Lipid Guideline Controversies in 2014: The Decision is Yours Carl E. Orringer, MD, FACC, FNLA

Objectives To provide an overview of the American College of Cardiology/American Heart Association and the National Lipid Association lipid management approaches for ASCVD prevention To identify the similarities and differences between the two approaches To provide the information needed to decide which approach to use and when

Two Different Perspectives Two ASCVD Prevention Approaches Two Different Perspectives External reviewer Author

US age-standardized death rates attributable to CVD, 2000 to 2010 Release of NCEP ATP III Release of NCEP ATP III Update US age-standardized death rates* attributable to CVD, 2000 to 2010. *Directly standardized to the age distribution of the 2000 US standard population. †Total CVD: International Classification of Diseases, 10th Revision (ICD-10) I00 to I99 and Q20 to Q28. §Stroke (all cerebrovascular disease): ICD-10 I60 to I69. ¶CHD: ICD-10 I20 to I25. **Other CVD: ICD-10 I00 to I15, I26 to I51, I70 to I78, I80 to I89, and I95 to I99. CHD indicates coronary heart disease; and CVD, cardiovascular disease. Source: Centers for Disease Control and Prevention, National Center for Health Statistics.5 Go A S et al. Circulation. 2014;129:e28-e292 Copyright © American Heart Association, Inc. All rights reserved.

Mean age-adjusted LDL-C trends 2001–2011 in the United States: Analysis of 105 million patient records from a single national diagnostic laboratory 10.1371/journal.pone.0063416

Is There a Need for a Dramatic Change in Approach to ASCVD Prevention? ACC/AHA CCS IAS EAS/ESC JAS NICE NLA Changes in: Evidence base Central focus Lipid goals Use of non-statins Risk calculator

The Rules The topic will be identified Common ground and differences will be noted Appropriate supporting evidence will be introduced Summary will be provided You make the decision

Evidence Base ACC/AHA NLA Randomized controlled trials (RCT) of statin therapy Meta-analyses of RCT RCT of statins and non-statin drug therapy Meta-analyses of RCT Observational epidemiologic studies Genetic studies Metabolic studies Mechanistic studies

Randomized Controlled Trials (RCT) Systematically test effect(s) of an intervention on pre-specified outcomes in defined populations Their use minimizes confounding Study populations are often not diverse and exclusion criteria may hamper physician’s ability to apply results to real-world patients Most are designed to gain regulatory registration for pharmaceutical agents; lifestyle trials, studies of generic drugs or of those produced by smaller companies may be under-represented in RCT due to inadequate financial support

Observational Epidemiologic Studies Worldwide in scope and may assess ASCVD risk across populations Cohort studies evaluate mortality and morbidity within populations Confounding may occur even after matching, stratification, and multivariate adjustment because of measurement error or unmeasured or unknown risk factors

Observational Epidemiologic Cohort Study of 2146 Patients with FH and no CHD at Baseline Versmissen J, et al BMJ 2008; 337: a2423

Genetic Studies Genetic epidemiology reduces the likelihood of confounding by focusing on single variables: genetic mutations Identification of specific mutations may serve to generate hypotheses for other types of trials Often limited in patient selection and costly

Data Demonstrating Genetic Variants Affecting ASCVD Risk Sequence variants in the gene encoding for PCSK9 resulting in loss of function mutations are associated with 28% reduction in LDL-C, an 88% reduction in CHD risk and provide support for the value of long term low LDL-C in promoting CHD risk reduction J Cohen et al. N Engl J Med 2006;354:1264-72

Evidence Base: Summary ACC/AHA By limiting the scope to RCT of statins and meta-analyses of RCT, only the highest level of evidence on statins in defined populations is employed to assess ASCVD outcomes NLA By including evidence from RCT and other sources, a broader evidence base for clinical decision making is employed . This approach is consistent with the perspective of previous NCEP ATP’s and the international community

Central Focus of Guideline ACC/AHA NLA Identification of statin benefit groups Initiation and maintenance of high or moderate intensity statin therapy Abandonment of lipid goals Avoidance of non-statin therapy because of “unfavorable risk/benefit ratio.” Identification of an individual patient’s ASCVD risk based on clinical parameters and risk factors Initiation of ASCVD risk-based lipid-lowering therapy Maintenance of lipid goals to assess effective reduction of atherogenic lipoproteins and enhace adherence Use of high or moderate dose statins, ±non-statins, if necessary, to achieve goals

ACC/AHA Statin Benefit Groups H=High intensity statin; M=Moderate intensity statin Individuals with clinical ASCVD without New York Heart Association class II-IV heart failure or receiving hemodialysis (H preferred; M if age >75 or if not candidate for H). Individuals with primary elevations of LDL-C ≥190 mg/dl (H preferred; M if not candidate for H). Individuals age 40-75 years with diabetes, and LDL-C 70-189 mg/dl without clinical ASCVD (M if 10 yr risk <7.5%; H if ≥7.5%). Individuals without clinical ASCVD or diabetes, who are age 40-75 years with LDL-C 70-189 mg/dl, and have an estimated 10-year ASCVD risk of ≥ 7.5% using Pooled Cohort Equations (M or H).

High- and Moderate-Intensity Daily Statin Therapy High Intensity (Lowers LDL-C ≥ 50%) Atorvastatin 40-80 mg Rosuvastatin 20-40 mg Moderate Intensity (Lowers LDL-C 30-50%) Atorvastatin 10 (20) mg Rosuvastatin (5) 10 mg Simvastatin 20–40 mg Simvastatin 80 mg* Pravastatin 40 (80) mg Lovastatin 40 mg Fluvastatin XL 80 mg Fluvastatin 40 mg 2x/day Pitavastatin 2–4 mg Bold = Tested in RCT and reviewed by Expert Panel Yellow= Not tested in RCT

Efficacy of Intensive Lowering of LDL-C in Subjects with Low Baseline LDL-C Meta-analysis of RCT’s of >1000 participants and ≥2 years treatment duration of more versus less intense statin trials involving 169,138 subjects The major vascular event reduction, among in those with baseline LDL-C <77mg/dL per further 39 mg/dL reduction was 29% (99% CI 2-48, p=0.007); in those with baseline LDL-C <70 mg/dl, similar reduction in LDL-C continued to demonstrate MVE reduction (RR 0.63, 99% CI 0.41-0.95, p=0.004). Cholesterol Treatment Trialists Collaboration. Lancet 2010;376:1670-81

ACC/AHA Perspective on Statin Therapy Statin intensity trials showed clear benefit for high intensity versus moderate intensity statins Because fixed doses, not dosage titrations, were employed, one should not assume that a dosage titration strategy is correct or that addition of non-statins to achieve low LDL-C is indicated

ACC/AHA Perspective on Non-Statin Lipid Drug Therapy Non-statin drugs without demonstrated ASCVD risk reduction may favorably alter lipids but have an unfavorable risk/benefit ratio Niacin in AIM-HIGH and HPS-2 THRIVE Fibrates in ACCORD-Lipid, FIELD Lack of ASCVD event end-point data on ezetimibe CETP inhibitors torcetrapib and dalcetrapib The use of non-statin drugs should generally be avoided

Overview of the NLA Recommendations All preventive therapy begins with risk assessment and a provider-patient discussion of the pros and cons of therapy Lifestyle therapy is at the basis of all ASCVD preventive recommendations, regardless of baseline risk Judicious use of evidence-based drug therapy, particularly moderate and high-dose statins, is associated with optimal ASCVD risk reduction When excessive circulating atherogenic cholesterol (non-HDL-cholesterol and LDL cholesterol) persists after appropriate lifestyle and statin therapy, the use of non-statin therapy may be considered Long-term follow-up fostered by provider-patient communication is essential for optimal ASCVD prevention This slide provides an overview of the NLA Recommendations for the Patient-Centered Management of Dyslipidemia. You will note that it begins with and ends with provider patient interaction.

NLA ASCVD Risk Category Criteria Very High ASCVD Diabetes mellitus (type 1 or 2) ≥2 other major ASCVD risk factors; or Evidence of end-organ damage High ≥3 major ASCVD risk factors 0-1 other major ASCVD risk factor, and no evidence of end-organ damage Chronic kidney disease Stage 3B or 4 LDL-C ≥190 or non-HDL-C ≥220 mg/dL Moderate 2 major ASCVD risk factors For specific clinical features, high quantitative risk score or specific biomarker levels, consider reclassification to high risk Low 0-1 major ASCVD risk factor For specific clinical features, consider reclassification to moderate risk The NLA Recommendations advocate risk categorization, beginning with very-high and moving to progressively lower risk groups. Specific criteria are used for each risk category. Once the clinician defines the highest risk category applicable to a patient, the next step is discussion with the patient about risk-appropriate atherogenic cholesterol lowering therapy.

NLA ASCVD Risk Categories, Levels for Consideration of Drug Therapy and Treatment Goals Risk Category Consider Drug Therapy Treatment Goal Non-HDL-C /LDL-C Goal (mg/dL) Non-HDL-C/LDL-C Goal (mg/dL) Very-high ≥100 ≥70 <100 <70 High ≥130 <130 Moderate ≥160 Low ≥190 Recognizing that lifestyle therapy is always advocated as the basis for ASCVD prevention, this slide provides the NLA perspective on non-HDL-C and LDL-C thresholds for considering drug therapy and treatment goals. The non-HDL-C and LDL-C levels at which drug therapy should be considered vary in accordance with the absolute risk of the patient. However, the treatment goals are all <130 for non-HDL-C and <100 mg/dL for LDL-C, except in the very high risk patient in whom the non-HDL-C goal is <100 and LDL-C goal <70 mg/dL. For patients with ASCVD or diabetes mellitus, consider use of moderate or high intensity statins, irrespective of baseline atherogenic cholesterol levels.

NLA Perspective on Statin Therapy Statin therapy is the most potent and evidence-based approach to lowering atherogenic lipoproteins (non-HDL-C and LDL-C) Statin intensity trials showed clear benefit for high-intensity versus moderate-intensity statins Broad-based evidence supports “lower is better” concept, and provides an opportunity for clinicians to address residual risk above that addressed by appropriately-dosed statin therapy

NLA Perspective on Non-Statin Lipid Drug Therapy If non-HDL-C and LDL-C goals are not achieved with maximal tolerated statin therapy, the addition of non-statin therapy should be considered to lower atherogenic cholesterol levels and to achieve goals Doctors can be instructed not to use niacin in patients on aggressive statin regimens As ezetimibe is safe and lowers atherogenic cholesterol, its use may be considered in selected patients with elevated non-HDL-C and/or LDL-C Resins may be considered in selected patients Meta-analyses of fibrate therapy in subgroups with atherogenic dyslipidemia suggest ASCVD risk reduction

Is High-Dose Statin Therapy the End of the Line?

Variability of Achieved LDL-C With High-Intensity Statin Therapy Boekholdt SM et al. J Am Coll Cardiol 2014;64: 485-94 Meta analysis of 8 statin RCT involving 38, 153 subjects of whom 5,387 had 6,286 major CV events and had baseline and 1 year lipids and lipoproteins Waterfall plot of Individual values From TNT, SPARCL, IDEAL and JUPITER demonstrating variability of LDL-C lowering. >40% did not achieve LDL-C <70 mg/dl on atorvastatin 80 or rosuvastatin 20 mg daily Boekholdt’s meta-analysis of 8 statin RCT’s involving 38,153 subjects focused on 5,387 subjects who had 6,286 major CV events and had baseline and 1-year lipids and lipoproteins. In those receiving atorvastatin 80 mg daily or rosuvastatin 20 mg daily more than 40 % did not achieve LDL-C <70 mg/dl.

Very Low LDL-C and Non-HDL-C in Statin Trials and Major CVD Event Risk On Treatment LDL- C, Non-HDL-C mg/dL Boekholdt et al. JACC 2014;64:485-494

Central Focus of Guidelines: Summary ACC/AHA: define statin benefit groups; risk/benefit discussion; use moderate or high-intensity statin therapy with lifestyle change as background therapy; generally avoid non-statin drug therapy; no lipid goals NLA: identify ASCVD risk level; risk/benefit discussion; emphasize healthy lifestyle and use moderate or high-intensity statin therapy, and if necessary, adjunctive non-statin therapy, to lower atherogenic cholesterol; maintain lipid goals (non-HDL-C is favored lipoprotein target)

Risk Calculators ACC/AHA NLA Use Pooled Cohort Risk calculator in non-Hispanic Whites and non-Hispanic African Americans age 40-79 without ASCVD and not on statin therapy; may be considered in other populations Assessment of lifetime risk may be considered in those aged 20-59 with no ASCVD and not at high short-term risk Consider 10-year FRS, ACC/AHA Pooled Cohort Risk calculator, or 30-year risk in those with 2 major ASCVD risk factors; re-classify to higher risk those with ≥10% 10-year FRS, ≥15% ACC/AHA risk, or ≥45% long-term risk One of the most contentious controversies between the ACC/AHA Guideline and the NLA Recommendations relates to the use of risk calculators. The ACC/AHA Guideline advocates the use of the Pooled Cohort Risk calculator in non-Hispanic Whites and non-Hipsanic African Americans age 40-79 without ASCVD and not on stain therapy, and suggested that it could reasonably be used in other populations. The use of the lifetime risk calculator was suggested for those individuals age 20-59 not found to be a high risk. The NLA Recommendations advises consideration of any of three risk calculators, including Framingham 10-year risk, Lifetime or 30 year risk, or the 10-year Pooled Cohort Risk in those with 2 major risk factors.

This is a screen shot of the ACC/AHA risk calculator used to demonstrate 10 year ASCVD risk using the Pooled Cohort Risk Equations calculator available on the Cardiosource.org website. The variables employed include gender, age, HDL-C, systolic blood pressure, treatment for hypertension, cigarette smoking status and diabetes mellitus. When all of these variable are entered the 10 year calculated risk appears at the top on the left. Below it there is a box that estimates the patient’s risk level if all of those risk factors were optimal. To the right of those numbers there are boxes that estimate lifetime or 30-year ASCVD risk, a calculation that is recommended for consideration for individuals between the age of 20 and 59 with 10-year risk <7.5%. The Risk Assessment Working Group of the ACC/AHA does not use lifetime risk to advise the initiation of drug therapy, but suggests that such information may be of value when counseling individuals on the value of initiating lifestyle therapy for ASCVD prevention. http://www.tools.cardiosource.org/ascvd-risk-estimator

Key Disclaimer of the ACC/AHA Pooled Cohort Risk Calculator It does not definitively recommend statin therapy for individuals with 10-year risk ≥7.5% It advises that for such individuals before initiating statin therapy “it is reasonable for clinicians and patients to engage in a discussion which considers the potential for ASCVD risk reduction benefits and for adverse effects, for drug-drug interactions and patient preferences for treatment.” The ACC/AHA Expert Panel emphasizes that the finding of a 7.5% 10 year risk is not an automatic recommendation for statin therapy, but represents a level at which joint patient provider discussion should be undertaken, explaining to the patient the benefit of starting statin therapy.

Pooled Cohort Equations: Criticism Early criticism of the Pooled Risk calculator centered around lack of long-term prospective validation and possible overestimation of risk When the Pooled Risk Equation calculator was applied to 3 large, more recent primary prevention cohorts, the Women’s Health Study, the Physician’s Health Study and the Women’s Health Initiative Observational Study, it systematically overestimated observed risk by 75-150%, roughly doubling the actual observed risk As the Pooled Cohort Equations lacked prospective long-term validation, debate ensued on the advisability of using this tool for clinical risk assessment and management reccommendations. Ridker PM, Cook NR. Statins: new American guidelines for prevention of cardiovascular disease. Lancet 2013 10.1016/S0140-6736(13)62388-0. published Nov 20. PubMed Copyright NLA 2014

Pooled Cohort Equations: Defense Validation was more extensive than any other previous risk equations The primary prevention cohorts of WHS, PHS and WHI were likely significantly healthier than the general population In WHS and PHS some risk factors were self-reported in ranges rather than directly measured, resulting in imprecision All 3 cohorts might have been subject to some downstream initiation of statins The response to these criticisms were centered on potential selection of usually healthy subjects in the Women’s Health Study, Physician’s Health Study and Women’s Health Initiative populations; on the potential inaccuracy of self-reported rather than measured risk factors; and on the possibility that downstream initiation of statins may have been responsible for overestimation of risk. Lloyd-Jones DM, Goff D and Stone N. The Lancet, Early Online Publication, 4 December 2013 doi:10.1016/S0140-6736(13)62348-X Copyright NLA 2014

Pooled Cohort Equations: Criticism An analysis of the Women’s Health Study examined data from 632 women who had an ASCVD event over a median follow-up of 10 years After adjustment for the intervention effects of statins, revascularization and hypothetical confounding by indication the ratios of predicted to observed events were 1.8 in those with 0-<5% and 5-7.4% estimated 10-year risk and 1.3 in the groups with 7.5-10% and >10% estimated 10 year risk groups A more recent analysis adjusted for intervention of statins, revascularization and hypothetical confounding and still found that the risk calculator over-estimated risk in the Women’s Health Study. Cook NR, Ridker PM. JAMA Intern Med Published Online October 6, 2014

Pooled Cohort Equations: Defense 10,997 adults, ages 45-79, in the Reasons for Geographic Distribution and Racial Differences in Stroke study, without ASCVD or diabetes, with total cholesterol 70-189 mg/dl and not taking statins were examined for incident ASCVD at 5 years, and additional analysis in 3,333 Medicare beneficiaries added additional ASCVD events as defined in Medicare Claims data. There was a high degree of correlation between observed and predicted ASCVD incidence per 1,000 person-years in groups with <5%, 5-<7.5%, 7.5-<10% and ≥10% (calibration) and moderate to good ability to accurately rank-order individuals (discrimination) (C-statistic ~0.7). Further validation of the Pooled Risk Equations was provided in this 2014 study, which demonstrated a high degree of correlation between observed and predicted ASCVD incidence in low, intermediate and high risk groups. Muntner P et al. JAMA 2014;311:1406-1415 Copyright NLA 2014

Pooled Cohort Equations: Criticism Application to elderly populations When the risk calculator was applied to 4,854 Rotterdam Heart Study participants with a mean age of 65 years, 96.4% of men and 65.8% of women would be recommended statins The average predicted versus observed cumulative incidence of hard ASCVD events was 21.5 vs. 12.7 % for men and 11.6 vs. 7.9% for women (poor calibration) There is continuing concern that the Pooled Cohort Equations recommend statin use in too many elderly patients. One example of this was noted when the Pooled Cohort Equations were applied to men and women participating in the Rotterdam Heart Study. Kavousi M et al. JAMA 2014: 311(14): 1416-23

ACC/AHA Risk Calculator: Possible Overtreatment in Older Patients? Age Total cholesterol HDL cholesterol Systolic BP Treatment for HBP Diabetes Smoker 10-year ASCVD risk 60 AA♂ 170 50 125 No 7.5% 65 AA♀ 178 130 60 C ♂ 47 Examining US populations, the individuals for who the Pooled Cohort Equations were designed, this slide provides 3 examples of patients for whom clinicians might not normally consider statin therapy.

NLA Perspective on Risk Calculators Although any of 3 risk calculators are suggested for consideration (10 year Framingham risk, lifetime Framingham risk, ACC/AHA Pooled Cohort Risk Calculator), risk calculators measure diverse endpoints and have limited application in various ethnic and age groups The interpretation of a particular risk level using any risk calculator in a given patient must be done using careful clinical judgment The NLA Expert Panel provides a more conservative approach to the use of risk calculators, suggesting consideration of their use only in intermediate risk individuals for whom the clinician feels that more information is required to make preventive lipid therapy management decisions. It recognizes that each risk calculator examines different endpoints, has its strengths and weaknesses in any given patient and that careful clinical judgment must be used in applying the results to the management of a specific patient.

Risk Calculators ACC/AHA NLA Consider 10-year FRS, ACC/AHA Pooled Cohort Risk calculator, or 30-year risk in those with 2 major ASCVD risk factors; re-classify to higher risk those with ≥10% 10-year FRS, ≥15% ACC/AHA risk, or ≥45%% long-term risk Use Pooled Cohort Risk calculator in non-Hispanic whites and non-Hispanic African Americans age 40-79 without ASCVD and not on statin therapy; may be considered in other populations Assessment of lifetime risk may be considered in those aged 20-59 with no ASCVD and not at high short-term risk

Lipid Guideline Controversies: Common Threads Between ACC/AHA and NLA Lifestyle therapy is warranted for ASCVD risk reduction, whether or not drug therapy is used Patients with ASCVD, FH and diabetes are candidates for moderate or high-dose statins Risk calculators aid in, but do not take the place of clinical judgment Whether or not lipid goals are set, regular lipid follow-up is warranted to assess adherence Patient engagement in preventive care decision making aids in long-term adherence

What’s Ahead in Guidelines? A new app for the NLA Recommendations Part 1 New ACC/AHA performance measures related to the 2013 Blood Cholesterol Guideline and another app related to the Guideline NLA Recommendations Part 2 on special populations (women, elderly, HIV, south Asian Indians, Hispanic, CKD, heart failure, rheumatoid arthritis) A new NLA Self-Assessment Program on Guidelines

The Decision is Yours ACC/AHA NLA