NCT01725646 on ClinicalTrial.gov, because the sponsor did not originally plan to publish the study internationally. It was thus registered in Taiwan’s CDE but not on ClinicalTrial.gov before recruiting the first patient. The authors confirm that all ongoing and related trials for this drug/intervention are registered.

Study ParticipantsPatient eligibility criteria were: 1) age between 20

and 79 years; 2) fasting serum TG level between 200 and 1000 mg/dL at screening and also at randomiza-tion; 3) having discontinued the following lipid-alter-ing agents: rosuvastatin, bile acid sequestrants, nico-tinic acid, probucol, cholesterol absorption inhibitors, gemfibrozil or fibric acid derivatives, for at least one month; 4) if a current smoker, having no plan to change smoking habits during the study; and 5) at the

tance, obesity, and acute pancreatitis, and low levels of high-density lipoprotein cholesterol (HDL-C) are present in patients with metabolic syndrome.

The standard recommendations for hypertriglyc-eridemic patients comprise measures to modify the root causes, such as weight loss, increased regular exer-cise, reduced intake of refined carbohydrates, as well as thyroid hormone replacement in patients with hypothyroidism, or alcohol cessation. At present, there are limited drugs specially indicated for the treatment of moderately elevated triglycerides; only omega-3 fatty acids, fibrates, and niacin3). The American Heart Association states that doses of fatty acids of 2–4 g/day of eicosapentaenoic acid (EPA) plus docosahexae-noic acid (DHA) can be used under a physician’s care to lower elevated triglycerides4).

Omacor ® (Pronova BioPharma), a capsule formed of 90% omega-3-acid ethyl esters, is the first US Food and Drug Administration (FDA) and EU-approved omega-3-acid drug. Omacor® is prepared from fish using various patented procedures; each cap-sule contains 460 mg of EPA and 380 mg of DHA5). The drug’s effects include lowering plasma triglyceride levels6), increasing conversion of very low-density lipo-protein cholesterol (VLDL-C) to low-density lipopro-tein cholesterol (LDL-C), depressing triglyceride syn-thesis6), and reducing postprandial lipemia7).

Several placebo-controlled studies have reported the efficacy of omega-3 fatty acids concentrates in lowering triglyceride in patients with moderately-ele-vated triglycerides. In addition, it has been observed that the degree of lowering depends on the dose and on baseline triglyceride levels3). The indicated doses of prescription omega-3 fatty acids range from 2 to 4g/day, with an efficacy on reduction of triglyceride levels of a 20%–45%3). Such a reduction is also consistent with that observed when it was combined with a statin8), such as simvastatin9). Additionally, baseline TG levels may be affected by ethnicity-related factors associated with different lifestyles and the frequency of fish intake. Indeed, the Eastern diet and lifestyle dif-fers substantially from other countries where Oma-cor® is currently authorized.

Therefore, this study aimed to demonstrate the efficacy and safety of Omacor®, administered at doses of 2 g/day and 4 g/day on hypertriglyceridemic patients in Taiwanese patients.

Materials and Methods

This was a randomized, double-blind, placebo-controlled, parallel study with two active dosage levels (1:1:1), conducted at four medical centers in Taiwan. The study consisted in five clinic visits: one screening

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

277

Including the leading period, the total follow up lasted for 13 weeks. After the five-week leading period, the double blind period covered two biweekly visits and one monthly visit. At each visit, a 12-hour fasting blood sample was obtained to determine the efficacy measurements: serum TG, TC, HDL-C, and LDL-C. The listed items were assessed locally. The TG, TC, HDL-C, and LDL-C were analyzed using a homoge-neous enzymatic method by Bayer-Siemens AVDIA 1800 Chemistry System with the corresponding reagent kits (AVDIA® Chemistry, Siemens).

Other additional information collected included: medical history review at screening and randomiza-tion, ECG at screening and at completion, dietary compliance (patient’s dietary history review), vital signs (blood pressure, pulse), safety hematology (white blood cell count, red blood cell count, platelet count, and hemoglobin), safety biochemistry test (fasting glu-cose, total protein, total bilirubin, aspartate amino-transferase (AST [SGOT]), alanine aminotransferase (ALT [SGPT]), alkaline phosphatase, total bilirubin, lactate dehydrogenase, γ- gamma-glutamyl transpepti-dase, blood urea nitrogen, creatinine, uric acid, sodium, potassium, chloride and high-sensitivity C-reactive protein). Blood samples were analyzed locally. Adverse events were reported spontaneously by the patient or elicited by open (non-leading) question-ing.

The diet control is also an important part. A reg-istered dietitian interviewed the enrolled patients monthly to evaluate their diet compliance. The patients had to record their diet in a diary card for three days before the interview with the dietitian. The dietitian reviewed the diary card in the interview and used a questionnaire to evaluate low fat diet compli-ance. In addition to the low fat diet, dietitians also evaluated fish consumption.

Study EndpointsThe primary efficacy endpoint was the effect of

Omacor® for lowering serum TG, measured by the percent change from baseline to week 8 of the given treatment. Secondary endpoints included the percent-age change in serum TG levels from baseline to week 4 of study treatment, the percentage change from baseline to weeks 4 and 8 in non-HDL-C concentra-tion, TC, HDL-C, and TC: HDL-C ratio. The safety endpoint included information on adverse events (AEs), vital signs, and clinical laboratory tests.

Statistical AnalysisSample size calculation was performed using a

superiority test based on the percentage change of TG levels. From a similar study with Omacor® in Japanese

randomization visit, either to have discontinued 3-hydroxy-3-methyl-glutaryl-CoA (HMG-CoA) reductase inhibitors, or to have been on a stable dose and schedule of HMG-CoA reductase inhibitors for at least eight weeks. The main reasons for exclusion were: 1) LDL-C control levels above the Bureau of National Health Insurance (BNHI) treatment goal at screening and randomization; 2) to be taking drugs such as non-study related omega-3, red yeast rice, weight loss drugs, immunomodulatory therapy, rosuvastatin, bile acid sequestrants, fibrates, niacin, probucol or choles-terol absorption inhibitors; 3) high consumption of fatty fish (＞two servings of 150 g of fatty fish per week during the leading period); 4) severe diseases, such as uncontrolled diabetes mellitus, thyroid dys-function, obstructive liver disease, chronic kidney dis-ease, nephrotic syndrome or any serious renal, pulmo-nary, hepatic, biliary, gastrointestinal diseases or cancer within 6 months prior to randomization; and 5) a his-tory of alcoholism (taking at least three glasses of wine or equivalent per day) during the previous three months. After the screening visit, patients who either had hypolipidemia with total cholesterol (TC) ＜120 mg/dL and LDL-C ＜50 mg/dL at any visit or had TG ＞1,000 mg/dL with a 30% elevation or more compared to baseline were requested to be withdrawn from study.

ProceduresFor this intervention study, eligible patients

entered a five-week leading period and were counseled on low fat diets throughout the study. After this lead-ing phase, patients needed serum triglycerides levels of 200 mg/dL or more and to meet the selection criteria to be randomized. Patients were equally stratified by their lipid-altering agents/statin (with or without) and by the baseline triglyceride level (200–499 mg/dL or 500–1,000 mg/dL) prior to being randomized. The randomization schedules were generated using a vali-dated SAS system that automates the random assign-ment of treatment groups to randomized numbers and were prepared with a 1:1:1 randomization ratio in block numbers as 6. Thus, patients were randomized to receive 1 g of Omacor® twice a day (2 g Omacor®/day), 2 g of Omacor® twice a day (4 g Omacor®/day) or placebo (i.e., olive oil in identical capsules) twice a day for a total of eight weeks. The Omacor® 2 g/day arm received one bottle with Omacor® capsules and one bottle with placebo capsules; the Omacor® 4 g/day arm received two bottles with Omacor® capsules; and the placebo arm received two bottles with placebo capsules. Every study medication bottle had the same appearance. The investigator kept individual blind-breaker envelopes containing the drug assignments.

Su et al.

278

For the analysis of TG levels, the geometric mean and 95% confidence interval (CI) were computed by taking the exponent of the mean and of the lower and upper limits of the 95% CI of the natural-log-trans-formed sizes. The change from baseline to the end-of-treatment in natural-log-transformed TG level was evaluated by an analysis of covariance (ANCOVA) model with terms for baseline natural-log-transformed TG value, baseline fatty fish consumption (i.e., none, ≤ 150 g or ＞150 g), stratification factors and treat-ment group. The test drug (Omacor®) was concluded to be superior to the control (placebo) on average if the null hypothesis was rejected. The test was per-formed separately for each dose versus placebo. The comparison between 2 g/day Omacor® and placebo could only be concluded if 4 g/day Omacor® was superior to placebo at the significant level of 0.05 (two-sided). For the secondary efficacy endpoints, hypothesis tests (all two-sided) were performed indi-vidually at the 5% significance level and there was no adjustment for multiple tests or for adjustment for multiplicity of endpoints. All data were analyzed using the SAS system.

The safety profile of Omacor® was analyzed according to adverse events reported during the study, vital signs measurements, serum chemistry and hema-tology results, and ECG reports. This analysis was

cohorts10), a standard deviation (SD) of approximately 30% may be expected. Similarly, in the cited study, a reduction in serum TG of 12% and 26.6% was observed for patients with plasma TG above 200 mg/dL treated with Omacor® 2 g/day and Omacor® 4 g/day at eight weeks.

With a SD of 30% and estimated 80% power, a sample size of 74 patients in each treatment group was required to detect a difference in TG reduction of 14% at the 5% significance level (two-sided). To adjust for an expected 10% drop-out or non-compli-ance rate, 83 patients were included in each treatment group.

The intent-to-treat (ITT) population comprised all patients who were randomized to the study treat-ment and who had taken at least one dose of study medication and have at least one follow-up efficacy endpoint evaluation. The primary efficacy analysis was based on the ITT population used a last observation carried forward (LOCF) method of imputation for missing response variable due to patient early termina-tion or incomplete assessment. The primary endpoint was also analyzed in the per protocol (PP) population, which included all patients who underwent any study treatment and had no major protocol violations affect-ing their efficacy assessments; and in the created strata based on statin usage or hypertriglyceridemia levels.

Patients assessedfor eligibility

Randomization(n = 253)

Allocated toOmacor 4 g group

(n = 84)

Allocated toOmacor 2 g group

(n = 82)

Allocated toPlacebo group

(n = 87)

WithdrawalLost to follow up (n = 1)Adverse event (n = 1)Lack of efficacy (n = 1)Investigator’s opinion

(n = 1)

Withdrawal

Investigator’s opinion(n = 2)

WithdrawalWithdrawal consent

(n = 1)Adverse event (n = 2)Lack of efficacy (n = 1)Investigator’s opinion

Analyzed forPer Protocol (n = 65)Intention to treat

Analyzed forPer Protocol (n = 77)Intention to treat

Analyzed forPer Protocol (n = 68)Intention to treat

Excluded (n = 245)Not meeting inclusioncriteria (n = 190)

Fig.1. Flowchart of Subject Disposition in Trial

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

279

tary Table 2. From baseline to the end of the study, at least about 80% of patients achieved low fat diet com-pliance in each treatment group, without statistically significant differences between treatment with Oma-cor® 4 g/day and Omacor® 2 g/day groups during the study follow-up. Regarding weekly fish consumption, the proportion of patients who consumed none or less than 150 g of fish per week reached at least 90% from the leading period to the end of eight weeks of treat-ment (Supplementary Table 3).

Efficacy AnalysisStatistically significant reductions in TG levels

were observed between Omacor® arms and placebo as early as week 2 and up to the end of the study (Fig.2). At week 8, the percent of change in both Omacor® 4 g/day (－32.1%) and 2 g/day (－29.7%) groups was significantly larger than in the placebo group (－5.4%), Omacor® 4 g/day vs. placebo, p＜0.0001; Omacor® 2 g/day vs. placebo, p＜0.0001. When the PP population was analyzed at week 8, the percent of change was －32.4%, －31.1%, and －10.0% in Omacor® 4 g/day, Omacor® 2 g/day and placebo groups, respectively, with statistically significant differ-ences between groups (Omacor® 4 g/day vs placebo, p=0.0001; Omacor® 2 g/day vs placebo, p=0.0002). Similar results were obtained at week 4, with percent changes of －27.8%, －28.7%, and －6.8% in Oma-cor® 4 g/day, Omacor® 2 g/day and placebo groups, respectively, with statistical differences between groups (Omacor® 4 g/day vs placebo, p＜0.0001; Omacor® 2 g/day vs placebo, p＜0.0001).

In addition to triglycerides, only LDL-C levels reached statistical significance when Omacor® and

conducted on the “Safety Population,” which included all randomized patients. The Medical Dictionary for Regulatory Activities (MedDRA, version 15.0) adverse event dictionary was used to map verbatim adverse events to preferred terms and system organ class.

Results

An overview of patient disposition from screen-ing to study termination is provided in Fig.1. This study was conducted from July 08, 2011 (first patient first visit) to May 24, 2013 (last patient last visit). A total of 498 patients were screened and 253 were ran-domized: 84 patients to Omacor® 4 g/day, 82 patients to Omacor® 2 g/day, and 87 patients to placebo. Of all randomized patients, 240 patients completed the study. The baseline characteristics of each group are shown in Table 1. A total of 166 (65.6%) patients were males and the mean age was 54 years old. At baseline, the mean (SD) TG levels were 375.1 (151.7) mg/dL, 363.5 (157.0) mg/dL, and 359.0 (138.8) mg/dL in Omacor® 4 g/day, Omacor® 2 g/day, and pla-cebo groups, respectively.

ComplianceMedian (interquartile range, IQR) compliance

during the double-blind treatment phase was 96.6% (50.0%–117.0%), 97.1% (70.7%–110.2%) and 95.8% (59.3%–123.6%) in the Omacor® 4 g/day, Omacor® 2 g/day and placebo groups, respectively (Supplementary Table 1). About 85%–95% of patients in each treatment groups achieved a good compliance with treatment percentage above 85%. The low fat diet compliance is shown in Supplemen-

Table 1. Baseline characteristics of participants by treatment groups

Characteristics Omacor® 4 g/dayN=84

Omacor® 2 g/dayN=82

PlaceboN=87

MaleAge (years)Body Mass Index (kg/m2)With Statin TreatmentTriglycerides level

≥ 200 and ＜500 mg/dL≥ 500 and ≤ 1,000 mg/dL

Any Cardiovascular DiseaseCerebral Vascular DiseaseCoronary Artery DiseaseHypertensionDiabetes mellitusSmoking habit

55 (65.5%)53.7 (11.0)26.63(3.73)29 (34.5%) 70 (83.3%)14 (16.7%)

4 (4.8%)1 (1.2%)3 (3.6%)

55 (65.5%)21 (25.0%)14 (16.7%)

47 (57.3%)54.7 (9.2)26.61(4.19)29 (35.4%) 70 (85.4%)12 (14.6%)

3 (3.7%)1 (1.2%)2 (2.4%)

51 (62.2%)19 (23.2%)13 (15.9%)

64 (73.6%)54.4 (10.7)26.66(3.86)31 (35.6%) 73 (83.9%)14 (16.1%)

4 (4.6%)1 (1.1%)4 (4.6%)

50 (57.5%)17 (19.5%)22 (25.3%)

Data was presented as mean (SD) if continuous variables and n (%) if binary variables

Su et al.

280

one patient treated with Omacor 4 g/day, who had an AE of mild myalgia, and then recovered without fur-ther treatment during the study, and the investigator thought it unlikely related to the study drug. The serum creatinine level also did not show significant elevation during the study in the three study groups (Supplementary Table 4). No drug-related severe AEs were observed during the study.

Discussion

The goal of the present study was to demonstrate the efficacy and safety of the doses of Omacor® 2 g/day and 4 g/day in the Taiwanese population. Based on the results obtained, the primary hypothesis was confirmed; significant improvements in triglyceride levels in both Omacor® arms versus placebo were achieved. Furthermore, it was confirmed in both the ITT and the PP population, which demonstrates the robustness of the outcome. These improvements are in agreement with the previous clinical evidence10). In fact, a review by Skulas et al.3) indicated an average triglyceride reduction for patients at the higher end of moderate hypertriglyceridemia of approximately 30% for Omacor® in studies with treatment duration between eight weeks and six months. Additionally, two randomized, double-blind, placebo-controlled studies assessing the effect of Omacor® 4 g/day reported a reduction of －38.9% after six weeks of treatment11) and －45% in TG levels after 16 weeks of treatment12), in patients with TG levels within the range of 500–2,000 mg/dL.

A reference study performed in the United States, the COMBOS trial7), which used a dose of 4 g/day in

placebo groups were compared (Table 2). With regard to LDL-C levels, significant changes from baseline were reported at week 4 with Omacor® 2 g/day (7.2%, p=0.0036) and Omacor® 4 g/day (6.3%, p=0.0096) and sustained until the end of the study (9.9%, p=0.0136 and 7.2%, p=0.0010, respectively). Additionally, the mean LDL-C levels remained within normal ranges in all patients among the three groups during the whole study duration.

In the subgroup analysis of the primary end-point, when the stratification was for statin or no lipid-altering medication use, the percentage change in TG levels was larger in the group that did not receive lipid-altering agents (Table 3). Similarly, the severe hypertriglyceridemia group experienced a larger change in TG levels than the group with moderate hypertriglyceridemia, although it was not statistically significant owing to the small patient numbers in the severe hypertriglyceridemia stratum (Table 4).

Safety AnalysisThe incidence of drug-related AEs was 0.0%,

1.2% (one case of nausea), and 0.0% in Omacor® 4 g/day, Omacor® 2 g/day, and placebo groups, respec-tively. Only one patient (1.2%) treated with Omacor® 4 g/day group had an AE with conjunctival hemor-rhage, which was unlikely related to the study drug. The incidence of serious AEs in Omacor® 2 g/day group was 1.2% (one case of coronary artery disease), and in placebo group was 1.1% (one case of sick sinus syndrome), both of them were unlikely related to the study drug. The incidence of AEs that led to the study drug discontinuation was 1.2% in the Omacor® 4 g/day group and 2.3% in placebo group. There was only

50

40

30

20

10

0

10

Baseline Week 2 Week 4 Week 8 *EOS

Omacor 4g/day

Omacor 2g/day

Placebo

Fig.2. Time-course of percent change in triglyceride levels (ITT population)

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

281

Table 2. Treatment effects after lipid-lowering therapy from baseline to end of study

Omacor® 4 g(n =82)

Omacor® 2 g(n =82)

Placebo (n =87)

p-value#

Geometric Mean

95%CIGeometric

Mean95%CI

Geometric Mean

95%CIOmacor 4 g/daya

Omacor 2 g/dayb

Total cholesterolBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline,

TriglyceridesBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %

HDL-CBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %

Non-HDL-CBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %

LDL-CBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %

184.0176.9－3.9%180.0－1.8%178.3－3.1%

351.2243.2－30.7%

251.9－27.8%

238.5－32.1%

37.138.02.3%37.91.9%38.02.4%

146.1137.9－5.6%141.4－2,7%139.1－4.8%

80.487.48.7%85.56.3%86.27.2%

177.5 – 190.8169.4 – 184.7－6.0 – －1.8%172.3 – 188.1－4.3 – 0.8%170.3 – 186.7－5.9 – －0.3%

325.2 – 379.3219.5 – 269.5－36.2 – －24.8%229.7 – 276.2－33.7 – －21.4%216.1 – 263.1－38.0 – －25.6%

35.8 – 38.536.6 – 39.4－0.5 – 5.2%36.5 – 39.4－1.9 – 5.9%36.4 – 39.7－1.3 – 6.4%

139.9 – 152.5131.0 – 145.3－8.3 – －2.8%134.3 – 148.8－5.7 – －0.4%131.7 – 146.9－8.0 – －1.4%

75.3 – 85.981.7 – 93.6

4.1 – 13.6%78.8 – 92.7

0.9 – 12.1%79.7 – 93.2

1.0 – 13.7%

186.9184.2－1.4%179.9－3.7%181.3－3.0%

338.6262.8－22.4%

241.6－28.7%

238.2－29.7%

39.339.81.5%39.40.4%39.91.6%

146.6143.4－2.2%139.5－4.9%140.4－4.2%

80.889.0

10.1%86.67.2%88.89.9%

179.9 – 194.0176.7 – 192.0－4.0 – 1.3%172.0 – 188.2－6.7 – －0.7%173.2 – 189.7－6.1 – 0.3%

313.0 – 366.4238.8 – 289.1－28.9 – －15.3%221.2 – 263.9－34.4 – －22.4%218.8 – 259.4－35.4– －23.4%

37.7 – 40.938.3 – 41.4－1.6 – 4.6%37.6 – 41.3－2.6 – 3.6%38.1 – 41.8－1.5 – 4.8%

140.1 – 153.4136.6 – 150.7－5.4 – 1.2%132.4 – 147.0－8.3 – －1.3%133.2 – 148.0－8.1 – －0.2%

75.2 – 86.882.6 – 96.0

4.9 – 15.7%79.3 – 94.6

1.0 – 13.6%81.5 – 96.9

3.9 – 16.3%

185.7184.1－0.9%182.0－1.9%186.80.6%

336.9311.3－7.6%311.9－6.8%318.7－5.4%

37.437.91.2%37.80.8%38.63.2%

147.4145.3－1.4%143.3－2.6%147.2－0.1%

83.483.2－0.3%

81.0－3.6%

80.5－3.5%

178.9 – 192.8177.1 – 191.2－3.0 – 1.3%173.8 – 190.5－4.5 – 0.7%178.9 – 195.1－2.2 – 3.5%

312.8 – 362.9282.8 – 342.6－14.4 – －0.2%282.1 – 344.8－14.1 – 1.1%283.6 – 358.2－14.8 – 5.1%

35.9 – 39.136.4 – 39.5－1.2 – 3.6%36.3 – 39.4－1.9 – 3.5%37.0 – 40.30.1 – 6.3%

141.3 – 153.8139.0 – 151.9－4.0 – 1.2%136.0 – 151.0－5.7 – 0.5%140.0 – 154.7－3.6 – 3.4%

77.6 – 89.776.6 – 90.4－4.8 – 4.4%74.3 – 88.3－8.0 – 0.9%73.5 – 88.2－8.9 – 2.2%

0.7320.035

0.997

0.068

0.442＜.001

＜.001

＜.001

0.7570.617

0.662

0.690

0.7650.018

0.891

0.059

0.4570.010

0.010

0.014

0.8120.747

0.309

0.092

0.9230.003

＜.001

＜.001

0.0940.348

0.587

0.967

0.8580.609

0.257

0.101

0.5240.001

0.004

0.001

Abbreviations: HDL-C and LDL-C, High-density and Low-density lipoprotein cholesterol#p-value: Omacor 4 g/daya indicates the differences between Omacor 4 g/day and placebo after treatment Omacor 2 g/dayb indicates the differences between Omacor 2 g/day and placebo after treatment

Su et al.

282

reported lower TG reduction in comparison with our study (－10.8% in the Omacor® 2 g group and －22.9% in the Omacor® 4 g group). A comparison among Tai-wan, Japan, and UK studies at dose 4 g was showed in Supplementary Table 5.

Surprisingly, the two doses used in our study, 4 g/day and 2 g/day, achieved a very similar percentage of TG reduction. Based on the clinical evidence, it is

combination with simvastatin for eight weeks versus placebo, reported a median of percent change of －29.5% in the Omacor®＋simvastatin arm and －6.3% in the placebo simvastatin arm. Another reference study performed in UK13) reported a mean of percent change of －22.4%. More specific to the Asian area is a Japanese phase Ⅲ study10), similar to ours in both the population of patients and doses used which

Table 3. Subgroup analyses by usage of statins and triglyceride-lowering effects at end of study (ITT population)

Subgroups Treatment Group NBaseline TG level

(Geometric Mean with 95% CI), mg/dL

Percent Change from Baseline

(Intra p-value)

Omacor® vs Placebo(Group Difference)

p-value

Statin usage

Stable statins

Omacor® 4 g 28339.8

(292.8 − 394.5)－28.7% (＜.001)

－29.7%(－43.4 − －12.5%)

0.002

Omacor® 2 g 29309.1

(273.9 − 348.9)－27.5% (＜.001)

－31.3%(－44.7 − －14.6%)

＜.001

Placebo 31325.0

(281.1 − 375.7)4.3% (0.678) -- --

No lipid- altering Agent

Omacor® 4 g 54357.3

(326.2 − 391.4)－33.8% (＜.001)

－25.0%(－36.0 − －12.0%)

＜.001

Omacor® 2 g 53356.0

(321.3 − 394.4)－30.8% (＜.001)

－22.0%(－33.5 − －8.4%)

0.003

Placebo 56343.7

(315.2 − 374.6)－10.4% (0.072) -- --

BaselineTGlevel

Moderatehypertrigly ceridemia

Omacor® 4 g 68311.3

(294.3 − 329.3)－28.4% (＜.001)

－25.3%(－34.5 − －14.9%)

＜.001

Omacor® 2 g 70302.3

(286.2 − 319.4)－25.2% (＜.001)

－22.7%(－32.1 − －12.0%)

＜.001

Placebo 73301.3

(284.4 − 319.2)－3.0% (0.588) -- --

Severe hypertrigly ceridemia

Omacor® 4 g 14631.4

(562.6 − 708.6)－47.7% (0.001)

－35.6%(－57.9 − －1.5%)

0.043

Omacor® 2 g 12656.4

(565.9 −761.4)－50.7% (＜.001)

－38.3%(－60.9 − －2.6%)

0.039

Placebo 14603.0

(546.7 − 665.1)－16.8% (0.231) -- --

p-value: Paired t-test for intragroup comparison; t-test per ANCOVA model for intergroup comparison.

Table 4. Comparison of triglyceride levels change between the subgroup of severe and moderate hyper-triglyceridemia after treatment

Subgroup 200 mg/dL ≤ TG ＜500 mg/dL TG ≥ 500 mg/dL p-value

n Change n Change

Omacor® 4 gOmacor® 2 g

6870

－28.4%－25.2%

1412

－47.7%－50.7%

0.60950.6654

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

283

the severe hypertriglyceridemia stratum experienced a larger change in TG levels than those in the moderate hypertriglyceridemia stratum, although it was not sta-tistically significant. In all cases, treatment groups ver-sus placebo were statistically significant in both the Omacor® 4 g/day and 2 g/day arms.

On the other hand, LDL-C levels slightly increased after four to eight weeks with both doses of Omacor®, comparing with placebo, although the mean levels of LDL-C remained within the normal ranges throughout the study in all groups. In this con-text, Maki et al. in a post hoc analysis of the COMBOS trial25) showed that Omacor® treatment also increased the levels of LDL-C but only in patients in the lowest tertile (＜80.4 mg/dL) level at the beginning of the study. In the present study, mean (SD) baseline LDL-C levels were 83.9 (24.3) mg/dL, 84.9 (25.8) mg/dL, and 88.1 (28.1) mg/dL in Omacor® 4 g/day, Omacor® 2 g/day, and placebo groups, respectively, which were close to the report of Maki et al. Another example is the TAK-085 study in Japan26), in which the mean (SD) baseline LDL-C levels were 129.0 (30.3) mg/dL, 133.2 (29.9) mg/dL, and 129.3 (33.0) mg/dL in Omacor® (TAK-085) 4 g/day, Omacor® (TAK-085) 2 g/day, and EPA-E groups, respectively, resulting in a slight increase or even decrease at the end of study (2.38±20.5%, －0.42±17.3%, and －1.49±16.8% in Omacor® 4 g/day, Omacor® 2 g/day, and EPA-E groups, respectively). However, it seems that the increment of LDL-C level was larger in Omacor® 2 g/day than that in Omacor® 4 g/day in the present study, but this was not statistically signifi-cant (p=0.391). Further studies are needed to clarify the mechanism.

Additionally, a subgroup analysis of the COM-BOS trial focusing on lipoprotein sizes and concentra-tions revealed that Omacor® 4 g/day produced signifi-cant reductions in intermediate density lipoprotein cholesterol (IDL-C) (p＜0.0001 vs. baseline) and small LDL-C (p=0.0035 vs. baseline)27). The TAK-085 study in Japan also showed a shift from small dense LDL-C to large buoyant LDL-C and increased LDL-C/Apo B ratios, as well as a decrement of Apo C-Ⅲ after treatment with Omacor®28). Higher levels of IDL-C and small dense LDL-C usually presenting in hypertriglyceridemic patients are considered more harmful than other lipid particles because they can more easily penetrate the arterial wall29). Our recent study also demonstrated that small dense LDL-C par-ticles might potentiate postchallenge hyperglycemia on the risk of arterial stiffness in middle-aged healthy adults30). The safety profile of both Omacor® arms was as expected, with a rate of adverse event reactions comparable to the 3% reported in the Summary of

widely accepted that the effect of this drug is dose-dependent14, 15). In addition, the Japanese study dem-onstrated the expected differences between both active groups. In the investigators’ opinion, one explanation could be moderation of Chinese foods with a large amount of carbohydrates may attenuate the additional dose of omega-3 fatty acids on triglyceride-lowering effects16). Besides, the low fat diet compliances for Omacor® 4 g and Omacor® 2 g groups were 81.0% and 91.5%, respectively. Although the difference was not statistically significant; however, it was close to statistical significance (p=0.067). The trend of better compliance of maintaining low fat diet may contrib-ute to the triglyceride-lowering effects for Omacor® 2 g group, thus attenuate the dose-dependent response of higher dose of Omacor® 4 g/day.

After treatment with Omacor® 2 g, a higher omega-3/omega-6 ratio also significantly improved insulin resistance and subsequently decreased the lev-els of triglycerides of study participants17, 18). Another explanation is the high percentage of overweight in our study patients, which had been demonstrated to have a higher risk of severe hypertriglyceridemia (TG ≥ 500 mg/dL) while concurrent with genetic polymor-phism in apolipoprotein A5 in Chinese19). An interac-tion between omega-3 fatty acids supplement and genetic variants in APOA5 and/or APOE4 may mod-ify the response to omega-3 fatty acids treatment sig-nificantly independent of dose-response pattern in these patients. Not only the insulin resistance, but also the arterial stiffness, can be improved by a higher EPA/AA ratio20).

In addition, our study demonstrated an even larger reduction in triglyceride levels in Taiwanese compared to Japanese participants. The difference in triglyceride-lowering effects between the Taiwanese and Japanese might be mediated by the difference in eating habits: more fish and seaweeds are consumed by the Japanese (rich in omega-3 fatty acids), which may attenuate the additional lipid-lowering effects of fish oil in Omacor®21).

It is worth mentioning that the significant improvement with Omacor® was achieved at week two and maintained until the end of our study. According to Rupp et al.5), when Omacor® 1 g/day is administered for 30 days, the bioavailability of the drug remains constant and optimal during the treat-ment period, reaching plateau levels from the tenth day. In our literature review, most studies also support such a reduction after week 413, 22-24), but no data is available for the preceding period and it therefore can-not be appropriately contextualized.

The effectiveness of Omacor® is influenced by baseline TG levels3). Indeed, in our study, patients in

Su et al.

284

pharma Norge AS, Norway. José Emilio Ruiz Olivar is an employee of Ferrer group, Spain. They offered technical support to Excelsior and the study group. Excelsior has a partnership with Ferrer and Pronova. Excelsior is an exclusive agent of Ferrer in Taiwan, and was authorized by Pronova to conduct Omacor® study in Taiwan.

References

1) Miller M, Stone NJ, Ballantyne C, Bittner V, Criqui MH, Ginsberg HN, et al.: Triglycerides and cardiovascular dis-ease: a scientific statement from the American Heart Asso-ciation. Circulation, 2011; 123: 2292-2333

2) Chapman MJ, Ginsberg HN, Amarenco P, Andreotti F, Boren J, Catapano AL, et al.: Triglyceride-rich lipopro-teins and high-density lipoprotein cholesterol in patients at high risk of cardiovascular disease: evidence and guid-ance for management. Eur Heart J, 2011; 32: 1345-1361

3) Skulas-Ray AC, West SG, Davidson MH, Kris-Etherton PM: Omega-3 fatty acid concentrates in the treatment of moderate hypertriglyceridemia. Expert Opin Pharmaco-ther, 2008; 9: 1237-1248

4) Kris-Etherton PM, Harris WS, Appel LJ: Omega-3 fatty acids and cardiovascular disease: new recommendations from the American Heart Association. Arterioscler Thromb Vasc Biol, 2003; 23: 151-152

5) Rupp H: Omacor (prescription omega-3-acid ethyl esters 90): From severe rhythm disorders to hypertriglyceride-mia. Adv Ther, 2009; 26: 675-690

6) Davidson MH: Mechanisms for the hypotriglyceridemic effect of marine omega-3 fatty acids. Am J Cardiol, 2006; 98(4A): 27i-33i

7) Berge RK, Madsen L, Vaagenes H, Tronstad KJ, Gottli-cher M, Rustan AC: In contrast with docosahexaenoic acid, eicosapentaenoic acid and hypolipidaemic deriva-tives decrease hepatic synthesis and secretion of triacylg-lycerol by decreased diacylglycerol acyltransferase activity and stimulation of fatty acid oxidation. Biochem J, 1999; 343: 191-197

8) Nambi V, Ballantyne CM: Combination therapy with statins and omega-3 fatty acids. Am J Cardiol, 2006; 98(4A): 34i-38i

9) Davidson MH, Stein EA, Bays HE, Maki KC, Doyle RT, Shalwitz RA, et al.: Efficacy and tolerability of adding prescription omega-3 fatty acids 4 g/d to simvastatin 40 mg/d in hypertriglyceridemic patients: an 8-week, ran-domized, double-blind, placebo-controlled study. Clin Ther, 2007; 29: 1354-1367

10) Tatsuno I, Saito Y, Kudou K, Ootake J: Efficacy and safety of TAK-085 compared with eicosapentaenoic acid in Japanese subjects with hypertriglyceridemia undergoing lifestyl modification: The omega-3 fatty acids randomized double-blind (ORD) study. J Clin Lipidol, 2013; 7: 199-207

11) Pownall HJ, Brauchi D, Kilinc C, Osmundsen K, Pao Q, Payton-Ross C, et al.: Correlation of serum triglyceride and its reduction by omega-3 fatty acids with lipid trans-fer activity and the neutral lipid compositions of high-density and low-density lipoproteins. Atherosclerosis,

Product Characteristics. Indeed, no significant safety concerns arose during the course of the trial.

In the COMBOS trial, with a treatment period of eight weeks, the short treatment duration has been noted as a study limitation. However, a 24-month extension of the COMBOS study confirmed the results about efficacy and safety31). In this case, the goal of the study was to confirm the efficacy of Oma-cor® and the duration was not considered a weakness for this purpose, but the authors admit that longer tri-als might better characterize clinical efficacy and safety. Additionally, the specificity of the population, all of which was Asian, offers valuable information regarding these patients, though it makes it difficult to extrapolate the results to other patient populations.

EPA also showed the effects on inflammatory markers, including high sensitivity CRP32). However, in the present study, there was no significant change in all of three treatment groups (Supplementary Table 6). This may have been due to the shorter period and the different remedies and doses, or differ-ent patient cohorts.

In conclusion, the use of Omacor® 4 g/day or Omacor® 2 g/day in Taiwanese patients significantly reduced TG levels compared to placebo in the present study. Although extrapolation of data from clinical tri-als to routine clinical practice is not always straightfor-ward, owing to the heterogeneous patient population, the presence of comorbidities, and the unstructured follow-up characteristics of daily practice, Omacor® appears a valid option for treating hypertriglyceride-mia in Taiwanese patients. Furthermore, differences in diet and behavior in the Chinese population reinforce the need to conduct confirmatory studies on the effi-cacy of Omacor® in this specific population of patients.

Acknowledgements

We thank doctors from medical centers, includ-ing Chia-Ti Tsai, Jyh-Ming Juang, and Yi-Chih Wang from National Taiwan University Hospital, and Ting-Hsing Chao, Chih-Chan Lin, and Cheng-Han Lee from National Cheng-Kung University Hospital, who help the recruitments of patients in this study.

Funding

This work was supported by Excelsior Pharmat-ech Labs, Taipei, Taiwan.

Disclosures

Runar Vige is an employee of Pronova Bio-

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

285

23) Eritsland J, Arnesen H, Gronseth K, Fjeld NB, Abdelnoor M: Effect of dietary supplementation with n-3 fatty acids on coronary artery bypass graft patency. Am J Cardiol, 1996; 77: 31-36

24) Leigh-Firbank EC, Minihane AM, Leake DS, Wright JW, Murphy MC, Griffin BA, et al.: Eicosapentaenoic acid and docosahexaenoic acid from fish oils: differential asso-ciations with lipid responses. Br J Nutr, 2002; 87: 435-445

25) Maki KC, Dicklin MR, Davidson MH, Doyle RT, Bal-lantyne CM: Baseline lipoprotein lipids and low-density lipoprotein cholesterol response to prescription omega-3 acid ethyl ester added to simvastatin therapy. Am J Car-diol, 2010; 105: 1409-1412

26) Tatsuno I, Saito Y, Ootake J: Long-term safety and effi-cacy of TAK-085 in Japanese subjects with hypertriglycer-idemia undergoing lifestyle modification: The omega-3 fatty acids randomized long-term (ORL) study. J Clin Lipidol, 2013; 7: 615-625

27) Maki KC, McKenney JM, Reeves MS, Lubin BC, Dicklin MR: Effects of adding prescription omega-3 acid ethyl esters to simvastatin (20 mg/day) on lipids and lipopro-tein particles in men and women with mixed dyslipid-emia. Am J Cardiol, 2008; 102: 429-433

28) Tatsuno I, Kudou K, Kagawa T: Effect of TAK-085 on low-density lipoprotein particle size in patients with hypertriglyceridemia: A double-blind randomized clinical study. Cardiovasc Ther, 2015; 33: 317-323

29) Austin MA, Breslow JL, Hennekens CH, Buring JE, Wil-lett WC, Krauss RM: Low-density lipoprotein subclass patterns and risk of myocardial infarction. JAMA, 1988; 260: 1917-1921

30) Ding C, Hsu SH, Wu YJ, Su TC: Additive effects of post-challenge hyperglycemia and low-density lipoprotein par-ticles on the risk of arterial stiffness in healthy adults. Lip-ids Health Dis, 2014; 13: 179

31) Bays HE, Maki KC, McKenney J, Snipes R, Meadowcroft A, Schroyer R, et al.: Long-term up to 24-month efficacy and safety of concomitant prescription omega-3-acid ethyl esters and simvastatin in hypertriglyceridemic patients. Curr Med Res Opin, 2010; 26: 907-915

32) Kohashi K, Nakagomi A, Saiki Y, Morisawa T, et al.: Effects of eicosapentaenoic acid on the levels of inflamma-tory markers, cardiac function and long-term prognosis in chronic heart failure patients with dyslipidemia. J Athero-scler Thromb, 2014; 21: 712-729

1999; 143: 285-29712) Harris WS, Ginsberg HN, Arunakul N, Shachter NS,

Windsor SL, Adams M, et al.: Safety and efficacy of Omacor in severe hypertriglyceridemia. J Cardiovasc Risk, 1997; 4: 385-391

13) Durrington PN, Bhatnagar D, Mackness MI, Morgan J, Julier K, Khan MA, et al.: An omega-3 polyunsaturated fatty acid concentrate administered for one year decreased triglycerides in simvastatin treated patients with coronary heart disease and persisting hypertriglyceridaemia. Heart, 2001; 85: 544-548

14) Balk E, Chung M, Lichtenstein A, Chew P, Kupelnick B, Lawrence A, et al.: Effects of omega-3 fatty acids on car-diovascular risk factors and intermediate markers of car-diovascular disease. Evid Rep Technol Assess (Summ), 2004; 93: 1-6

15) Bradberry JC, Hilleman DE: Overview of omega-3 Fatty Acid therapies. P T, 2013; 38: 681-691

16) Harris WS, Connor WE, Inkeles SB, Illingworth DR: Dietary omega-3 fatty acids prevent carbohydrate-induced hypertriglyceridemia. Metabolism, 1984; 33: 1016-1019

17) Simopoulos AP: The importance of the ratio of omega-6/omega-3 essential faty acids. Biomed & Pharmacother, 2002; 56: 365-379

18) Lin CY, Chen MF, Lin LY, Liau CS, Lee YT, Su TC: Insu-lin resistance is the major determinant for microalbumin-uria in severe hypertriglyceridemia: Implication for high-risk stratification. Int Med, 2008; 47: 1091-1097

19) Lee MJ, Chien KL, Chen MF, Stephenson DA, Su TC: Overweight modulates APOE and APOA5 alleles on the risk of severe hypertriglyceridemia. Clin Chim Acta, 2013; 416: 31-35

20) Ito R, Satoh-Asahara N, Yamakage H, Sasaki Y, et al.: An increase in the EPA/AA ratio is associated with improved arterial stiffness in obese patients with dyslipidemia. J Atheroscler Thromb, 2014; 21: 248-260

21) Davidson MH, Maki KC, Bays H, Carter R, Ballantyne CM: Effects of prescription omega-3-acid ethyl esters on lipoprotein particle concentrations, apolipoproteins AI and CIII, and lipoprotein-associated phospholipase A(2) mass in statin-treated subjects with hypertriglyceridemia. J Clin Lipidol, 2009; 3: 332-340

22) Calabresi L, Donati D, Pazzucconi F, Sirtori CR, France-schini G: Omacor in familial combined hyperlipidemia: effects on lipids and low density lipoprotein subclasses. Atherosclerosis, 2000; 148: 387-396

Su et al.

286

Supplementary Table 1. Summary of study drug compliance

Group difference

Omcor 4 g/day Omacor 2 g/day Placebo Omacor 4 g/day vs Placebo

Omacor 2 g/day vs Placebo

Omacor 4 g/day vs Omacor 2 g/day

Omacor vs Placebo

NMean (SD), %Median, %(Min, Max), %Mean, %95% CI, %p-value (a)p-value (b)

8294.9 (10.6)

96.6(50.0, 117.0)

8296.0 (6.4)

97.1(70.7, 110.2)

8793.5 (10.3)

95.8(59.3, 123.6)

1.40－1.43 ~ 4.22

0.33140.3051

2.56－0.26 ~ 5.38

0.07550.0891

－1.16－4.03 ~ 1.70

0.42460.5039

1.98－0.46 ~ 4.41

0.1108

p-value(a): t-test for intergroup comparisonp-value(b): Rank transformation ANOVA for intergroup comparison

Supplementary Table 2. Summary of low fat diet compliance

Omacor 4 g/day Omacor 2 g/day Placebo p-value(Overall)N % N % N %

Baseline 0.262Compliance 62 75.6% 69 84.1% 74 85.1%Not Compliance 20 24.4% 13 15.9% 13 14.9%p-value (Group Comparison) 0.242 (4 g vs 2 g) 1.000 (2 g vs Placebo) 0.174 (4 g vs Placebo)

First Month 1.000Compliance 64 79.0% 63 77.8% 67 78.8%Not Compliance 17 21.0% 18 22.2% 18 21.2%p-value (Group Comparison) 1.000 (4 g vs 2 g) 1.000 (2 g vs Placebo) 1.000 (4 g vs Placebo)

Second Month 0.129Compliance 64 81.0% 75 91.5% 69 83.1%Not Compliance 15 19.0% 7 8.5% 14 16.9%p-value (Group Comparison) 0.067 (4 g vs 2 g) 0.160 (2 g vs Placebo) 0.838 (4 g vs Placebo)

p-value: Fisher’s exact test

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

287

Supplementary Table 3. Summary of fish consumption

Omacor 4 g/day Omacor 2 g/day Placebo p-value(overall)N % N % N %

Baseline 0.4452None per week 27 32.9% 36 43.9% 28 32.2%≤ 150g per week 50 61.0% 40 48.8% 54 62.1%＞150g per week 5 6.1% 6 7.3% 5 5.7%p-value (Group comparison) 0.299 (4 g vs 2 g) 0.211 (2 g vs Placebo) 1.000 (4 g vs Placebo)

First Month 0.070None per week 40 79.0% 34 42.0% 26 30.6%≤ 150g per week 36 21.0% 45 55.6% 52 61.2%＞150g per week 5 2 2.5% 7 8.2%p-value (Group comparison) 0.263 (4 g vs 2 g) 0.125 (2 g vs Placebo) 0.051 (4 g vs Placebo)

Second Month 0.695None per week 30 38.0% 35 42.7% 27 32.5%≤ 150g per week 47 59.5% 44 53.7% 52 62.7%＞150g per week 2 2.5% 3 3.7% 4 4.8%p-value (Group comparison) 0.695 (4 g vs 2 g) 0.378 (2 g vs Placebo) 0.631 (4 g vs Placebo)

p-value: Fisher’s exact test

Supplementary Table 4. Changes of serum creatinine level during the study period by treatment groups

Creatinine level (mg/dL)

Omacor® 4 g/day (n=84) Omacor® 2 g/day (n=82) Placebo (n=87) p-value

Mean (SD) Min – Max Mean (SD) Min – Max Mean (SD) Min – Max Omacor 4 g/day

Omacor 2 g/day

Baseline 0.88 (0.24) 0.42 – 2.18 0.86 (0.24) 0.47 – 1.72 0.87 (0.25) 0.46 – 1.71 0.811 0.630

Week 2 0.87 (0.24) 0.43 – 2.31 0.86 (0.25) 0.43 – 1.81 0.88 (0.25) 0.49 – 1.88 0.380 0.396

Change from baseline －0.01 －0.04 – 0.02 (95%CI)

－0.01 －0.03 – 0.01 (95%CI)

0.00 －0.02 – 0.02 (95%CI)

Week 4 0.87 (0.24) 0.51 – 2.22 0.87 (0.27) 0.43 – 1.87 0.87 (0.26) 0.45 – 1.87 0.571 0.407

Change from baseline －0.01 －0.04 – 0.02 (95%CI)

0.01 －0.01 – 0.03 (95%CI)

－0.00 －0.02 – 0.02 (95%CI)

Week 8 0.86 (0.20) 0.48 – 1.47 0.88 (0.27) 0.42 – 1.94 0.88 (0.24) 0.48 – 1.70 0.735 0.252

Change from baseline 0.00 －0.03 – 0.03 (95%CI)

0.03 0.00 – 0.06 (95%CI)

0.01 －0.02 – 0.03 (95%CI)

End of Study 0.89 (0.25) 0.48 – 2.22 0.89 (0.28) 0.42 – 1.94 0.88 (0.24) 0.48 – 1.70 0.918 0.960

Change from baseline 0.00 －0.03 – 0.03 (95%CI)

0.03 0.00 – 0.06 (95%CI)

0.01 －0.02 – 0.03 (95%CI)

Su et al.

288

Supplementary Table 5. Comparisons of lipid-lowering effects among the trials in Taiwan, Japan, and UK at the dose of 4 g/day

Taiwan Japan UK

Geometric Mean

95%CI Mean (SD)

95%CI Mean (SD)

Min – Max

TriglyceridesBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %Week 10, mg/dLChange from baseline, %Week 12, mg/dLChange from baseline, %

351.2243.2－30.7%

251.9－27.8%

238.5－32.1%

————

325.2 – 379.3219.5 – 269.5－36.2 – －24.8%229.7 – 276.2－33.7 – －21.4%216.1 – 263.1－38.0 – －25.6%

————

277.5 (97.3)

——

200 (87.9)－26.9%

212.5 (107.4)－22.5%

205.9 (99.3)－23.4%

208.8 (86.0)－22.9%

————

－29.8 – －24.1%—

－26.6 – －18.4%—

－27.8 – －19.0%—

－26.0 – －19.7%

284.3 (92.1)

——

222.3 (93.0)－21.8%

217.0 (83.3)－23.7%

——

220.5 (98.3)－22.4%

177.1 – 554.5

——

103.6 – 524.4—

92.1 – 430.5———

68.2 – 544.7—

Total cholesterolBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %Week 10, mg/dLChange from baseline, %Week 12, mg/dLChange from baseline, %

184.0176.9－3.9%180.0－1.8%178.3－3.1%

————

177.5 – 190.8169.4 – 184.7－6.0 – －1.8%

172.3 – 188.1－4.3 – 0.8%170.3 – 186.7－5.9 – －0.3%

————

212.0 (30.2)

——

206.2 (32.5)－2.7%

206.1 (32.9)－2.9%

203.5 (31.6)－3.9%

203.9 (31.5)－3.7%

————

－4.0 – －1.5%—

－4.2 – －1.5%—

－5.4 – －2.5%—

－5.0 – －2.4

288.5 (43.7)

——

287.7 (50.3)－0.3%

282.3 (60.7)－0.3%

——

291.2 (54.9)0.9%

205.0 – 380.9

——

193.4 – 406.0—

166.3 – 464.0———

177.9 – 444.7—

HDL-CBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %Week 10, mg/dLChange from baseline, %Week 12, mg/dLChange from baseline, %

37.138.02.3%37.91.9%38.02.4%————

35.8 – 38.536.6 – 39.4－0.5 – 5.2%36.5 – 39.4－1.9 – 5.9%36.4 – 39.7－1.3 – 6.4%

————

45.7 (10.0)

——

47.6 (11.2)4.1%

47.5 (11.6)3.9%

47.4 (11.3)4.2%

47.6 (11.0)4.3%

————

2.6 – 5.6%—

2.2 – 5.6%—

2.5 – 5.9%—

2.8 – 5.8%

41.0 (10.4)

——

42.9 (12.4)4.7%

41.0 (10.8)0.0%——

42.5 (11.2)3.8%

35.9 – 39.1

——

15.5 – 77.3—

19.3 – 69.6———

19.3 – 65.7—

LDL-CBaseline, mg/dLWeek 2, mg/dLChange from baseline, %Week 4, mg/dLChange from baseline, %Week 8, mg/dLChange from baseline, %Week 10, mg/dLChange from baseline, %Week 12, mg/dLChange from baseline, %

80.487.48.7%85.56.3%86.27.2%————

75.3 – 85.981.7 – 93.6

4.1 – 13.6%78.8 – 92.7

0.9 – 12.1%79.7 – 93.2

1.0 – 13.7%————

125.7 (28.5)

——

127.9 (30.3)2.3%

125.2 (31.9)－0.4%

124.2 (29.7)－0.9%

123.6 (29.0)－1.1%

—

—0——

0.1 – 4.39%—

－2.9 – 2.0%—

－3.4 – 1.6%—

－3.35 – 1.2%

174.0 (42.5)

——

191.4 (51.8)10.0%

184.5 (45.6)6.0%——

183.3 (56.1)5.3%

99.8 – 245.9

——

75.8 – 309.4—

83.5 – 304.7———

84.7 – 290.4—

Abbreviations: HDL-C and LDL-C, High-density and Low-density lipoprotein cholesterol

Effects of Omega-3 Polyunsaturated Fatty Acid in Taiwanese Hypertriglyceridemia

289

Supplementary Table 6. Summary of mean high-sensitivity CRP level before and after Omacor treatment

Unit: % Group difference

Omcor 4 g/day

Omacor 2 g/day

Placebo Omacor 4 g/day vs Placebo

Omacor 2 g/day vs Placebo

Omacor 4 g/day vs Omacor 2 g/day

Omacorvs Placebo

BaselineNMean (SD)Median(Min, Max)95% CIp-value (a)

End of StudyNMean (SD)Median(Min, Max)Change95% CIp-value (a)p-value (b)

84

0.2272 (0.2820)0.1130

(0.0120, 1.1830)

790.2438 (0.4548)

0.1030(0.0120, 3.6870)

0.0162－0.0873 ~ 0.1197

0.75620.9939

82

0.1631 (0.1676)0.1070

(0.0120, 0.8700) 82

0.2182 (0.4560)0.0830

(0..0120, 3.1250)0.0550

0.0399 ~ 0.15000.25220.5316

87

0.1652 (0.2475)0.0810

(0.0120, 1.5390)

860.1508 (0.2525)

0.0520(0.0120, 1.4410)－0.0151

－0.0737 ~ 0.04350.60930.1599

0.0620

－0.0097 ~ 0.13360.0897

0.0617－0.0551 ~ 0.1784

0.29940.1331

－0.0020

－0.0741 ~ 0.07010.9556

0.0688－0.0462 ~ 0.1838

0.23970.8708

0.0640

－0.0087 ~ 0.13670.0842

－0.0072－0.1253 ~ 0.1110

0.90510.1845

0.0300

－0.0320 ~ 0.09200.3420

0.0652－0.0345 ~ 0.1649

0.19870.3298

p-value (a): t-test for intergroup comparisonp-value (b): Rank transformation ANOVA for intergroup comparison

Clinical Therapeutics/Volume 40, Number 1, 2018

Efficacy and Safety of Adding Omega-3 Fatty Acidsin Statin-treated Patients with ResidualHypertriglyceridemia: ROMANTIC(Rosuvastatin-OMAcor iN residualhyperTrIglyCeridemia), a Randomized,Double-blind, and Placebo-controlled Trial

Chee Hae Kim, MD1; Kyung Ah Han, MD, PhD2; Jaemyung Yu, MD, PhD3;Sang Hak Lee, MD, PhD4; Hui Kyung Jeon, MD, PhD5; Sang Hyun Kim, MD, PhD6;Seok Yeon Kim, MD, PhD7; Ki Hoon Han, MD, PhD8; Kyungheon Won, MD, PhD7;Dong-Bin Kim, MD, PhD9; Kwang-Jae Lee, MD, PhD10; Kyungwan Min, MD, PhD2;Dong Won Byun, MD, PhD11; Sang-Wook Lim, MD, PhD12; Chul Woo Ahn, MD, PhD13;SeongHwan Kim, MD, PhD14; Young Joon Hong, MD, PhD15; Jidong Sung, MD, PhD16;Seung-Ho Hur, MD, PhD17; Soon Jun Hong, MD, PhD18; Hong-Seok Lim, MD, PhD19;Ie Byung Park, MD, PhD20; In Joo Kim, MD, PhD21; Hyoungwoo Lee, MD, PhD22; andHyo-Soo Kim, MD, PhD1

1Department of Internal Medicine, Seoul National University Hospital, Cardiovascular Centre, Seoul, Korea;2Diabetes Center, Eulji University, Seoul Eulji hospital, Seoul, Korea; 3Kangnam Sacred Heart Hospital, Seoul,Korea; 4Cardiology Division, Yonsei University College of Medicine, Severance Cardiovascular Hospital, Seoul,Korea; 5Department of Internal Medicine, Division of Cardiology, The Catholic University of Korea, Seoul, Korea;6Department of Cardiology, Boramae Medical Center, Seoul National University College of Medicine, Seoul, Korea;7Department of Internal Medicine, Seoul Medical Center, Seoul, Korea; 8Departments of Cardiology, IschemicHeart Disease Center, Asan Medical Center Heart Institute, Seoul, Korea; 9Department of Cardiology, CatholicUniversity of Korea College of Medicine, Seoul, Korea; 10Department of Endocrinology, Daedong Hospital, Seoul,Korea; 11Endocrinology and Metabolism, Soonchunhyang University Hospital, Seoul, Korea; 12Department ofCardiology, Bundang Cha General Hospital, Seoul, Korea; 13Department of Internal Medicine, Gangnam SeveranceHospital, Seoul, Korea; 14Division of Cardiology, Department of Medicine, Korea University Ansan Hospital, Seoul,Korea; 15Heart Center of Chonnam National University Hospital, Chonnam National University, Seoul, Korea;16Division of Cardiology, Cardiac and Vascular Center, Samsung Medical Center, Seoul, Korea; 17Department ofCardiology, Keimyung University Dongsan Hospital, Daegu, Korea; 18Korea University Medical Center, KoreaUniversity, Seoul, Korea; 19Department of Cardiology, Ajou University School of Medicine, Suwon, Korea;20Division of Endocrinology and Metabolism, Department of Internal Medicine, Gachon University Gil MedicalCenter, Incheon, Korea; 21Department of Internal Medicine, Pusan National University College of Medicine, Busan,Korea; and 22Department of Internal Medicine, Yeungnam University College of Medicine, Daegu, Korea

Accepted for publication November 14, 2017.https://doi.org/10.1016/j.clinthera.2017.11.0070149-2918/$ - see front matter

& 2018 The Authors. Published by Elsevier HS Journals, Inc. This is anopen access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

ABSTRACT

Purpose: The purpose of this study was to examinethe efficacy and safety of adding ω-3 fatty acids torosuvastatin in patients with residual hypertriglyceridemiadespite statin treatment.

January 2018 83

Clinical Therapeutics

Methods: This study was a multicenter, random-ized, double-blind, placebo-controlled study. After a4-week run-in period of rosuvastatin treatment, thepatients who had residual hypertriglyceridemia wererandomized to receive rosuvastatin 20 mg/d plus ω-3fatty acids 4 g/d (ROSUMEGA group) or rosuvastatin20 mg/d (rosuvastatin group) with a 1:1 ratio andwere prescribed each medication for 8 weeks.

Findings: A total of 201 patients were analyzed(mean [SD] age, 58.1 [10.7] years; 62.7% male). After8 weeks of treatment, the percentage change frombaseline in triglycerides (TGs) and non–HDL-C wassignificantly greater in the ROSUMEGA group than inthe rosuvastatin group (TGs: −26.3% vs −11.4%,P o 0.001; non–HDL-C: −10.7% vs −2.2%,P ¼ 0.001). In the linear regression analysis, thelipid-lowering effect of ω-3 fatty acids was greaterwhen baseline TG or non−HDL-C levels were highand body mass index was low. The incidence ofadverse events was not significantly different betweenthe 2 groups.

Implications: In patients with residual hypertrigly-ceridemia despite statin treatment, a combination ofω-3 fatty acids and rosuvastatin produced a greaterreduction of TGs and non−HDL-C than rosuvastatinalone. Further study is needed to determine whetherthe advantages of this lipid profile of ω-3 fatty acidsactually leads to the prevention of cardiovascularevent. ClinicalTrials.gov identifier: NCT03026933.(Clin Ther. 2018;40:83–94) & 2018 The Authors.Published by Elsevier HS Journals, Inc.

Key words: combination, hypertriglyceridemia,non–HDL-C, ω-3 fatty acids, rosuvastatin,triglycerides.

INTRODUCTIONControl of blood cholesterol levels apparently reducesatherosclerotic cardiovascular disease.1 The firstrecommended therapy for dyslipidemia is statins,which effectively prevents cardiovascular disease bylowering LDL-C levels. However, hypertriglyceridemiais also well known as an independent risk factorassociated with cardiovascular events.2–4 Statins arenot effective at lowering triglycerides (TGs), whichpartly explains the reason why cardiovascular eventsoccur even with the use of statins. Therefore, inpatients with mixed dyslipidemia, controlling TG levels

84

in addition to lowering LDL-C levels should beconsidered.

Treatment options to lower TG levels are fibrates,niacin, and ω-3 fatty acids.5 Among these, fibratesand niacin are associated with tolerability problems.Contrariwise, ω-3 fatty acids have proven itsTG-lowering effect with good tolerability.6 However,there are mild LDL-C–increasing effects in ω-3fatty acids.7 For appropriate combination therapyof statin and ω-3 fatty acids, further studies areneeded.

Previous studies have found the efficacy of combin-ing ω-3 fatty acids with several statins on controllingTG levels.8–10 However, the efficacy and tolerabilityof the combination of ω-3 fatty acids and rosuvastatin,which is the most potent statin currently used, havenot yet been proven. This Phase III study aimed toexamine the efficacy and safety of the combinationof ω-3 fatty acids and rosuvastatin compared withrosuvastatin alone in patients with residual hyper-triglyceridemia despite statin treatment.

METHODSStudy Design

The study was an 8-week, prospective, randomized,double-blind, parallel group, Phase III multicenter trialconducted in 33 centers in South Korea. The studyperiod was from June 18, 2014, through March 31,2016.

Patients with hypercholesterolemia at high risk forcardiovascular disease according to the NationalCholesterol Education Program (NCEP): Adult Treat-ment Panel III (ATP III) were screened.11 To be eligiblein first screening, participants were required to meetthe following criteria: (1) age from 19 to 80 years, (2)fasting TG level ≥300 mg/dL and LDL-C level ≥100mg/dL and o160 mg/dL for individuals who were nottaking statins for 4 weeks, (3) TG level ≥200 mg/dLand o500 mg/dL, and LDL-C level o110 mg/dL forindividuals who were taking statins for last 4 weeks,and (4) nonsmoking during the study period. Theneligible participants underwent a 4-week run-inperiod. During the run-in period, all participantsreceived 20 mg/d of open-label rosuvastatin calciumand discontinued use of other lipid-lowering agents.After the run-in period, the levels of LDL-C and TGswere measured repeatedly. To be eligible in the secondscreening, participants were required to meet the

Volume 40 Number 1

ScreeningTreatment

(run-in)

Day -35 to -28 Day 1 Day 29 ± 3 Day 57 ± 3

ROSUMEGA (KI1107) 4 capsules + placebo (rosuvastatin 20 mg) once

Placebo (ROSUMEGA (KI1107)) 4 capsules + (rosuvastatin 20 mg) once

Rosuvastatin 20 mg

Randomization Treatment visit Last visitScreening visit

200 mg/dL ≤ TG < 500 mg/dLLDL-C < 110 mg/dLNo treatment with statin for 4 weeks: reduction of LDL-C comparing screening> 80% Adherence

Figure 1. Study protocol. After a 4-week run-inperiod with rosuvastatin 20 mg, eligi-ble patients were randomized to therosuvastatin 20 mg/d plus ω-3 fattyacids 4 g/d (ROSUMEGA) or rosuvas-tatin group. Lipid and lipoproteinlevels and adverse events were evalu-ated at 8 weeks after treatment. TG ¼triglyceride.

C.H. Kim et al.

following criteria: (1) residual hypertriglyceridemiawith fasting TG level ≥200 mg/dL and o500 mg/dL, (2) well-controlled LDL-C level of o110 mg/dL,(3) reduction of LDL-C levels during the run-in periodcomparing first screening if individuals were nottaking statins before, and (4) those who had anadherence rate of ≥80% for medication during therun-in period. The exclusion criteria included (1)history of unstable angina, acute myocardialinfarction, coronary artery revascularization, includingcoronary artery bypass surgery, transient ischemicattack, or stroke 3 months before screening; (2)history of operation for aortic aneurysm within 6months before screening; (3) symptom of unexplainedmyalgia or a diagnosis of myalgia or rhabdomyolysis atscreening; (4) history of pancreatitis before screening;(5) uncontrolled hypertension; (6) serum creatinine level≥2 times the upper limit of normal; (6) alanineaminotransferase and/or aspartate aminotransferase≥3 times the upper limit of normal; (7) creatininephosphokinase levels 45 times the upper limit ofnormal; (8) genetic disorder, including galactoseintolerance, Lapp lactase deficiency, or glucose-galactose malabsorption; (9) history of positiveantibody HIV-1 or HIV test result; (10) history ofmalignant tumor within 5 years; and (11) the use ofprohibited concomitant medications. The eligiblepatients were randomly assigned with a 1:1 ratio to 2groups and prescribed ROSUMEGA or rosuvastatin for8 weeks. Individuals in the ROSUMEGA group wereprescribed 4 capsules of ω-3 fatty acids 1 g plusrosuvastatin calcium 5 mg and 1 tablet of placebo ofrosuvastatin 20 mg/d. Each 1 g of ω-3 fatty acidscontains 380 mg of docosahexaenoic acid and 460 mgof eicosapentaenoic acid. The rosuvastatin groupreceived 4 capsules of placebo of ROSUMEGA and 1tablet of rosuvastatin calcium 20 mg (Figure 1). On thebasis of baseline data collected, an individual’s 10-yearrisk of coronary heart disease was calculated accordingto NCEP ATP III.11 The study protocol was approvedby the institutional review board or ethics committee ateach participating center, and all patients providedwritten informed consent.

Efficacy and Tolerability AssessmentsThe efficacy end points were the percentage change

of lipid and lipoprotein levels, including TGs, non–HDL-C, total cholesterol, LDL-C, HDL-C, VLDL-C,

January 2018

and apolipoprotein A1 and B (Apo A1 and Apo B)after 8 weeks from baseline.

Subgroup analyses were performed according todiabetes mellitus (DM), chronic kidney disease (CKD),age (≥65 vs o65 years), and sex (male vs female) todetermine whether the effects of ω-3 fatty acidsdiffered in each subgroup. The study participantswere classified into the DM group if they had aprevious diagnosis of DM or who were currentlytaking oral hypoglycemic agents or insulins. Thedefinition of CKD was a glomerular filtration rateo60 mL/min per 1.73 m2.12

Tolerability was assessed by monitoring of adverseevents and performing physical examinations andlaboratory tests, including serum chemical analysesand urinalysis. Adverse events included all unintendedconsequences of the individuals receiving treat-ment regardless of causality. Adverse drug reactionmeant a harmful, unintended reaction, and thecausal relationship with the treatment cannot beexcluded. Adverse events were categorized asdefinitely related, probably related, possibly related,probably not related, and definitely not related to thestudy drug.

Statistical AnalysisContinuous variables were expressed as mean (SD)

or mean (SE) as appropriate and compared with the

85

Clinical Therapeutics

2-sample t test or Wilcoxon rank sum test accordingto normal distribution. Categorical variables wereexpressed as frequencies and percentages, and the χ2

test was used for comparison. The differences ofpercentage changes in TG and other cholesterol valuesbetween the treatment groups were comparedusing the 2-sample t test if normally distributed andthe Wilcoxon rank sum test if not normallydistributed. To identify the clinical factors associatedwith the greater effect of ω-3 fatty acids in loweringTG and non–HDL-C levels, linear regression analyseswere performed and the prediction model wasobtained.

The efficacy analyses were performed using the fullanalysis set population and the safety analyses withthe safety set population. Two-sided P o 0.05 wasconsidered statistically significant. All analyses wereperformed using SAS, version 9.3 (SAS Institute, Cary,North Carolina) and SPSS, version 22.0 (IBM Co,Armonk, New York).

RESULTSBaseline Characteristics

Of the 750 patients who were screened, 469patients entered the run-in period, and 215 patientswere randomly assigned to treatment groups. A totalof 104 patients were assigned to the ROSUMEGAgroup and 111 patients to the rosuvastatin group.From the randomized population, 1 participant ineach group were excluded from the safety set becausethey were not actually treated. Then the equal numberof 6 patients in both groups was excluded from fullanalysis set population because of violation fromeligibility criteria or inadequate laboratory test. Fi-nally, 201 patients were analyzed for evaluatingefficacy end points (Figure 2).

Baseline characteristics are summarized inTable I. Clinical characteristics were similar betweenthe 2 groups except for age, with older subjects in theROSUMEGA group than in the rosuvastatingroup (mean [SD] age, 59.7 [10.8] vs 56.6 [10.5]years; P ¼ 0.040). The mean (SD) 10-year risk scorefor coronary heart disease was 10.5% (6.9%) in theROSUMEGA group and 8.8% (6.5%) in therosuvastatin group (P ¼ 0.085). There was nodifference in the proportion of patients whopreviously used statins between the 2 groups (92.8%vs 95.2%, P ¼ 0.471).

86

Efficacy End PointsTable II summarizes the results for the changes of

lipid and lipoprotein variables. The percentage changeat 8 weeks from baseline in TG levels was significantlygreater with the ROSUMEGA group compared withthe rosuvastatin group (−26.3% vs −11.4%, P o0.001). There was also a greater reduction of non–HDL-C levels in the ROSUMEGA group than in therosuvastatin group (−10.7% vs −2.2%, P ¼ 0.001).Among other lipid parameters, total cholesterol,VLDL-C, Apo A1, and Apo B also had a greaterdecrease in the ROSUMEGA group than in therosuvastatin group after 8 weeks of treatment(P o 0.05 for each). Meanwhile, LDL-C and HDL-Clevels slightly increased after 8 weeks of treatment inboth groups, but the difference between the groups wasnot statistically significant (LDL-C: 1.8% vs 4.3%,P ¼ 0.335; HDL-C: 0.9% vs 2.8%, P ¼ 0.377).

Subgroup Analysis of EfficacyWe performed subgroup analyses in changes of TG

and non–HDL-C levels according to DM, CKD, sex,and age (≥65 vs o65 years). Regardless of thepresence of DM, the ROSUMEGA group had agreater reduction in TG levels after 8 weeks comparedwith the rosuvastatin group (patients with DM:−25.4% vs −9.4%, P ¼ 0.002; patients withoutDM: −29.5% vs −16.8%, P ¼ 0.045). This tendencyof additional TG-lowering effects of ROSUMEGAwas present in both of insulin-dependent DM andnon–insulin-dependent DM (insulin-dependent DM:−27.6% vs −14.1%, P ¼ 0.224; non–insulin-depend-ent DM: −24.6% vs −9.4%, P ¼ 0.011). Likewise,regardless of CKD, sex, and age groups, greaterreductions in TG levels were achieved by the ROSU-MEGA group compared with the rosuvastatin group(Figure 3).

In the case of non–HDL-C, the effect of ROSU-MEGA on reducing the levels of non–HDL-C after 8weeks was different by subgroups. In patients withDM, ROSUMEGA had a greater lowering effect onnon–HDL-C than rosuvastatin (−12.6% vs −1.6%,P ¼ 0.002), but in patients without DM, ROSUMEGAhad similar effects on non-HDL-C as rosuvastatin(−4.3% vs −3.8%, P ¼ 0.250). Among patients withDM, ROSUMEGA tended to decrease non–HDL-Clevels more than rosuvastatin, regardless of insulindependency (insulin-dependent DM: −19.4% vsþ6.8%, P ¼ 0.018; non–insulin-dependent DM:

Volume 40 Number 1

Screening

First screening failure (N = 281)

Second screening failure (n = 254)

Violation from inclusion and exclusion (n=269)

Violation from inclusion and exclusion (n = 239)

Violation from inclusion and exclusion (n=5)

Not assessed non–HDL-C (n = 2)

Withdrawal consent (n = 10)

Withdrawal consent (n = 11)

Others (n = 2)

Follow-up loss (n = 1)

Adverse event (n = 3)

Randomized

ROSUMEGA Rosuvastatin

(n = 215)

(n = 104)

Exclusion from safety set (n = 1)

Not treated (n = 1)

Exclusion from safety set (n = 1)

Exclusion from FAS (n* = 6)

Violation from inclusion and exclusion (n=5)

Not assessed non–HDL-C (n = 1)

Exclusion from FAS (n=6)

Not treated (n = 1)

(n = 111)

ROSUMEGA Rosuvastatin(n = 103) (n = 110)

Run-in

(n = 469)

(N = 750)

Safety set

(n = 213)

ROSUMEGA Rosuvastatin

(n = 97) (n = 104)

(n = 201)

FAS

Figure 2. Study population. FAS ¼ full analysis set; ROSUMEGA ¼ rosuvastatin 20 mg/d plus ω-3 fatty acids4 g/d. *Double counting.

C.H. Kim et al.

−11.2% vs −4.0%, P ¼ 0.169). In the subgroups ofnormal renal function, female, and elderly (465 yearsold), ROSUMEGA had a greater lowering effect onnon–HDL-C than rosuvastatin alone (Figure 4).

Clinical Factors That Affect Lipid-lowering Effectsof ω-3 Fatty Acids

Multiple linear regression analyses were performedto find the clinical factors associated with the greater

January 2018

effect of ω-3 fatty acids in lowering TG and non–HDL-C levels. In the case of TGs, the higher thebaseline TG level and the lower the body mass index(BMI), the greater the decrease in TGs obtained byadding ω-3 fatty acids. However, DM, age, CKD,and sex did not significantly affect TG reduction(Table III).

Similar to TGs, high baseline non–HDL-Clevels and low BMI were associated with a large

87

Table I. Baseline characteristics.*

Characteristic ROSUMEGA (n ¼ 97) Rosuvastatin (n ¼ 104) P

Age, mean (SD), y 59.7 (10.8) 56.6 (10.5) 0.040Male 59 (60.8) 67 (64.4) 0.598Height, mean (SD), cm 162.7 (9.3) 163.1 (8.6) 0.767Weight, mean (SD), kg 72.5 (12.0) 73.5 (12.4) 0.586BMI, mean (SD), kg/m2 27.4 (3.7) 27.6 (3.6) 0.410Hypertension 75 (77.3) 79 (76.0) 0.820Diabetes mellitus 75 (77.3) 75 (72.1) 0.397Insulin-dependent diabetes mellitus 13 (13.4) 17 (16.3) 0.558Chronic kidney disease 14 (14.4) 15 (14.7) 0.957Current smoker 25 (25.8) 24 (23.1) 0.656Alcohol drinking 44 (45.4) 50 (48.1) 0.884History of coronary artery disease 36 (37.1) 48 (46.2) 0.194History of cerebrovascular disease 5 (5.2) 7 (6.7) 0.63710-Year risk score for coronary heart disease 10.5 (6.9) 8.8 (6.5) 0.085Previous use of statin 90 (92.7) 99 (95.2) 0.471

BMI ¼ body mass index; ROSUMEGA ¼ rosuvastatin 20 mg/d plus ω-3 fatty acids 4 g/d.⁎Data are presented as number (percentage) of patients unless otherwise indicated.

Clinical Therapeutics

non–HDL-C decrease. In addition, the effect ofω-3 fatty acids on reducing non–HDL-C wasgreater in patients with DM than in patients without

Table II. Lipid and lipoprotein levels at baseline and 8 w

Variable

ROSUMEGA (n ¼ 97)

Baseline,Mean (SD),

mg/dL

Week 8,Mean (SD),

mg/dL

PercentChange,Mean(SEM)

M

Triglycerides 284.0 (68.6) 205.9 (91.4) −26.3 (3.1) 2Non–HDL-C 99.1 (23.7) 86.0 (25.4) −10.7 (2.9)Total cholesterol 141.2 (24.5) 128.3 (27.2) −8.1 (1.9) 1LDL-C 61.9 (19.6) 61.5 (22.2) 1.8 (3.1)HDL-C 42.1 (7.5) 42.3 (8.8) 0.9 (1.5)VLDL-C 37.2 (16.4) 24.6 (15.1) −28.5 (4.4)Apolipoprotein

A1140.1 (23.0) 133.8 (24.5) −1.5 (4.1) 1

ApolipoproteinB

75.4 (20.5) 71.1 (18.3) −3.4 (2.5)

88

DM. Age, CKD, and sex did not significantlyaffect non–HDL-C reduction by ω-3 fatty acids(Table IV).

eeks after treatment.

Rosuvastatin (n ¼ 104)

Baseline,ean (SD),mg/dL

Week 8,Mean (SD),

mg/dL

PercentChange,Mean(SEM)

P for PercentChange Between

Groups

79.6 (64.2) 241.7 (97.7) −11.4 (3.4) o0.00196.7 (22.0) 94.1 (30.7) −2.2 (2.5) 0.00139.4 (24.0) 137.2 (31.9) −1.2 (1.7) o0.00162.5 (18.4) 64.7 (25.5) 4.3 (2.9) 0.33542.6 (10.1) 43.1 (8.9) 2.8 (1.6) 0.37734.2 (13.4) 29.4 (19.5) −12.2 (4.9) 0.00439.9 (23.8) 138.0 (22.7) −0.5 (1.2) 0.009

75.7 (16.5) 75.3 (20.4) 0.3 (2.0) 0.049

Volume 40 Number 1

A B

C D

TG Change by DM

TG Change by Sex TG Change by Age

DM

Male Female ≥65years <65years

Non-DM CKD Non-CKD

TG Change by CKD

-25.4 -9.4 -29.5 -28.5 -20.3-16.8RRm

-25.2 -13.0

RRm

-28.1 -8.7

RRm

-27.5 -7.5

RRm

-25.6 -12.9

RRm

R RRm Rm-25.9 -9.5

RRm

0.0

Cha

nge

(%)

Cha

nge

(%)

Cha

nge

(%)

Cha

nge

(%)

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

-30.0

-35.0

-40.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

-30.0

-35.0

-40.0

-5.0

-10.0

-15.0

-20.0

-25.0

-30.0

-35.0

-40.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

-30.0

-35.0

-40.0n = 75

n = 59 n = 67 n = 38 n = 37 n = 34 n = 28 n = 63 n = 76

n = 75 n = 22 n = 29 n = 14 n = 15 n = 83 n = 87

P = 0.002∫ P = 0.045∫ P = 0.276§ P < 0.001∫

P = 0.014∫ P = 0.003∫ P = 0.004∫ P = 0.013∫

Figure 3. Subgroup analyses in percentage changes of triglyceride (TG) levels after 8 weeks. There was agreater reduction of TG levels in the rosuvastatin 20 mg/d plus ω-3 fatty acids 4 g/d.(ROSUMEGA) group than in the rosuvastatin group regardless of diabetes mellitus (DM), chronickidney disease (CKD), sex, and age.

C.H. Kim et al.

Safety analysesThere was no significant difference between groups

in adverse events (15.5% in the ROSUMEGA groupvs 17.3% in the rosuvastatin group, P ¼ 0.732).Serious adverse events occurred in 2 of 103 individ-uals (1.9%) in the ROSUMEGA group and 2 of 110(1.8%) in the rosuvastatin group. In particular, the 2serious adverse events in the ROSUMEGA group wereovarian cancer and parkinsonism, and the 2 seriousadverse events in the rosuvastatin group were breastcancer and tarsal tunnel syndrome. None ofthese were considered related to study treatment.There were no significant increases in creatininephosphokinase levels (4 5.0 times the upper limitof normal) or alanine aminotransferase levels(43.0 times the upper limit of normal) in bothgroups. One patient in the ROSUMEGA grouphad mild significant aspartate aminotransferaseelevation (43.0 times the upper limit of normal)

January 2018

but o5 times the upper limit of normal. No adversedrug reactions were observed in the rosuvastatingroup.

DISCUSSIONThis study found that a combination of ω-3 fatty acidsand rosuvastatin in patients with residual hypertrigly-ceridemia achieved a greater reduction in TG, non–HDL-C, and other lipid and lipoprotein levels thanrosuvastatin alone did. In addition, the higherthe baseline TG and non-HDL levels and the lowerthe BMI, the better the effect of ROSUMEGA, and theeffect of lowering non–HDL-C levels was more no-ticeable in patients with DM. Finally, there was nosignificant adverse event according to adding 4 g/d ofω-3 fatty acids on rosuvastatin.

The most important treatment in patients withdyslipidemia is statins, which mainly lowers LDL-C

89

A B

C D

Non–HDL-C Change by DM

Non–HDL-C Change by Sex Non–HDL-C Change by Age

Non–HDL-C Change by CKD

DM

Rm-12.6

-7.6 -5.2 -15.5

3.1

-18.3

2.0

-6.6 -3.7

-1.6 -4.3 -3.8 -12.7 -10.6 -10.4 -0.4

R

Rm R Rm R Rm R Rm R

Rm R Rm R Rm R

Non-DM CKD Non-CKD5.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

10.0

5.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

10.0

5.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

Cha

nge

(%)

Cha

nge

(%)

Cha

nge

(%)

Cha

nge

(%)

5.0

0.0

-5.0

-10.0

-15.0

-20.0

-25.0

n = 34 n = 28

P = 0.002∫

n = 59 n = 67

P = 0.237∫

n = 38 n = 37 n = 34 n = 28 n = 63 n = 76

P < 0.001∫ P = 0.001∫ P = 0.124∫

n = 63 n = 76

P = 0.250∫

n = 14 n = 15 n = 83 n = 87

P = 0.719§ P < 0.001∫

Male Female ≥65years <65years

Figure 4. Subgroup analyses in percentage changes of non–HDL-C levels after 8 weeks. rosuvastatin 20 mg/dplus ω-3 fatty acids 4 g/d (ROSUMEGA) found a greater lowering effect on non–HDL-C thanrosuvastatin alone in subgroups of patients with diabetes mellitus (DM), patients without chronickidney disease (CKD), female patients, and elderly patients.

Clinical Therapeutics

levels. However, statins do not control TG levelseffectively. TG levels are classified as normal (o150mg/dL), borderline (150-199 mg/dL), high (200-499mg/dL), and very high (≥500 mg/dL), and hyper-glyceridemia is defined as TG levels ≥200 mg/dL,generally. According to Third National Health andNutrition Examination Survey (NHANES III) in theUnited States, 35% of men and 25% of women havehypertriglyceridemia (TG ≥150 mg/dL).13 In Korea,the prevalence of hypertriglyceridemia was 38% inmen and 20% in women.14 In patients with DM,the prevalence of hypertriglyceridemia increases,reaching 49%.14 Hypertriglyceridemia is alsoknown as an independent risk factor associated withcardiovascular event.2–4 TG is a major component ofTG-rich lipoproteins, including VLDL-C andchylomicrons. A recent study suggested that TG isalso an independent risk factor for ischemic heartdisease.15 Even non–HDL-C levels were claimed to be

90

a better predictor of cardiovascular risk thanLDL-C.16 Considering the high prevalence ofhypertriglyceridemia and subsequent independentrisk for cardiovascular event, TG should also beactively considered in the treatment of dyslipidemia.

Metabolic consequences of hypertriglyceridemiaare not well established yet. However, increased TG-rich lipoproteins, a result of hypertriglyceridemia,directly influence composition and metabolism oflipoproteins, making smaller and dense HDL-C andLDL-C particles.17 This disadvantageously affectscardiovascular disease because hypertriglyceridemicHDL-C (small, dense HDL-C) loses its function todeliver cholesteryl esters18 and small and denseLDL-C is more susceptible to oxidative modification.19

In addition, TG-rich lipoproteins have atherogeniceffects by inducing macrophage dysfunction,endothelial cell inflammation, and coagulationabnormality.17

Volume 40 Number 1

Table III. Predictive model for reduction in TGs by ω-3 fatty acids.*

Variable Reduction of TGs, mg/dL (95% CI) P

Higher baseline TG level (by 10-mg/dL increment) 5.4 (2.8 to 8.0) o0.001Lower BMI (by 1-kg/m2 decrement) 5.3 (0.2 to 10.4) 0.041Non-DM (vs DM) 14.3 (−28.7 to 57.3) 0.510Younger age (by 10-year decrement) 8.9 (−10.3 to 28.1) 0.360CKD (vs non-CKD) 32.6 (−24.1 to 89.3) 0.256Female (vs. male) 27.0 (−13.0 to 66.9) 0.183

BMI ¼ body mass index, CKD ¼ chronic kidney disease, DM ¼ diabetes mellitus, TG ¼ triglyceride.⁎TG level reduction after adding ω-3 fatty acids compared with rosuvastatin treatment during the run-in period wasestimated by linear regression analysis. The coefficient of determination (R2) of the model was 21.3% (P ¼ 0.001).

C.H. Kim et al.

Treatment options for reducing TG levels includeω-3 fatty acids, niacin, and fibrates. However, niacincauses flushing, liver toxic effects, and myopathy andhas the risk to increase serum glucose, especially whencoadministered with a statin.20 Fibrates also has a riskof myopathy and the potential to blunt the beneficialeffect of statins.21 Therefore, only ω-3 fatty acids areavailable to use for reducing TG levels withoutburdens of additional adverse effects.

Historically, ω-3 fatty acids suggested its benefit inobservational studies, reporting that intake of ω-3fatty acid-enriched food reduced cardiovascularrisk.22 The TG-lowering mechanisms of ω-3 fattyacids are direct inhibition of TG synthesis, reductionof hepatic synthesis of VLDL-Apo B, stimulation offatty acid oxidation, and enhancing TG clearance withincreased plasma lipolytic activity.23 Currently used

Table IV. Predictive model for reduction in non–HDL-C

Variable R

Higher baseline non–HDL-C (by 10-mg/dL increment)Lower BMI (by 1-kg/m2 decrement)DM (vs non-DM)Younger age (by 10-year decrement)CKD (vs non-CKD)Female (vs male)

BMI ¼ body mass index, CKD ¼ chronic kidney disease, DM ¼⁎Non–HDL-C level reduction after adding ω-3 fatty acids comparestimated by linear regression analysis. The coefficient of deter

January 2018

concentrated forms of ω-3 fatty acids includeeicosapentaenoic acid and docosahexaenoic acid,which have similar TG reduction ranges to eachother (eicosapentaenoic acid: þ1.8% to −34.9%;docosahexaenoic acid: −8.0% to −43.7%).24 TG-lowering effects of ω-3 fatty acids are dependent onbaseline TG levels. The higher the baseline TG level,the greater the TG-lowering effect of ω-3 fatty acids.In previous randomized controlled trials, 2 and 4 g/dof ω-3 fatty acids had a reduction of TG levels of 20%to 26% and 31% to 33%, respectively, in patientswith severe hyperglyceridemia of TG levels ≥500mg/dL.25,26 In case of hyperglyceridemia with a TGlevel o500 mg/dL, several studies have compared acombination of ω-3 fatty acids and statin with statinalone. A combination of ω-3 fatty acids 4 mg/d withsimvastatin 40 mg/d for 8 weeks produced a greater

by ω-3 fatty acids.*

eduction of Non–HDL-C, mg/dL (95% CI) P

5.7 (3.8 to 7.7) o0.0011.6 (0.3 to 2.9) 0.015

14.3 (3.2 to 25.5) 0.0121.3 (−3.6 to 6.2) 0.5977.0 (−7.4 to 21.5) 0.3375.1 (−5.1 to 15.3) 0.322

diabetes mellitus.ed with rosuvastatin treatment during the run-in period wasmination (R2) of the model was 33.5% (P o 0.001).

91

Clinical Therapeutics

reduction in TG levels by 30% and non–HDL-C levelsby 9%, whereas simvastatin alone reduces TG levelsby 6% and non–HDL-C levels by 2%.8 In anotherrandomized study, ω-3 fatty acids 4 g/d combinedwith several kinds of statins also reduced TG levels bya greater amount than statin alone (18% vs 6%).27 Inour study, the percentage changes from baseline TGand non–HDL-C were significantly greater in theROSUMEGA group than in the rosuvastatin group(TG: −26.3% vs −11.4%, P o 0.001; non–HDL-C:−10.7% vs −2.2%, P ¼ 0.001), which are consistentwith the results of previous studies. These resultssuggest that ω-3 fatty acids have definitely additiveand complementary effects on TG and non–HDL-Ccontrol with different mechanisms. As in our study,LDL-C levels were mildly increased in previousstudies of ω-3 fatty acids. ω-3 fatty acids reduceVLDL-C secretion from liver; as a result, the levelof LDL-C is slightly increased by the process ofconverting VLDL-C to LDL-C.28 Although LDL-Cincreases, it is less harmful because the larger andless atherosclerotic LDL-C subcomponent mainlyincreases.23

In subgroup analyses, we can notice that additiveTG-lowering effects of ω-3 fatty acids are observedregardless of DM, CKD, sex, and age. Therefore, ω-3fatty acids can be applied to any clinical setting ofhypertriglyceridemia. However, in terms of non–HDL-C, a combination of ω-3 fatty acids withrosuvastatin had different effects in each subgroupcompared with rosuvastatin alone. There was amore prominent non–HDL-C–lowering effect ofROSUMEGA in patients with DM, normal renalfunction, female, and older age. In addition, furtherTG and non–HDL-C–lowering effects of ω-3 fattyacids in DM were observed regardless of insulindependency.

In multiple linear regression analyses, the effect ofω-3 fatty acids on TG reduction was greater withhigher baseline TG levels and lower BMI. Similarly, innon–HDL-C, the higher the baseline non–HDL-Clevel and the lower the BMI, the greater the non–HDL-C reduction. Previous studies have found thatthe higher the baseline lipid level, the greater thereduction in ω-3 fatty acids.29 In addition, our resultssuggest that lowering weight and maintainingadequate BMI may be helpful in controlling TG andnon–HDL-C levels. Interestingly, the effect of ω-3fatty acids in lowering non–HDL-C levels is greater

92

in patients with DM, although TG reduction isnot affected by the presence of DM. Non–HDL-Cis a more inclusive measure of all atherogenicApo B–containing lipoproteins: VLDL-C, IDL-C,chylomicron remnants, lipoprotein A, and LDL-C.30

Non–HDL-C is known to be particularly elevated inpatients with DM31 and serves as a strong predictor ofcardiovascular disease in patients with DM.32

Therefore, our results, which report the obviousnon–HDL-C–reducing effect of ROSUMEGA inpatients with DM, emphasize that the addition ofω-3 fatty acids should be considered in patients withDM.

Rosuvastatin has proven to be more potent thanany other statins in many randomized controlledtrials.33–35 However, the combination of rosuvastatinand ω-3 fatty acids has not been previously studied.Our study was the first to examine the effect of addingω-3 fatty acids to rosuvastatin and proved an additiveeffect of adding ω-3 fatty acids to rosuvastatin onreducing TG levels, without any significant adverseevent.

However, the present study has several limitations.First, the duration of treatment was short. Additionallong-term studies on the efficacy and tolerability ofω-3 fatty acids are needed. It is also necessary to studywhether the reduction in TG levels using ω-3fatty acids is preventing cardiovascular events.Nevertheless, this study proved that even short-termtreatment with ω-3 fatty acids effectively reducedTG and non–HDL-C levels. Second, the study pop-ulation was exclusively middle-aged Asians, whichlimits the generalizability of the results. Third, therewere age differences in baseline characteristics be-tween the ROSUMEGA and rosuvastatin groups.However, in the subgroup analysis, ROSUMEGAhad a tendency to lower TG and non–HDL-C levelsmore than rosuvastatin regardless of age group.In addition, age was not a significant indepen-dent variable in multiple linear regression analysesof the ROSUMEGA effect, so the age differencebetween the 2 study groups was determined to beacceptable.

CONCLUSIONSIn patients with residual hypertriglyceridemia despitestatin treatment, a combination of ω-3 fatty acids androsuvastatin decreases TG and non–HDL-C levels to a

Volume 40 Number 1

C.H. Kim et al.

greater extent than rosuvastatin alone. Further studyis needed to determine whether the advantages of thislipid profile of ω-3 fatty acids actually leads to theprevention of cardiovascular event.

ACKNOWLEDGMENTSHyo-Soo Kim contributed to the study conception anddesign with Sang Hak Lee, Hui Kyung Jeon, SangHyun Kim, Ki Hoon Han, Young Joon Hong, andJidong Sung. All the authors, except Chee Hae Kim,contributed to data collection and interpretation ofdata. All authors confirmed the final version of thearticle. Chee Hae Kim and Hyo-Soo Kim contributedto writing, analysis, and interpretation of data.

FUNDING SOURCESThis study was funded by the Kuhnil PharmaceuticalCo Ltd.

CONFLICTS OF INTERESTThe authors have indicated that they have no conflictsof interest regarding the content of this article.

REFERENCES1. Ray KK, Kastelein JJ, Boekholdt SM, et al. The ACC/AHA

2013 guideline on the treatment of blood cholesterol toreduce atherosclerotic cardiovascular disease risk inadults: the good the bad and the uncertain: a comparisonwith ESC/EAS guidelines for the management of dyslipi-daemias 2011. Eur Heart J. 2014;35:960–968.

2. Freiberg JJ, Tybjærg-Hansen A, Jensen JS, NordestgaardBG. Nonfasting triglycerides and risk of ischemic stroke inthe general population. JAMA. 2008;300:2142–2152.

3. Labreuche J, Touboul P-J, Amarenco P. Plasma triglyceridelevels and risk of stroke and carotid atherosclerosis: asystematic review of the epidemiological studies. Athero-sclerosis. 2009;203:331–345.

4. Sarwar N, Danesh J, Eiriksdottir G, et al. Triglycerides andthe risk of coronary heart disease 10,158 incident casesamong 262,525 participants in 29 western prospectivestudies. Circulation. 2007;115:450–458.

5. Chapman MJ, Ginsberg HN, Amarenco P, et al, EuropeanAtherosclerosis Society Consensus Panel. Triglyceride-richlipoproteins and high-density lipoprotein cholesterol inpatients at high risk of cardiovascular disease: evidenceand guidance for management. Eur Heart J. 2011;32:1345–1361.

January 2018

6. Harris WS, Ginsberg HN, Arunakul N, et al. Safety andefficacy of Omacor in severe hypertriglyceridemia.J Cardiovasc Risk. 1997;4:385–391.

7. Maki KC, Lawless AL, Kelley KM, et al. Effects ofprescription omega-3-acid ethyl esters on fasting lipidprofile in subjects with primary hypercholesterolemia.J Cardiovasc Pharmacol. 2011;57:489–494.

8. Davidson MH, Stein EA, Bays HE, et al. COMBination ofprescription Omega-3 with Simvastatin (COMBOS) Inves-tigators. Efficacy and tolerability of adding prescriptionomega-3 fatty acids 4 g/d to simvastatin 40 mg/d inhypertriglyceridemic patients: an 8-week, randomized,double-blind, placebo-controlled study. Clin Ther. 2007;29:1354–1367.

9. Dunbar RL, Nicholls SJ, Maki KC, et al. Effects of omega-3carboxylic acids on lipoprotein particles and other cardio-vascular risk markers in high-risk statin-treated patients withresidual hypertriglyceridemia: a randomized, controlled,double-blind trial. Lipids Health Dis. 2015;14:1.

10. Ng TW, Ooi EM, Watts GF, et al. Atorvastatin plusomega-3 fatty acid ethyl ester decreases very-low-densitylipoprotein triglyceride production in insulin resistantobese men. Diabetes Obes Metab. 2014;16:519–526.

11. Expert Panel on Detection, Evaluation, and Treatment ofHigh Blood Cholesterol in Adults. Executive summary ofthe Third Report of the National Cholesterol EducationProgram (NCEP) expert panel on detection, evaluation,and treatment of high blood cholesterol in adults (AdultTreatment Panel III). JAMA. 2001;285:2486–2497.

12. Eknoyan G, Lameire N, Eckardt K, et al. KDIGO 2012clinical practice guideline for the evaluation and manage-ment of chronic kidney disease. Kidney Int. 2013;3:5–14.

13. Alexander CM, Landsman PB, Teutsch SM, Haffner SM.NCEP-defined metabolic syndrome, diabetes, and preva-lence of coronary heart disease among NHANES IIIparticipants age 50 years and older. Diabetes. 2003;52:1210–1214.

14. Roh E, Ko SH, Kwon HS, et al, Taskforce Team ofDiabetes Fact Sheet of the Korean Diabetes Association.Prevalence and management of dyslipidemia inKorea: Korea National Health and Nutrition ExaminationSurvey during 1998 to 2010. Diabetes Metab J. 2013;37:433–449.

15. Varbo A, Benn M, Tybjærg-Hansen A, et al. Remnantcholesterol as a causal risk factor for ischemic heartdisease. J Am Coll Cardiol. 2013;61:427–436.

16. Di Angelantonio E, Sarwar N, Perry P, et al. Major lipids,apolipoproteins, and risk of vascular disease. JAMA.2009;302:1993–2000.

17. Miller M, Stone NJ, Ballantyne C, et al. Triglyceridesand cardiovascular disease a scientific statement fromthe American Heart Association. Circulation. 2011;123:2292–2333.

93

Clinical Therapeutics

18. Greene DJ, Skeggs JW, Morton RE.Elevated triglyceride content dimin-ishes the capacity of high densitylipoprotein to deliver cholesteryl es-ters via the scavenger receptor classB type I (SR-BI). J Biol Chem. 2001;276:4804–4811.

19. Chait A, Brazg RL, Tribble DL,Krauss RM. Susceptibility of small,dense, low-density lipoproteins tooxidative modification in subjectswith the atherogenic lipoproteinphenotype, pattern B. Am J Med.1993;94:350–356.

20. Ito MK. Long-chain omega-3 fattyacids, fibrates and niacin as thera-peutic options in the treatment ofhypertriglyceridemia: a review of theliterature. Atherosclerosis. 2015;242:647–656.

21. Davidson MH, Rosenson RS, MakiKC, et al. Effects of Fenofibric Acidon Carotid Intima-Media Thicknessin Patients With Mixed Dyslipidemiaon Atorvastatin Therapy Random-ized, Placebo-Controlled Study(FIRST). Arterioscler Thromb Vasc Biol.2014;34:1298–1306.

22. Kris-Etherton PM, Harris WS, AppelLJ, Committee N. Fish consumption,fish oil, omega-3 fatty acids, andcardiovascular disease. Circulation.2002;106:2747–2757.

23. Jacobson TA. Role of n− 3 fattyacids in the treatment of hypertri-glyceridemia and cardiovasculardisease. Am J Clin Nutr. 2008;87:1981S–1990S.

24. Jacobson TA, Glickstein SB,Rowe JD, Soni PN. Effects ofeicosapentaenoic acid and docosa-hexaenoic acid on low-density lip-oprotein cholesterol and otherlipids: a review. J Clin Lipidol. 2012;6:5–18.

25. Maki KC 1, Orloff DG, Nicholls SJ,et al. A highly bioavailable omega-3free fatty acid formulation improvesthe cardiovascular risk profile inhigh-risk, statin-treated patientswith residual hypertriglyceridemia(the ESPRIT trial). Clin Ther. 2013;35:1400–1411.

94

26. Bays HE, Ballantyne CM, KasteleinJJ, et al. Eicosapentaenoic acidethyl ester (AMR101) therapy inpatients with very high triglyceridelevels (from the Multi-center,plAcebo-controlled, Randomized,double-blINd, 12-week study withan open-label Extension [MARINE]trial). Am J Cardiol. 2011;108:682–690.

27. Ballantyne CM 1, Bays HE, KasteleinJJ, et al. Efficacy and safety ofeicosapentaenoic acid ethyl ester(AMR101) therapy in statin-treatedpatients with persistent high trigly-cerides (from the ANCHOR study).Am J Cardiol. 2012;110:984–992.

28. Chan DC, Watts GF. Dyslipidaemiain the metabolic syndrome and type2 diabetes: pathogenesis, priorities,pharmacotherapies. Expert Opin

Pharmacother. 2011;12:13–30.29. Pirillo A, Catapano AL. Omega-3

polyunsaturated fatty acids in thetreatment of hypertriglyceridaemia.Int J Cardiol. 2013;170:S16cS20.

30. Blaha MJ, Blumenthal RS, BrintonEA, Jacobson TA, National LipidAssociation Taskforce on Non-HDLCholesterol. The importance ofnon–HDL cholesterol reporting in

lipid management. J Clin Lipidol. 2008;2:267–273.

31. Sniderman AD, Scantlebury T, Cian-flone K. Hypertriglyceridemic hyper-apob: the unappreciated atherogenicdyslipoproteinemia in type 2 diabetesmellitus. Ann Intern Med. 2001;135:447–459.

32. Lu W, Resnick HE, Jablonski KA,et al. Non-HDL cholesterol as apredictor of cardiovascular diseasein type 2 diabetes. Diabetes Care.2003;26:16–23.

33. Jones PH, Davidson MH, Stein EA,et al. Comparison of the efficacy andsafety of rosuvastatin versus atorvas-tatin, simvastatin, and pravastatinacross doses (STELLAR Trial). Am J

Cardiol. 2003;92:152–160.34. Strandberg TE, Feely J, Sigurdsson EL.

Twelve-week, multicenter, randomized,open-label comparison of the effectsof rosuvastatin 10 mg/d and atorvas-tatin 10 mg/d in high-risk adults: aDISCOVERY study. Clin Ther. 2004;26:1821–1833.

35. Olsson AG, Istad H, Luurila O, et al.Effects of rosuvastatin and atorvastatincompared over 52 weeks of treatmentin patients with hypercholesterolemia.Am Heart J. 2002;144:1044–1051.

Address correspondence to: Hyo-Soo Kim, MD, PhD, Department ofInternal Medicine, Seoul National University Hospital, Cardiovascular Centre,101 Daehak-ro, Jongro-gu, Seoul 110-744, Korea. E-mail: usahyosoo@gmail.com, hyosoo@snu.ac.kr

Volume 40 Number 1