LP(A) AND TRIGLYCERIDE RICH LIPOPROTEINS

Transcription

1 LP(A) AND TRIGLYCERIDE RICH LIPOPROTEINS Alan S Brown, MD FACC FNLA President, National Lipid Association Director, Division of Cardiology Advocate Lutheran General Hospital Park Ridge, Illinois Clinical Associate Professor of Medicine Loyola Stritch School of Medicine

2 DISCLOSURES FOR ALAN S BROWN, MD Speakers Bureau: Amgen, Regeneron, Sanofi, Advisory board: Akcea, Amgen, Regeneron,Sanofi

3 WHAT IS LP(A) AND IT S STRUCTURE?

4 Leibundgut et al JACC 2012 and JLR 2013

5 STRUCTURE OF HUMAN LP(A) Lp(a) consists of an LDL-like particle and apo(a), which are covalently bound via a disulfide bond between Cys4326 of apob-100 and Cys4057 of apo(a) located in kringle IV (KIV) type 9 (KIV9) The apo(a) comprises 10 KIV subunits, of which KIV2 is present in variable identically sized repeats, kringle V (KV), and an inactive protease domain. The KIV2 has variable repetitive identical repeats (3 to >40) determined by the LPA gene The apo (a) shows a high degree of homology (75-100%) with plasminogen at both the nucleotide and amino acid level The mrna for apo (a) is primarily experessed in the liver with minimal amounts in the testes, brain, adrenal glands,and pituitary Lp(a) levels are genetically determined by the LPA alleles present in an individual Graham, et al J. Lipid Res :

6 WHY DO WE HAVE LP(A)? It is highly likely that it plays a role in mediation of wound healing, as immunohistochemical analysis of healing wounds stained positively for apo(a)/apob during the infiltration of immune cells, production of granulation tissue, and initiation of revascularisation. In addition, a proteomics study determined that many of the proteins associated with Lp(a) were involved with the wound healing response. Riches, et al, Volume 2012, Article ID , 10 pages doi: /2012/923289

7 LP(A) AND ATHEROSCLEROSIS The species distribution of Lp(a) is limited to humans and old world monkeys (a distant homolog is present in hedgehogs) due to expression of the apo(a) gene. Transgenic animals have been generated to express human apo(a), or human apo(a) and apob-100 to aid the study of Lp(a) and in general these have confirmed the observation that Lp(a) is atherogenic regardless of species.

8 LP(A) AND ATHEROSCLEROSIS

9 LP(A) ATHEROGENICITY Smaller size is due to fewer KIV2 repeats Smaller size is correlated with increased levels of Lp(a) in the circulation and greater particle numbers Smaller size is correlated with increased risk of atherosclerosis Ramesh Saeedi et al. Clinical Diabetes and Endocrinology2016 2:7

10 Tsimikas, S. (2017). A Test in Context: Lipoprotein(a) Diagnosis, Prognosis, Controversies, and Emerging Therapies. Journal of the American College of Cardiology, 69(10),

11 Tsimikas, S. (2017). A Test in Context: Lipoprotein(a) Diagnosis, Prognosis, Controversies, and Emerging Therapies. Journal of the American College of Cardiology, 69(6),

12 LP(A) AND AORTIC STENOSIS

13 The Author Sotirios Tsimikas JACC 2017;69:

14 POTENTIAL THERAPIES TO REDUCE CV RISK IN PATIENTS WITH ELEVATED LP(A) Statin therapy to lower LDL (Lp(a) is not decreased) High dose Niacin Estrogen CETP inhibitors PCSK9 antibodies Apheresis mrnai/antisense therapy

15 RESIDUAL RISK DESPITE STATINS In the Jupiter Trial, though patients with elevated Lp(a) had similar reduction in risk from rosuvastation, the overall risk remained higher in those with elevated Lp(a) In the present cohort of asymptomatic white JUPITER participants with low LDL cholesterol and elevated hscrp, Lp(a) was a significant determinant of residual risk. Furthermore, the efficacy of rosuvastatin in reducing CVD was similar among participants with high or low Lp(a) concentrations. Khera, et al Circulation. 2014;129:

16 Efficacy of rosuvastatin according to baseline lipoprotein(a) [Lp(a)] concentration. Amit V. Khera et al. Circulation. 2014;129: Copyright American Heart Association, Inc. All rights reserved.

17 APHERESIS FOR ELEVATED LP(A) A single treatment reduces both LDL and Lp(a) by 60-70% No prospective randomized trials for apheresis in high Lp(a) patientrs exist Lipoprotein apheresis has beneficial effects regarding endothelial function and myocardial perfusion in patients with high levels of Lp(a) Case reports showing decreased CV events post apheresis compared to pre apheresis are somewhat encouraging One retrospective evaluation indicates that patients with elevated Lp(a) irrespective of the LDL-c level have a greater benefit from lipoprotein apheresis than patients with low Lp(a) and high levels of LDL-c Can be considered for patients with recurrent clinical CV events despite maximal LDL-C lowering and who have elevated Lp(a) Von Dreyender, et al. Differences in the atherogenic risk of patients treated by lipoprotein apheresis according to their lipid pattern. Atheroscler Suppl 14(1): Vogt, Clin Res Cardiol Suppl (2017) (Suppl) 12:12 17

18 LIPOPROTEIN APHERESIS IN PATIENTS WITH MAXIMALLY TOLERATED LIPID LOWERING THERAPY, LP(A)HYPERLIPOPROTEINEMIA AND PROGRESSIVE CARDIOVASCULAR DISEASE: PROSPECTIVE OBSERVATIONAL MULTICENTER STUDY Circulation. published online September 20, 2013; 166 patients with CAD and Lp(a) with progressive CAD Event rates after 2 years of apheresis were compared to the rates during the prior 2 years Mean annual MACE rates declined from.41 for 2 years prior to apheresis to.09 for the 2 years during apheresis (p<.0001) All vascular event annual rates declined from 0.61 to.16(p<.0001) LEEBMANN, et al DOI: /CIRCULATIONAHA

19 ANTISENSE TECHNOLOGY REDUCES DISEASE CAUSING PROTEIN LEVELS BY TARGETING MRNA Gene mrna Disease-Causing Protein Traditional Small Molecule Drugs Inhibitors ora gonists of proteins Translation Transcription DISEASE DISEASE Biologics Inhibitors orm im ic s of proteins Disease-Causing Protein Translation X X DISEASE Antisense Oligonucleotide Inhibition of RN A fu nc tion (no prod u c tion of d isease c au sing protein)

20 ISIS-APO(A)RX MECHANISM OF ACTION IN REDUCING PLASMA LP(A) Tsimikas et al, Lancet 2015

21 PHASE 1 ISIS-APO(A)RX STUDY MEAN PERCENT CHANGE IN LP(A) OVER TIME BY TREATMENT GROUP MULTIPLE-DOSE COHORTS (N=29) **p<0.01 ***p Tsimikas et al, The Lancet 2015

22 ISIS-APO(A)RX PHASE I TRIAL RELATIONSHIP OF PLASMA ISIS-APO(A)RX TROUGH CONCENTRATIONS AND MEAN PERCENT CHANGE IN LP(A), OXPLAPOB AND OXPL-APO(A) 22 Tsimikas et al, Lancet 2015

23 IN WHOM SHOULD YOU ORDER LP(A)? Patients with Familial Hypercholesterolemia Patients with progressive CAD despite maximal therapy Patients with a family history of CAD Consider in patients with less than expected LDL response to statin therapy?aortic stenosis patients early in their course?

24 CONCLUSIONS Lp(a) levels are primarily genetically determined and increase the risk of atherosclerosis through multiple mechanisms Randomized trials to determine effect of lowering Lp(a) have not been performed in the past due to absence of targeted reliable Lp(a) lowering data Smaller studies suggest attenuation of risk by aggressively lowering LDL but more recent data suggests that residual risk remains Encouraging observational studies with apheresis have been published but no prospective randomized trials New therapies with antisense oligonucleotides against apo a provide promise for future prospective outcome trials to determine if Lp(a) should be a target of therapy

25 TRIGLYCERIDE RICH LIPOPROTEINS An independant CV risk factor in patients with established atherosclerosis Predict risk in patients with low HDL and/or high LDL when the LDL/HDL is >5 May be secondary to metabolic syndrome or insulin resistance May not represent risk when HDL is high such as women on estrogen or those with heavy Etoh consumption

26 ADDING TRIGLYCERIDE LOWERING TO STATINS IN CLINICAL TRIALS HAS YIELDED DISAPPOINTING RESULTS DESPITE OBSERVATION OF RISK ENHANCEMENT Aim High (Niacin) HPS2 Thrive (Niacin with antiflushing agent) ACCORD LIPID Trial (fenofibrate) Recent VITAL Trial (Low dose EPA/DHA omega 3)

27 REDUCTION OF CARDIOVASCULAR EVENTS WITH ICOSAPENT ETHYL INTERVENTION TRIAL Deepak L Bhatt, MD, MPH, Ph. Gabriel Steg, MD, Michael Miller, MD, Eliot A. Brinton, MD, Terry A. Jacobson, MD, Steven B. Ketchum, PhD, Ralph T. Doyle, Jr., BA, Rebecca A. Juliano, PhD, Lixia Jiao, PhD, Craig Granowitz, MD, PhD, Jean-Claude Tardif, MD, Christie M. Ballantyne, MD, on Behalf of the REDUCE-IT Investigators

28 LOW DOSE OMEGA-3 MIXTURES SHOW NO SIGNIFICANT CARDIOVASCULAR BENEFIT No. of Events (%) Source Treatment Control Rate Ratios (CI) Favors Treatment Favors Control Coronary heart disease Nonfatal myocardial infarction 1121 (2.9) 1155 (3.0) 0.97 ( ) Coronary heart disease 1301 (3.3) 1394 (3.6) 0.93 ( ) Any 3085 (7.9) 3188 (8.2) 0.96 ( ) P=.12 Stroke Ischemic 574 (1.9) 554 (1.8) 1.03 ( ) Hemorrhagic 117 (0.4) 109 (0.4) 1.07 ( ) Unclassified/other 142 (0.4) 135 (0.3) 1.05 ( ) Any 870 (2.2) 843 (2.2) 1.03 ( ) P=.60 Revascularization Coronary Noncoronary Any 3044 (9.3) 3040 (9.3) 1.00 ( ) 305 (2.7) 330 (2.9) 0.92 ( ) 3290 (10.0) 3313 (10.2) 0.99 ( ) P=.60 Any major vascular event 5930 (15.2) 6071 (15.6) 0.97 ( ) P= Rate Ratio 2.0 Adapted with permissionǂ from Aung T, Halsey J, Kromhout D, et al. Associations of omega-3 fatty acid supplement use with cardiovascular disease risks: Meta-analysis of 10 trials involving individuals. JAMA Cardiol. 2018;3: [ǂ

29 JELIS SUGGESTS CV RISK REDUCTION WITH EPA IN JAPANESE HYPERCHOLESTEROLEMIC PATIENTS Kaplan-Meier Estimates of Incidence of Coronary Events Total Population Primary Prevention Cohort Major coronary events (%) Control Control EPA* 2 EPA* Hazard ratio: 0.81 ( ) p=0.011 Hazard ratio: 0.82 ( ) p= Hazard ratio: 0.81 ( ) p= Years Numbers at risk Treatment group Control 8.0 EPA* Control group Secondary Prevention Cohort Years Years *1.8 g/day Adapted with permission from Yokoyama M, Origasa H, Matsuzaki M, et al. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet. 2007;369:

30 CONSORT DIAGRAM Countries Sites Screened N=19, Randomized N=8179 (43% of screened) Screen Fails N=11,033* Incl./Excl. criteria not met 10,429 Withdrawal of consent 340 Adverse event 13 Primary Prevention category closed 4 Death 5 Lost to follow-up 108 Enrollment closed 3 Other 135 *4 patients presented 2 screen failure reasons. Icosapent Ethyl N=4089 (100%) Placebo N=4090 (100%) Completed Study N=3684 (90.1%) Completed Study N=3630 (88.8%) Early Discontinuation from Study N=405 (9.9%) Actual vs. potential total follow-up time (%) 93.6% Known vital status 4083 (99.9%) Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Early Discontinuation from Study N=460 (11.2%) Actual vs. potential total follow-up time (%) 92.9% Known vital status 4077 (99.7%) Median trial follow up duration was 4.9 years.

31 Key Inclusion Criteria 1. Age 45 years with established CVD (Secondary Prevention Cohort) or 50 years with diabetes with 1 additional risk factor for CVD (Primary Prevention Cohort) 2. Fasting TG levels 150 mg/dl and <500 mg/dl* 3. LDL-C >40 mg/dl and 100 mg/dl and on stable statin therapy (± ezetimibe) for 4 weeks prior to qualifying measurements for randomization *Due to the variability of triglycerides, a 10% allowance existing in the initial protocol, which permitted patients to be enrolled with qualifying triglycerides 135 mg/dl. protocol amendment 1 (May 2013) changed the lower limit of acceptable triglycerides from 150 mg/dl to 200 mg/dl, with no variability allowance. Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

32 Inclusion Criteria for Secondary Prevention Cohort One or more of the following: 1. Documented coronary artery disease Multi vessel CAD ( 50% stenosis in 2 major epicardial coronary arteries with or without antecedent revascularization Prior MI Hospitalization for high-risk non-st-segment elevation acute coronary syndrome with ST-segment deviation or biomarker positivity Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

33 Inclusion Criteria for Secondary Prevention Cohort One or more of the following: 1. Documented coronary artery disease Multi vessel CAD ( 50% stenosis in 2 major epicardial coronary arteries with or without antecedent revascularization Prior MI Hospitalization for high-risk non-st-segment elevation acute coronary syndrome with ST-segment deviation or biomarker positivity 2. Documented cerebrovascular or carotid disease Prior ischemic stroke Symptomatic carotid artery disease with 50% carotid arterial stenosis Asymptomatic carotid artery disease with 70% carotid arterial stenosis History of carotid revascularization Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

34 Inclusion Criteria for Secondary Prevention Cohort One or more of the following: 1. Documented coronary artery disease Multi vessel CAD ( 50% stenosis in 2 major epicardial coronary arteries with or without antecedent revascularization Prior MI Hospitalization for high-risk non-st-segment elevation acute coronary syndrome with ST-segment deviation or biomarker positivity 2. Documented cerebrovascular or carotid disease Prior ischemic stroke Symptomatic carotid artery disease with 50% carotid arterial stenosis Asymptomatic carotid artery disease with 70% carotid arterial stenosis History of carotid revascularization 3. Documented peripheral artery disease Ankle-brachial index <0.9 with symptoms of intermittent claudication History of aorto-iliac or peripheral artery intervention Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

35 Inclusion Criteria for Primary Prevention Cohort 1. Diabetes mellitus requiring medication AND Patients with diabetes and CVD are counted under Secondary Prevention Cohort Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

36 Inclusion Criteria for Primary Prevention Cohort 1. Diabetes mellitus requiring medication AND years of age AND Patients with diabetes and CVD are counted under Secondary Prevention Cohort Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

37 Inclusion Criteria for Primary Prevention Cohort 1. Diabetes mellitus requiring medication AND years of age AND 3. 1 additional risk factor for CVD Men 55 years and women 65 years Cigarette smoker or stopped smoking within 3 months Hypertension ( 140 mmhg systolic OR 90 mmhg diastolic) or on antihypertensive medication; HDL-C 40 mg/dl for men or 50 mg/dl for women hscrp >3.0 mg/l Renal dysfunction: Creatinine clearance >30 and <60 ml/min Retinopathy Micro- or macroalbuminuria ABI <0.9 without symptoms of intermittent claudication Patients with diabetes and CVD are counted under Secondary Prevention Cohort Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

38 Key Exclusion Criteria 1. Severe (NYHA class IV) heart failure 2. Severe liver disease 3. History of pancreatitis 4. Hypersensitivity to fish and/or shellfish Adapted with permissionǂ from: Bhatt DL, Steg PG, Brinton EA, et al; on behalf of the REDUCE-IT Investigators. Rationale and design of REDUCE-IT: Reduction of Cardiovascular Events with Icosapent Ethyl Intervention Trial. Clin Cardiol. 2017;40: [ǂ

39 EFFECTS ON BIOMARKERS FROM BASELINE TO YEAR 1 Icosapent Ethyl (N=4089) Median Placebo (N=4090) Median Median Between Group Difference at Year 1 Baseline Year 1 Baseline Year 1 Absolute Change from Baseline Triglycerides (mg/dl) < Non-HDL-C (mg/dl) < LDL-C (mg/dl) < HDL-C (mg/dl) < Apo B (mg/dl) < hscrp (mg/l) < EPA (µg/ml) < Biomarker* *Apo B and hscrp were measured at Year 2. Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed % Change from Baseline % Change P-value

40 PRIMARY END POINT: CV DEATH, MI, STROKE, CORONARY REVASC, UNSTABLE ANGINA 30 Patients with an Event (%) 28.3% 20 Placebo Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed

41 PRIMARY END POINT: CV DEATH, MI, STROKE, CORONARY REVASC, UNSTABLE ANGINA % Patients with an Event (%) Hazard Ratio, 0.75 (95% CI, ) 20 Placebo 23.0% Icosapent Ethyl Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed

42 PRIMARY END POINT: CV DEATH, MI, STROKE, CORONARY REVASC, UNSTABLE ANGINA % Patients with an Event (%) Hazard Ratio, 0.75 (95% CI, ) 20 Placebo 23.0% Icosapent Ethyl Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed RRR = 24.8% ARR = 4.8% NNT = 21 (95% CI, 15 33)

43 PRIMARY END POINT: CV DEATH, MI, STROKE, CORONARY REVASC, UNSTABLE ANGINA % Patients with an Event (%) Hazard Ratio, 0.75 (95% CI, ) 20 Placebo 23.0% Icosapent Ethyl Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed RRR = 24.8% ARR = 4.8% NNT = 21 (95% CI, 15 33) P=

44 KEY SECONDARY END POINT: CV DEATH, MI, STROKE Patients with an Event (%) % 20 Placebo Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed

45 KEY SECONDARY END POINT: CV DEATH, MI, STROKE 30 Patients with an Event (%) Hazard Ratio, % 20 Placebo 16.2% 10 Icosapent Ethyl Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed (95% CI, )

46 KEY SECONDARY END POINT: CV DEATH, MI, STROKE 30 Patients with an Event (%) Hazard Ratio, % 20 Placebo 16.2% 10 Icosapent Ethyl Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed (95% CI, ) RRR = 26.5% ARR = 3.6% NNT = 28 (95% CI, 20 47)

47 KEY SECONDARY END POINT: CV DEATH, MI, STROKE 30 Patients with an Event (%) Hazard Ratio, % 20 Placebo 16.2% 10 Icosapent Ethyl Years since Randomization Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed (95% CI, ) RRR = 26.5% ARR = 3.6% NNT = 28 (95% CI, 20 47) P=

48 PRIMARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 559/2892 (19.3%) 146/1197 (12.2%) 738/2893 (25.5%) 163/1197 (13.6%) 0.73 ( ) 0.88 ( ) Region Western Eastern Asia Pacific 551/2906 (19.0%) 143/1053 (13.6%) 11/130 (8.5%) 713/2905 (24.5%) 167/1053 (15.9%) 21/132 (15.9%) 0.74 ( ) 0.84 ( ) 0.49 ( ) Ezetimibe Use No Yes 649/3827 (17.0%) 56/262 (21.4%) 834/3828 (21.8%) 67/262 (25.6%) 0.75 ( ) 0.82 ( ) Male Female 551/2927 (18.8%) 154/1162 (13.3%) 715/2895 (24.7%) 186/1195 (15.6%) 0.73 ( ) 0.82 ( ) White vs Non-White White Non-White 646/3691 ( 17.5%) 59/398 (14.8%) 812/3688 (22.0%) 89/401 (22.2%) 0.77 ( ) 0.60 ( ) Age Group <65 Years 65 Years 322/2232 (14.4%) 383/1857 (20.6%) 460/2184 (21.1%) 441/1906 (23.1%) 0.65 ( ) 0.87 ( ) US vs Non-US US Non-US 281/1548 (18.2%) 424/2541 (16.7%) 394/1598 (24.7%) 507/2492 (20.3%) 0.69 ( ) 0.80 ( ) Baseline Diabetes Diabetes No Diabetes 433/2394 (18.1%) 272/1695 (16.0%) 536/2393 (22.4%) 365/1694 (21.5%) 0.77 ( ) 0.73 ( ) Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 197/905 (21.8%) 380/2217 (17.1%) 128/963 (13.3%) 263/911 (28.9%) 468/2238 (20.9%) 170/939 (18.1%) 0.71 ( ) 0.80 ( ) 0.70 ( ) Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl 430/2481 (17.3%) 275/1605 (17.1%) 559/2469 (22.6%) 342/1620 (21.1%) 0.73 ( ) 0.79 ( ) Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl 640/3674 (17.4%) 65/412 (15.8%) 811/3660 (22.2%) 90/429 (21.0%) 0.75 ( ) 0.79 ( ) Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No 149/823 (18.1%) 554/3258 (17.0%) 214/794 (27.0%) 687/3293 (20.9%) 0.62 ( ) 0.79 ( ) Baseline Statin Intensity High Moderate Low 232/1290 (18.0%) 424/2533 (16.7%) 48/254 (18.9%) 310/1226 (25.3%) 543/2575 (21.1%) 45/267 (16.9%) 0.69 ( ) 0.76 ( ) 1.12 ( ) Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl 211/1319 (16.0%) 221/1223 (18.1%) 186/1133 (16.4%) 267/1222 (21.8%) 277/1209 (22.9%) 256/1188 (21.5%) 0.70 ( ) 0.79 ( ) 0.73 ( ) Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l 288/1919 (15.0%) 417/2167 (19.2%) 407/1942 (21.0%) 494/2147 (23.0%) 0.68 ( ) 0.81 ( ) Primary Composite End Point (ITT) HR (95% CI) Int P Val Subgroup Sex Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better 1.8

49 KEY SECONDARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 361/2892 (12.5%) 98/1197 (8.2%) 489/2893 (16.9%) 117/1197 (9.8%) 0.72 ( ) 0.81 ( ) Region Western Eastern Asia Pacific 358/2906 (12.3%) 93/1053 (8.8%) 8/130 (6.2%) 473/2905 (16.3%) 117/1053 (11.1%) 16/132 (12.1%) 0.73 ( ) 0.78 ( ) 0.47 ( ) Ezetimibe Use No Yes 426/3827 (11.1%) 33/262 (12.6%) 569/3828 (14.9%) 37/262 (14.1%) 0.73 ( ) 0.87 ( ) 353/2927 (12.1%) 106/1162 (9.1%) 474/2895 (16.4%) 132/1195 (11.0%) 0.72 ( ) 0.80 ( ) White vs Non-White White Non-White 418/3691 (11.3%) 41/398 (10.3%) 538/3688 (14.6%) 68/401 (17.0%) 0.76 ( ) 0.55 ( ) Age Group <65 Years 65 Years 200/2232 (9.0%) 259/1857 (13.9%) 290/2184 (13.3%) 316/1906 (16.6%) 0.65 ( ) 0.82 ( ) 187/1548 (12.1%) 272/2541 (10.7%) 266/1598 (16.6%) 340/2492 (13.6%) 0.69 ( ) 0.77 ( ) Baseline Diabetes Diabetes No Diabetes 286/2394 (11.9%) 173/1695 (10.2%) 391/2393 (16.3%) 215/1694 (12.7%) 0.70 ( ) 0.80 ( ) Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 152/905 (16.8%) 229/2217 (10.3%) 78/963 (8.1%) 205/911 (22.5%) 296/2238 (13.2%) 105/939 (11.2%) 0.71 ( ) 0.77 ( ) 0.70 ( ) Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl 290/2481 (11.7%) 169/1605 (10.5%) 371/2469 (15.0%) 235/1620 (14.5%) 0.75 ( ) 0.71 ( ) Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl 421/3674 (11.5%) 38/412 (9.2%) 546/3660 (14.9%) 60/429 (14.0%) 0.74 ( ) 0.66 ( ) Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No 101/823 (12.3%) 356/3258 (10.9%) 136/794 (17.1%) 470/3293 (14.3%) 0.68 ( ) 0.75 ( ) Baseline Statin Intensity High Moderate Low 151/1290 (11.7%) 270/2533 (10.7%) 37/254 (14.6%) 210/1226 (17.1%) 361/2575 (14.0%) 32/267 (12.0%) 0.66 ( ) 0.74 ( ) 1.20 ( ) Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl 135/1319 (10.2%) 135/1223 (11.0%) 121/1133 (10.7%) 171/1222 (14.0%) 182/1209 (15.1%) 178/1188 (15.0%) 0.71 ( ) 0.73 ( ) 0.69 ( ) Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l 183/1919 (9.5%) 276/2167 (12.7%) 245/1942 (12.6%) 361/2147 (16.8%) 0.73 ( ) 0.73 ( ) Key Secondary Composite Endpoint (ITT) HR (95% CI)* Int P Val Subgroup Subgroup Hazard Ratio (95% CI) Sex Male Female Icosapent Ethyl Placebo 0.44 HR (95% CI) Int P Val 0.13 Risk Category Secondary Prevention Cohort Primary Prevention Cohort /2892 (12.5%) 98/1197 (8.2%) US vs Non-US US Non-US 489/2893 (16.9%) 117/1197 (9.8%) Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better ( ) 0.81 ( )

50 KEY SECONDARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 361/2892 (12.5%) 98/1197 (8.2%) 489/2893 (16.9%) 117/1197 (9.8%) 0.72 ( ) 0.81 ( ) Region Western Eastern Asia Pacific 358/2906 (12.3%) 93/1053 (8.8%) 8/130 (6.2%) 473/2905 (16.3%) 117/1053 (11.1%) 16/132 (12.1%) 0.73 ( ) 0.78 ( ) 0.47 ( ) Ezetimibe Use No Yes 426/3827 (11.1%) 33/262 (12.6%) 569/3828 (14.9%) 37/262 (14.1%) 0.73 ( ) 0.87 ( ) Male Female 353/2927 (12.1%) 106/1162 (9.1%) 474/2895 (16.4%) 132/1195 (11.0%) 0.72 ( ) 0.80 ( ) White vs Non-White White Non-White 418/3691 (11.3%) 41/398 (10.3%) 538/3688 (14.6%) 68/401 (17.0%) 0.76 ( ) 0.55 ( ) Age Group <65 Years 65 Years 200/2232 (9.0%) 259/1857 (13.9%) 290/2184 (13.3%) 316/1906 (16.6%) 0.65 ( ) 0.82 ( ) Key Secondary Composite Endpoint (ITT) HR (95% CI)* Int P Val Subgroup Sex Subgroup Hazard Ratio (95% CI) US vs Non-US US Non-US Placebo /1548 (12.1%) 272/2541 (10.7%) 266/1598 (16.6%) 340/2492 (13.6%) 0.69 ( ) 0.77 ( ) 286/2394 (11.9%) 173/1695 (10.2%) 391/2393 (16.3%) 215/1694 (12.7%) 0.70 ( ) 0.80 ( ) 152/905 (16.8%) 229/2217 (10.3%) 78/963 (8.1%) 205/911 (22.5%) 296/2238 (13.2%) 105/939 (11.2%) 0.71 ( ) 0.77 ( ) 0.70 ( ) 290/2481 (11.7%) 169/1605 (10.5%) 371/2469 (15.0%) 235/1620 (14.5%) 0.75 ( ) 0.71 ( ) Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl 421/3674 (11.5%) 38/412 (9.2%) 546/3660 (14.9%) 60/429 (14.0%) 0.74 ( ) 0.66 ( ) Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No 101/823 (12.3%) 356/3258 (10.9%) 136/794 (17.1%) 470/3293 (14.3%) 0.68 ( ) 0.75 ( ) Baseline Statin Intensity High Moderate Low 151/1290 (11.7%) 270/2533 (10.7%) 37/254 (14.6%) 210/1226 (17.1%) 361/2575 (14.0%) 32/267 (12.0%) 0.66 ( ) 0.74 ( ) 1.20 ( ) Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl 135/1319 (10.2%) 135/1223 (11.0%) 121/1133 (10.7%) 171/1222 (14.0%) 182/1209 (15.1%) 178/1188 (15.0%) 0.71 ( ) 0.73 ( ) 0.69 ( ) Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l 183/1919 (9.5%) 276/2167 (12.7%) 245/1942 (12.6%) 361/2147 (16.8%) 0.73 ( ) 0.73 ( ) Baseline Diabetes Diabetes No Diabetes Sex Male Female Icosapent Ethyl Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 HR (95% CI) /2927 (12.1%) 106/1162 (9.1%) Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl 474/2895 (16.4%) 132/1195 (11.0%) Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better 1.8 Int P Val 0.72 ( ) 0.80 ( )

51 KEY SECONDARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 361/2892 (12.5%) 98/1197 (8.2%) 489/2893 (16.9%) 117/1197 (9.8%) 0.72 ( ) 0.81 ( ) Region Western Eastern Asia Pacific 358/2906 (12.3%) 93/1053 (8.8%) 8/130 (6.2%) 473/2905 (16.3%) 117/1053 (11.1%) 16/132 (12.1%) 0.73 ( ) 0.78 ( ) 0.47 ( ) Ezetimibe Use No Yes 426/3827 (11.1%) 33/262 (12.6%) 569/3828 (14.9%) 37/262 (14.1%) 0.73 ( ) 0.87 ( ) Male Female 353/2927 (12.1%) 106/1162 (9.1%) 474/2895 (16.4%) 132/1195 (11.0%) 0.72 ( ) 0.80 ( ) White vs Non-White White Non-White 418/3691 (11.3%) 41/398 (10.3%) 538/3688 (14.6%) 68/401 (17.0%) 0.76 ( ) 0.55 ( ) Age Group <65 Years 65 Years 200/2232 (9.0%) 259/1857 (13.9%) 290/2184 (13.3%) 316/1906 (16.6%) 0.65 ( ) 0.82 ( ) US vs Non-US US Non-US 187/1548 (12.1%) 272/2541 (10.7%) 266/1598 (16.6%) 340/2492 (13.6%) 0.69 ( ) 0.77 ( ) Baseline Diabetes Diabetes No Diabetes 286/2394 (11.9%) 173/1695 (10.2%) 391/2393 (16.3%) 215/1694 (12.7%) 0.70 ( ) 0.80 ( ) Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 152/905 (16.8%) 229/2217 (10.3%) 78/963 (8.1%) 205/911 (22.5%) 296/2238 (13.2%) 105/939 (11.2%) 0.71 ( ) 0.77 ( ) 0.70 ( ) 290/2481 (11.7%) 169/1605 (10.5%) 371/2469 (15.0%) 235/1620 (14.5%) 0.75 ( ) 0.71 ( ) 421/3674 (11.5%) 38/412 (9.2%) 546/3660 (14.9%) 60/429 (14.0%) 0.74 ( ) 0.66 ( ) Key Secondary Composite Endpoint (ITT) HR (95% CI)* Int P Val Subgroup Sex Subgroup Hazard Ratio (95% CI) Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl US vs Non-US US Non-US Icosapent Ethyl Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl Placebo HR (95% CI) 0.62 Int P Val 0.68 Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No 187/1548 (12.1%) 272/2541 (10.7%) /794 (17.1%) 470/3293 (14.3%) 0.68 ( ) 0.75 ( ) 151/1290 (11.7%) 270/2533 (10.7%) 37/254 (14.6%) 210/1226 (17.1%) 361/2575 (14.0%) 32/267 (12.0%) 0.66 ( ) 0.74 ( ) 1.20 ( ) Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl 135/1319 (10.2%) 135/1223 (11.0%) 121/1133 (10.7%) 171/1222 (14.0%) 182/1209 (15.1%) 178/1188 (15.0%) 0.71 ( ) 0.73 ( ) 0.69 ( ) Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l 183/1919 (9.5%) 276/2167 (12.7%) 245/1942 (12.6%) 361/2147 (16.8%) 0.73 ( ) 0.73 ( ) Baseline Statin Intensity High Moderate Low /1598 (16.6%) 340/2492 (13.6%) 101/823 (12.3%) 356/3258 (10.9%) Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better ( ) 0.77 ( )

52 KEY SECONDARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 361/2892 (12.5%) 98/1197 (8.2%) 489/2893 (16.9%) 117/1197 (9.8%) 0.72 ( ) 0.81 ( ) Region Western Eastern Asia Pacific 358/2906 (12.3%) 93/1053 (8.8%) 8/130 (6.2%) 473/2905 (16.3%) 117/1053 (11.1%) 16/132 (12.1%) 0.73 ( ) 0.78 ( ) 0.47 ( ) Ezetimibe Use No Yes 426/3827 (11.1%) 33/262 (12.6%) 569/3828 (14.9%) 37/262 (14.1%) 0.73 ( ) 0.87 ( ) Male Female 353/2927 (12.1%) 106/1162 (9.1%) 474/2895 (16.4%) 132/1195 (11.0%) 0.72 ( ) 0.80 ( ) White vs Non-White White Non-White 418/3691 (11.3%) 41/398 (10.3%) 538/3688 (14.6%) 68/401 (17.0%) 0.76 ( ) 0.55 ( ) Age Group <65 Years 65 Years 200/2232 (9.0%) 259/1857 (13.9%) 290/2184 (13.3%) 316/1906 (16.6%) 0.65 ( ) 0.82 ( ) US vs Non-US US Non-US 187/1548 (12.1%) 272/2541 (10.7%) 266/1598 (16.6%) 340/2492 (13.6%) 0.69 ( ) 0.77 ( ) Baseline Diabetes Diabetes No Diabetes 286/2394 (11.9%) 173/1695 (10.2%) 391/2393 (16.3%) 215/1694 (12.7%) 0.70 ( ) 0.80 ( ) Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 152/905 (16.8%) 229/2217 (10.3%) 78/963 (8.1%) 205/911 (22.5%) 296/2238 (13.2%) 105/939 (11.2%) 0.71 ( ) 0.77 ( ) 0.70 ( ) 290/2481 (11.7%) 169/1605 (10.5%) 371/2469 (15.0%) 235/1620 (14.5%) 0.75 ( ) 0.71 ( ) 421/3674 (11.5%) 38/412 (9.2%) 546/3660 (14.9%) 60/429 (14.0%) 0.74 ( ) 0.66 ( ) 101/823 (12.3%) 356/3258 (10.9%) 136/794 (17.1%) 470/3293 (14.3%) 0.68 ( ) 0.75 ( ) Key Secondary Composite Endpoint (ITT) HR (95% CI)* Int P Val Subgroup Sex Subgroup Baseline Diabetes Diabetes No Diabetes Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl 0.62 Hazard Ratio (95% CI) Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl Icosapent Ethyl Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No Placebo HR (95% CI) 0.68 Int P Val 0.50 Baseline Statin Intensity High Moderate Low 286/2394 (11.9%) 173/1695 (10.2%) Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l /2393 (16.3%) 215/1694 (12.7%) 151/1290 (11.7%) 270/2533 (10.7%) 37/254 (14.6%) 210/1226 (17.1%) 361/2575 (14.0%) 32/267 (12.0%) 0.66 ( ) 0.74 ( ) 1.20 ( ) 135/1319 (10.2%) 135/1223 (11.0%) 121/1133 (10.7%) 171/1222 (14.0%) 182/1209 (15.1%) 178/1188 (15.0%) 0.71 ( ) 0.73 ( ) 0.69 ( ) 183/1919 (9.5%) 276/2167 (12.7%) 245/1942 (12.6%) 361/2147 (16.8%) 0.73 ( ) 0.73 ( ) Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better ( ) 0.80 ( )

53 KEY SECONDARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 361/2892 (12.5%) 98/1197 (8.2%) 489/2893 (16.9%) 117/1197 (9.8%) 0.72 ( ) 0.81 ( ) Region Western Eastern Asia Pacific 358/2906 (12.3%) 93/1053 (8.8%) 8/130 (6.2%) 473/2905 (16.3%) 117/1053 (11.1%) 16/132 (12.1%) 0.73 ( ) 0.78 ( ) 0.47 ( ) Ezetimibe Use No Yes 426/3827 (11.1%) 33/262 (12.6%) 569/3828 (14.9%) 37/262 (14.1%) 0.73 ( ) 0.87 ( ) Male Female 353/2927 (12.1%) 106/1162 (9.1%) 474/2895 (16.4%) 132/1195 (11.0%) 0.72 ( ) 0.80 ( ) White vs Non-White White Non-White 418/3691 (11.3%) 41/398 (10.3%) 538/3688 (14.6%) 68/401 (17.0%) 0.76 ( ) 0.55 ( ) Age Group <65 Years 65 Years 200/2232 (9.0%) 259/1857 (13.9%) 290/2184 (13.3%) 316/1906 (16.6%) 0.65 ( ) 0.82 ( ) US vs Non-US US Non-US 187/1548 (12.1%) 272/2541 (10.7%) 266/1598 (16.6%) 340/2492 (13.6%) 0.69 ( ) 0.77 ( ) Baseline Diabetes Diabetes No Diabetes 286/2394 (11.9%) 173/1695 (10.2%) 391/2393 (16.3%) 215/1694 (12.7%) 0.70 ( ) 0.80 ( ) Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 152/905 (16.8%) 229/2217 (10.3%) 78/963 (8.1%) 205/911 (22.5%) 296/2238 (13.2%) 105/939 (11.2%) 0.71 ( ) 0.77 ( ) 0.70 ( ) Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl 290/2481 (11.7%) 169/1605 (10.5%) 371/2469 (15.0%) 235/1620 (14.5%) 0.75 ( ) 0.71 ( ) Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl 421/3674 (11.5%) 38/412 (9.2%) 546/3660 (14.9%) 60/429 (14.0%) 0.74 ( ) 0.66 ( ) Key Secondary Composite Endpoint (ITT) HR (95% CI)* Int P Val Subgroup Sex Subgroup Hazard Ratio (95% CI) Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No Baseline Statin Intensity High Moderate Low Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl Icosapent Ethyl 0.68 ( ) 0.75 ( ) 151/1290 (11.7%) 270/2533 (10.7%) 37/254 (14.6%) 210/1226 (17.1%) 361/2575 (14.0%) 32/267 (12.0%) 0.66 ( ) 0.74 ( ) 1.20 ( ) 135/1319 (10.2%) 135/1223 (11.0%) 121/1133 (10.7%) 171/1222 (14.0%) 182/1209 (15.1%) 178/1188 (15.0%) 0.71 ( ) 0.73 ( ) 0.69 ( ) 183/1919 (9.5%) 276/2167 (12.7%) Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better Int P Val /794 (17.1%) 470/3293 (14.3%) 290/2481 (11.7%) 169/1605 (10.5%) Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l Placebo 101/823 (12.3%) 356/3258 (10.9%) 371/2469 (15.0%) 235/1620 (14.5%) 245/1942 (12.6%) 361/2147 (16.8%) ( ) 0.73 ( ) HR (95% CI) 0.75 ( ) 0.71 ( )

54 KEY SECONDARY END POINT IN SUBGROUPS End Point/Subgroup Hazard Ratio (95% CI) Icosapent Ethyl Placebo 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) Risk Category Secondary Prevention Cohort Primary Prevention Cohort 361/2892 (12.5%) 98/1197 (8.2%) 489/2893 (16.9%) 117/1197 (9.8%) 0.72 ( ) 0.81 ( ) Region Western Eastern Asia Pacific 358/2906 (12.3%) 93/1053 (8.8%) 8/130 (6.2%) 473/2905 (16.3%) 117/1053 (11.1%) 16/132 (12.1%) 0.73 ( ) 0.78 ( ) 0.47 ( ) Ezetimibe Use No Yes 426/3827 (11.1%) 33/262 (12.6%) 569/3828 (14.9%) 37/262 (14.1%) 0.73 ( ) 0.87 ( ) Male Female 353/2927 (12.1%) 106/1162 (9.1%) 474/2895 (16.4%) 132/1195 (11.0%) 0.72 ( ) 0.80 ( ) White vs Non-White White Non-White 418/3691 (11.3%) 41/398 (10.3%) 538/3688 (14.6%) 68/401 (17.0%) 0.76 ( ) 0.55 ( ) Age Group <65 Years 65 Years 200/2232 (9.0%) 259/1857 (13.9%) 290/2184 (13.3%) 316/1906 (16.6%) 0.65 ( ) 0.82 ( ) US vs Non-US US Non-US 187/1548 (12.1%) 272/2541 (10.7%) 266/1598 (16.6%) 340/2492 (13.6%) 0.69 ( ) 0.77 ( ) Baseline Diabetes Diabetes No Diabetes 286/2394 (11.9%) 173/1695 (10.2%) 391/2393 (16.3%) 215/1694 (12.7%) 0.70 ( ) 0.80 ( ) Baseline egfr <60 ml/min/1.73m2 60-<90 ml/min/1.73m2 90 ml/min/1.73m2 152/905 (16.8%) 229/2217 (10.3%) 78/963 (8.1%) 205/911 (22.5%) 296/2238 (13.2%) 105/939 (11.2%) 0.71 ( ) 0.77 ( ) 0.70 ( ) Baseline Triglycerides 200 vs <200 mg/dl Triglycerides 200 mg/dl Triglycerides <200 mg/dl 290/2481 (11.7%) 169/1605 (10.5%) 371/2469 (15.0%) 235/1620 (14.5%) 0.75 ( ) 0.71 ( ) Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl 421/3674 (11.5%) 38/412 (9.2%) 546/3660 (14.9%) 60/429 (14.0%) 0.74 ( ) 0.66 ( ) Baseline Triglycerides 200 and HDL-C 35 mg/dl Yes No 101/823 (12.3%) 356/3258 (10.9%) 136/794 (17.1%) 470/3293 (14.3%) 0.68 ( ) 0.75 ( ) 151/1290 (11.7%) 270/2533 (10.7%) 37/254 (14.6%) 210/1226 (17.1%) 361/2575 (14.0%) 32/267 (12.0%) 0.66 ( ) 0.74 ( ) 1.20 ( ) 135/1319 (10.2%) 135/1223 (11.0%) 121/1133 (10.7%) 171/1222 (14.0%) 182/1209 (15.1%) 178/1188 (15.0%) 0.71 ( ) 0.73 ( ) 0.69 ( ) Key Secondary Composite Endpoint (ITT) HR (95% CI)* Int P Val Subgroup Sex Subgroup Hazard Ratio (95% CI) Baseline Statin Intensity High Moderate Low Icosapent Ethyl Baseline LDL-C (Derived) by Tertiles 67 mg/dl >67-84 mg/dl >84 mg/dl Placebo 0.10 HR (95% CI) Int P Val 0.90 Baseline Triglycerides 150 vs <150 mg/dl Triglycerides 150 mg/dl Triglycerides <150 mg/dl 421/3674 (11.5%) 38/412 (9.2%) Baseline hscrp 2 vs >2 mg/l 2 mg/l >2 mg/l 183/1919 (9.5%) 276/2167 (12.7%) Icosapent Ethyl Better Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed Placebo Better /3660 (14.9%) 60/429 (14.0%) 245/1942 (12.6%) 361/2147 (16.8%) ( ) 0.73 ( ) 0.74 ( ) 0.66 ( )

55 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Primary Composite (ITT) Icosapent Ethyl Placebo 705/4089 (17.2%) 901/4090 (22.0%) 0.4 Icosapent Ethyl Better Placebo Better Hazard Ratio (95% CI) RRR P-value 0.75 ( ) 25% RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

56 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 0.74 ( ) 26% 0.4 Icosapent Ethyl Better Placebo Better 606/4090 (14.8%) RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

57 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% 0.4 Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

58 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% 0.4 Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

59 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% 0.4 Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

60 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% Cardiovascular Death 174/4089 (4.3%) 213/4090 (5.2%) 0.80 ( ) 20% Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

61 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% Cardiovascular Death 174/4089 (4.3%) 213/4090 (5.2%) 0.80 ( ) 20% 0.03 Hospitalization for Unstable Angina 108/4089 (2.6%) 157/4090 (3.8%) 0.68 ( ) 32% Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

62 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% Cardiovascular Death 174/4089 (4.3%) 213/4090 (5.2%) 0.80 ( ) 20% 0.03 Hospitalization for Unstable Angina 108/4089 (2.6%) 157/4090 (3.8%) 0.68 ( ) 32% Fatal or Nonfatal Stroke 98/4089 (2.4%) 134/4090 (3.3%) 0.72 ( ) 28% Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

63 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% Cardiovascular Death 174/4089 (4.3%) 213/4090 (5.2%) 0.80 ( ) 20% 0.03 Hospitalization for Unstable Angina 108/4089 (2.6%) 157/4090 (3.8%) 0.68 ( ) 32% Fatal or Nonfatal Stroke 98/4089 (2.4%) 134/4090 (3.3%) 0.72 ( ) 28% /4089 (13.4%) 690/4090 (16.9%) 0.77 ( ) 23% Total Mortality, Nonfatal Myocardial Infarction, or Nonfatal Stroke 0.4 Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

64 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% Cardiovascular Death 174/4089 (4.3%) 213/4090 (5.2%) 0.80 ( ) 20% 0.03 Hospitalization for Unstable Angina 108/4089 (2.6%) 157/4090 (3.8%) 0.68 ( ) 32% Fatal or Nonfatal Stroke 98/4089 (2.4%) 134/4090 (3.3%) 0.72 ( ) 28% /4089 (13.4%) 690/4090 (16.9%) 0.77 ( ) 23% Total Mortality, Nonfatal Myocardial Infarction, or Nonfatal Stroke 0.4 Icosapent Ethyl Better Placebo Better RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.

65 PRESPECIFIED HIERARCHICAL TESTING Endpoint Hazard Ratio (95% CI) Icosapent Ethyl Placebo Hazard Ratio (95% CI) RRR P-value Primary Composite (ITT) 705/4089 (17.2%) 901/4090 (22.0%) 0.75 ( ) 25% Key Secondary Composite (ITT) 459/4089 (11.2%) 606/4090 (14.8%) 0.74 ( ) 26% Cardiovascular Death or Nonfatal Myocardial Infarction 392/4089 (9.6%) 507/4090 (12.4%) 0.75 ( ) 25% Fatal or Nonfatal Myocardial Infarction 250/4089 (6.1%) 355/4090 (8.7%) 0.69 ( ) 31% Urgent or Emergent Revascularization 216/4089 (5.3%) 321/4090 (7.8%) 0.65 ( ) 35% Cardiovascular Death 174/4089 (4.3%) 213/4090 (5.2%) 0.80 ( ) 20% 0.03 Hospitalization for Unstable Angina 108/4089 (2.6%) 157/4090 (3.8%) 0.68 ( ) 32% Fatal or Nonfatal Stroke 98/4089 (2.4%) 134/4090 (3.3%) 0.72 ( ) 28% 0.01 Total Mortality, Nonfatal Myocardial Infarction, or Nonfatal Stroke 549/4089 (13.4%) 690/4090 (16.9%) 0.77 ( ) 23% Total Mortality 274/4089 (6.7%) 0.87 ( ) 13% Icosapent Ethyl Better Placebo Better 310/4090 (7.6%) RRR denotes relative risk reduction Bhatt DL, Steg PG, Miller M, et al. N E ngljm ed.2018.