Non-HDL cholesterol, ApoB and LDL particle concentration in coronary heart disease risk prediction and treatment

Transcription

1 Clinical Lipidology ISSN: (Print) (Online) Journal homepage: Non-HDL cholesterol, ApoB and LDL particle concentration in coronary heart disease risk prediction and treatment Carl E Orringer To cite this article: Carl E Orringer (2013) Non-HDL cholesterol, ApoB and LDL particle concentration in coronary heart disease risk prediction and treatment, Clinical Lipidology, 8:1, To link to this article: Copyright 2013 Future Medicine Ltd Published online: 18 Jan Submit your article to this journal Article views: 113 View related articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 08 December 2017, At: 23:07

2 review Non-HDL cholesterol, ApoB and LDL particle concentration in coronary heart disease risk prediction and treatment LDL cholesterol has been identified as the primary therapeutic target for lipid management to reduce the future risk of coronary events. Despite the use of LDL cholesterol-targeted therapy with statins and other lipid-altering agents, many patients still suffer coronary events. Residual risk for such events has been attributed, at least in part, to persistently elevated atherogenic particle concentration. When patients are found to have elevated non-hdl cholesterol, ApoB-100 or LDL particle concentrations, they are presumed to be at increased risk and are often prescribed more aggressive lipid-altering therapy. This review examines the evidence that these biomarkers provide additional clinically useful information both in coronary risk prediction and in lipid management decision-making. Keywords: ApoB biomarkers for coronary heart disease coronary heart disease risk prediction LDL particle concentration LDL particle number non-hdl cholesterol The traditional lipid profile, including a fasting plasma measurement of total cholesterol (TC), HDL cholesterol (HDL-C), triglycerides and calculated LDL cholesterol (LDL-C) has been recognized as the standard by which coronary heart disease (CHD) risk assessment and on-treatment management decisions are made. Biomarker assessment has been proposed to refine risk prediction and to more accurately identify those who may require more aggressive lipid management. This article focuses on the controversy related to the use of CHD risk prediction information derived from standard lipids versus that obtained from the measurement of atherogenic lipoprotein particle concentrations. Biomarkers: definitions & standards for clinical utility A biomarker is a clinical test that is used to predict disease and/or to assist in more effective treatment of disease. The role of the standard lipid profile is well established in clinical decisionmaking. Global CHD risk scoring that uses multiple traditional cardiovascular risk factors is a class I recommendation (benefit greatly exceeds risk and the test should be performed) of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and is advised in all asymptomatic adults without a clinical history of CHD [1]. These scores combine individual risk factor measurements into a single, clinically useful, quantitative estimate of risk that can be used to guide preventive therapies. The lipid values employed in the most widely used global risk scores are TC or LDL-C and HDL-C (Framingham Risk Score [101], Systematic Coronary Risk Evaluation [102] and Reynolds Risk Score [103]), or LDL-C, HDL-C and triglycerides (Prospective Cardiovascular Münster Study [104]). The decision about whether to use a new biomarker depends upon how readily it is measured, its potential to add new information and its ability to aid the clinician in managing patients. Quality of measurement depends upon accuracy and reproducibility of analytical methods, whether preanalytical issues, including stability, have been properly assessed, accessibility of the assay, availability of the assay and reasonable cost of the test. A test s potential to add new information is a function of strong and consistent association between the biomarker and the disease in question in multiple studies, whether it adds to or improves upon existing tests, whether decision limits, if recommended, are validated in multiple studies and whether the evaluation includes data from multiple populations. The usefulness of the biomarker for patient management requires superior performance to existing /CLP Future Medicine Ltd Clin. Lipidol. (2013) 8(1), Carl E Orringer University Hospitals Case Medical Center, Harrington Heart & Vascular Institute, Euclid Avenue, LKS 5038, Cleveland, OH 44106, USA Tel.: Fax: carl.orringer@uhhospitals.org part of ISSN

3 Review Orringer diagnostic tests, evidence that associated risk is modifiable with specific therapy or evidence that biomarker-guided triage or monitoring enhances care [2]. ApoB 100 particle synthesis, metabolism & role in atherogenesis ApoB 100 (ApoB) is an essential component of atherogenic particles. Its synthesis in the liver is controlled by the apob gene. The expression of this gene is not autoregulated as hepatic ApoB content is determined by proteolysis. After ApoB is synthesized it interacts with cholesteryl esters, resulting in conformational changes in ApoB, decreased degradation and increased production. Subsequent incorporation of triglycerides into this complex through the action of microsomal transfer proteins, the addition of phospholipids, and apolipoproteins E, C I, C II and C III, readies the VLDL particle to be secreted into the plasma. Following secretion, VLDL triglycerides are hydrolyzed by lipoprotein lipase and its cofactor, ApoC II. The free fatty acids that are produced are taken up by muscle cells for energy utilization and are re-esterified to triglycerides by adipose tissue for energy storage. The remaining lipoprotein particles are VLDL remnants, of which the smallest is intermediate-density lipoprotein (IDL). Some IDL interacts with ApoE and is taken up by the LDL receptor in the liver and the remainder is hydrolyzed by hepatic lipase, resulting in the formation of LDL, the end product of VLDL catabolism. ApoA is another lipoprotein that is synthesized in the liver. It has been demonstrated to associate extracellularly with circulating ApoB lipoproteins to form lipoprotein(a) particles [3]. Particles of VLDL, VLDL remnants, IDL, LDL and lipoprotein(a) each carry one molecule of ApoB that acts to facilitate attachment of these particles to cellular receptors [4]. Such attachment is necessary to allow LDL particles to deliver cholesteryl esters to the tissues for cell membrane formation and hormone, vitamin and bile acid synthesis. Approximately 90% of ApoB-containing particles are LDL particles, a finding that is responsible for the strong correlation between the plasma concentrations of ApoB and LDL particles [5]. This observation is strong supporting evidence for the identical recommendations for the clinical use of ApoB and LDL particle concentration (LDL P) in initial clinical assessment and on-treatment management decisions advocated by the 70 Clin. Lipidol. (2013) 8(1) National Lipid Association expert panel of lipid specialists [6]. The process of atherosclerosis is propagated by excessive plasma atherogenic particle concentration and endothelial dysfunction, resulting in the gradient-driven diffusion of ApoB-containing particles from the vascular lumen into the arterial intima. Circulating monocytes are attracted to activated endothelial cells by leukocyte adhesion molecules. These monocytes migrate into the intima where they mature into macrophages that multiply and produce proatherogenic chemokines [7]. Proteoglycan-mediated LDL particle reten tion, oxidative modification and macrophage engulfment of these LDL particles result in the formation of foam cells. The ingress of smooth muscle cells promotes formation of the fibrous cap overlying these plaques. Recruitment of inf lammatory cells and the expression of growth factors and additional cytokines result in thinning of the fibrous cap, setting the stage for plaque rupture, vascular thrombosis and acute coronary syndromes [8]. Reliability of lipoprotein measurement for CHD risk assessment The LDL C concentration, the concentration of cholesterol in plasma LDL particles, is the primary lipid marker of CHD risk and the target of lipid-altering therapy [9 11]. It is most often reported as a calculated value using the Friedewald equation, in which the concentration of LDL C equals that of TC minus HDL C minus VLDL cholesterol (VLDL C). This equation is based upon the assumption that the concentration of VLDL C equals the plasma triglyceride concentration divided by five. In patients with plasma triglycerides <400 mg/dl, the Friedewald equation provides a reasonable estimate of LDL C, except in those with chylomicronemia and excessive quantities of b-vldl. However, when plasma triglycerides are higher, the Friedewald equation should not be used to estimate LDL C as the ratio of plasma triglycerides:vldl C may be considerably greater than five due to triglyceride enrichment of VLDL and/or hyperchylomicronemia [12]. Confounding inf luences on calculated LDL C-based CHD risk assessment occur in hypertriglyceridemic states. An analysis of data from the National Health and Nutrition Examination Survey (NHANES) and the American Heart Association comparing population data from 1994 to 2002 versus

4 Non HDL cholesterol, ApoB & LDL particle concentration in coronary heart disease risk those of 2003 to 2010 shows that there has been a progressive increase in the incidence of hypertriglyceridemia, hypertriglyceridemia with low HDL C, Type 2 diabetes and impaired fasting glucose [13]. The NHANES estimated that 35.7% of adults were obese as defined by a BMI of 30 kg/m2 [105]. This finding is frequently associated with the lipoprotein abnormalities of the metabolic syndrome. In the setting of increased prevalence of hypertriglyceridemic states and in order to more accurately assess CHD risk in patients with hypertriglyceridemia, the use of non HDL C concentration has been advocated as a secondary target of therapy in patients with plasma triglycerides of mg/dl. The physiologic importance of non HDL C is that it is a measure of the cholesterol content of all atherogenic particles. It is calculated by subtracting the concentration of HDL C from TC. Because measurement of TC and HDL C are independent of the postprandial state, non HDL C may be used in clinical decisionmaking in patients in the nonfasting state. The targeted goal for non HDL C has been established as <30 mg/dl above the patient s LDL C, based upon the observation that when triglyceride levels are 150 mg/dl, VLDL C values are usually 30 mg/dl. When plasma triglycerides are >150 mg/dl, VLDL C is usually >30 mg/dl and additional therapy directed at lowering non HDL C is warranted (Figure 1) [9]. As the concentration of non HDL C has been demonstrated to be highly correlated with that of ApoB [14], non HDL C has been used as a risk marker and target of therapy [15]. The validity of this premise should be examined based upon an understanding of the reliability of LDL C, non HDL C and ApoB measurement, and upon an appreciation of the difference between correlation and concordance of LDL C and LDL P. Analytic laboratory error related to lack of reproducibility (a measurement of how often repeated measurements agree with one another) and to inaccuracy (a measurement of the systematic difference in results obtained by using any given method versus the gold standard for that method) is substantial for all of the lipid-based measurements. A study examining the correlation between 145 pairs of Friedewald-calculated and directly measured LDL C found that one-third of the measurements had a >15 mg/dl difference and 25% had a >20 mg/dl difference [16]. Biologic variation of serial measurements of TC, HDL C Review Total cholesterol + IDL-C + RLP-C + Lp(a)-C + Cholesterol in HDL particles + Cholesterol in LDL particles Cholesterol in VLDL particles Non-HDL-C Use clinically when plasma TG 200 mg/dl Non-HDL-C = total cholesterol cholesterol in HDL particles IDL-C + RLP-C + Lp(a)-C is usually minimal Because desirable TG is <150 mg/dl and VLDL-C generally = TG/5, goal for non-hdl-c is <30 mg/dl above LDL-C goal When non-hdl-c is 30 mg/dl above LDL-C goal, more intensive lipid therapy is warranted Figure 1. Understanding non-hdl cholesterol. HDL-C: HDL cholesterol; IDL-C: Intermediate-density lipoprotein cholesterol; LDL-C: LDL cholesterol; Lp(a)-C: Lipoprotein(a) cholesterol; RLP-C: Remnant lipoprotein cholesterol; TG: Triglyceride; VLDL-C: VLDL cholesterol. Data taken from [9]. and calculated LDL C may also be substantial [17,18] and potentially alter clinical decisionmaking. The importance of avoiding imprecision of measurement was emphasized by The National Cholesterol Education Program Working Group on Lipoprotein Measurement, which stated that risk categorization should not be based on a single LDL C measurement [19]. Implementation of this recommendation in the USA may be obstructed by variability in insurance reimbursement for two lipid panels obtained within a relatively short time frame. The observation that the degree of corre lation between non HDL C and ApoB is high does not imply concordance between the two measurements. Concordance between two variables means that for any given value of one, there is a limited range of values of the other. Sniderman et al. examined the relationship between non HDL C and ApoB in the NHANES survey of the American population [20]. While the correlation between the two variables was very high (0.95), as well as a high concordance at the extremes of values, there was considerable discordance in the mid-range of values, a variance that is likely 71

5 Review Orringer due to the compositional differences related to the cholesterol content of VLDL and LDL particles [20]. Direct measurement of ApoB and LDL P provides another option to assess atherogenic particle concentration. Although the bias and imprecision of ApoB measurement appears to be quite low, more data are needed to allow rigorous assessment of currently available ApoB assays, especially in patients with lipoprotein disorders [21]. Measurement of LDL P by nuclear magnetic resonance (NMR) spectroscopy employs the assessment of characteristic signals broadcasted by the number of terminal methyl groups of the lipids contained within the lipoprotein particle core. Cholesteryl esters and triglycerides in the particle core each contribute three methyl groups and phospholipids and unesterified cholesterol contribute two methyl groups [22]. The LDL P measured by NMR spectroscopy includes large and small LDL particles and IDL particles. Although its use has been validated against other methods of lipoprotein subclass measurement, there are no international validation standards [23]. Non HDL-C as a CHD risk marker & on-treatment guide to therapy One means of examining non HDL C as a marker of atherosclerosis is to assess its value in postmortem atherosclerotic lesions. The Pathobiological Determinants of Atherosclerosis in Youth Study examined serum lipids and lipoproteins obtained during autopsy within 72 h following death in 715 cases of accidental death, homicide or suicide in subjects aged years to determine whether the measurement of ApoA1 and B, lipoprotein(a), and sizes of lipoproteins improved the ability to predict the extent of fatty streaks in the thoracic and abdominal aorta, and in the right coronary artery more than the lipid measurements of HDL C and non HDL C. None of the apolipoprotein measurements were as strongly or consistently correlated with the extent of lesions as was the measurement of HDL C or non HDL C. Beyond the basic model that included sex, age, race, smoking status, hypertension, HDL C and non HDL C, the addition of ApoA1 and ApoB measurements added only an average 1.3% explanatory ability to the model, whereas the lipid measures of HDL C plus non HDL C added an average 2.5% [24]. Another means of assessing the clinical value of non HDL C measurement is to examine 72 Clin. Lipidol. (2013) 8(1) its ability to predict CHD risk in prospective epidemiological trials. The Emerging Risk Factors Collaboration, a patient-level meta-analysis, analyzed individual records of 302,430 people without initial vascular disease from 68 long-term studies. There were 2.79 million patient-years of follow-up, during which there were 8857 nonfatal myocardial infarctions, 3928 CHD deaths, 2534 ischemic strokes, 513 hemorrhagic strokes and 2536 unclassified strokes. Hazard ratios (HRs) adjusted for several conventional risk factors were calculated for traditional lipids, directly measured LDL C, non HDL C and ApoA1 and B. The study demonstrated that HRs associated with non HDL C and HDL C were nearly identical to those of ApoA1 and B. In addition, HRs were similar with non HDL C versus directly measured LDL C [25]. Other authors have disagreed with this conclusion and argued for the greater relative value of non HDL C and ApoB versus LDL C in initial CHD risk prediction. Sniderman et al. carried out a study-level meta-analysis including three large studies that were not part of the Emerging Risk Factors Collaboration metaanalysis [26]. They reviewed 12 independent epidemiological studies, including 233,455 sub jects and 22,950 cardiovascular events, and examined published risk estimates, converted them to standardized risk ratios, and analyzed them, employing quantitative meta-analysis using a random effects model. For each standard deviation increase, the relative risk (RR) ratio and 95% CI for LDL C was 1.25 ( ), non HDL C 1.34 ( ) and ApoB 1.43 ( ). The comparisons of withinstudy differences showed that the ApoB RR ratio was 5.7% >non HDL C (p < 0.001) and 12% >LDL C (p < ). The non HDL C RR ratio was 5.0% >LDL C (p = 0.017) [26]. A reduction in the concentration of non HDL C has been found to be a consistent marker of lower CHD risk on therapy. A meta-analysis of 14 statin trials, seven fibrate trials, seven trials of niacin monotherapy or combination therapy, one study of ileal bypass, one of a diet high in polyunsaturated fatty acids and one of bile acidbinding therapy, showed a one-to-one relationship between the percentage of non HDL C lowering and CHD risk reduction [27]. The use of non HDL C concentrations may be of value in CHD risk prediction even in populations in which the plasma triglycerides are <200 mg/dl. The Lipid Research Clinics Program Follow-Up study was a primary prevention

6 Non HDL cholesterol, ApoB & LDL particle concentration in coronary heart disease risk study of 4462 subjects aged years, in whom mean baseline plasma triglycerides were 153 mg/dl in men and 117 mg/dl in women. The participants were followed for an average of 19 years. Non HDL C was found to be a stronger predictor of all-cause mortality and cardiovascular disease (CVD) mortality than LDL C (c2-test for non HDL C 24.3 vs 5.0 for LDL C) [28]. The European Prospective Investigation into Cancer and Nutrition-Norfolk Prospective Population Study followed 21,448 participants, aged years, without diabetes or CHD, for 11 years. The mean plasma triglyceride levels in subjects without and with CHD were 150 and 159 mg/dl, respectively, in men and 115 and 150 mg/dl, respectively, in women. A total of 2086 participants developed clinical CHD during follow-up. After adjustment for age, smoking, waist circumference, physical activity, systolic blood pressure and hormone replacement therapy for women, increasing levels of non HDL C were a better predictor of risk for future CHD (HR: 2.39; 95% CI: ) than LDL C, triglycerides and total cholesterol:hdl C ratio. Among individuals with LDL C <100 mg/dl, those with non HDL C >130 mg/dl had a HR for future CHD of 1.84 (95% CI: ), a finding that confirms that increased risk is associated with elevated non HDL C, even in those with low LDL C concentrations [29]. The Bypass Angioplasty Revascularization Investigation examined baseline lipid levels in 1514 patients (73% men; mean age: 61 years). All of the patients had multivessel coronary artery disease. The study followed patients for a mean of 5 years, examining outcomes of death, nonfatal myocardial infarction and death, or myocardial infarction, using univariate and multivariate time-dependent proportional hazard methods. Angina pectoris at 5 years was modeled using univariate and multivariate logistic regression. While LDL C and HDL C did not predict events at follow-up, non HDL C was a strong and independent predictor of nonfatal myocardial infarction (RR: 1.049; 95% CI: ) and angina pectoris (RR: 1.049; 95% CI: ; p < 0.05 for both, but not mortality) [30]. In order to assess whether non HDL C and ApoB were more strongly associated with the risk of future cardiovascular events than LDL C (primary objective) among statin-treated patients and to determine whether non HDL C and ApoB explained a larger proportion of Review the atheroprotective effects of statin therapy than LDL C (secondary objective), Boekholdt et al. performed a meta-analysis of randomized controlled statin trials in which lipids and apolipoproteins were determined in all study participants at baseline and at 1 year follow-up [31]. They identified eight trials published between 1994 and 2008, contacted the investigators and evaluated individual patient data of 62,154 patients. HR and corresponding 95% CI for risk of major cardiovascular events were adjusted for established risk factors by one standard deviation increase in LDL C, non HDL C and ApoB. Among 38,153 statintreated subjects, the adjusted HR and 95% CI per one standard deviation were 1.13 ( ) for LDL C, 1.16 ( ) for non HDL C and 1.14 ( ) for ApoB. These HRs were significantly higher for non HDL C than LDL C (p = 0.002) and ApoB (p = 0.02). Thus, among these statin-treated patients, on-treatment levels of all three measures were associated with increased risk of cardiovascular events, but the association slightly favored non HDL C [31]. This mild statistical advantage is of questionable clinical significance. Among the prospective trials examining statin therapy, the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin is unique in that inclusion in the study was restricted to adults without diabetes or CVD, and who had baseline LDL C <130 mg/dl, high-sensitivity CRP 2 mg/l and triglyceride concentrations <500 mg/dl. Subjects received rosuvastatin 20-mg daily or placebo, and the primary end point was incident nonfatal myocardial infarction or stroke, hospitalization for unstable angina, arterial revascularization, or cardiovascular death. The authors used separate multivariate Cox models and found statistically significant associations of a similar magnitude (with 95% CI) with residual risk for CVD for on-treatment LDL C (1.31; ), non HDL C (1.25; ), ApoB (1.27; ), TC/HDL C (1.22; ), LDL C/HDL C (1.29; ) and ApoB/A1 (1.27; ). Among those subgroups achieving an LDL C <70 mg/dl, non HDL C 100 mg/dl or ApoB 80 mg/dl, the residual risk associated with these measures were mostly no longer statistically significant. This study demonstrated that on-treatment LDL C was as valuable as non HDL C in determining residual risk in this aggressively treated population [32]. 73

7 Review Orringer A study examined the association of mean absolute ApoB reduction with RR of CHD (nonfatal myocardial infarction and CHD death), stroke (nonfatal and fatal) or CVD (CHD, stroke and coronary revascularization), and compared its ability to predict CVD events with non HDL C. This Bayesian random effects meta-analysis included 25 studies (n = 131,134): 12 on statin, four on fibrate, five on niacin, two on simvastatin/ezetimibe, one on ileal bypass surgery and one on aggressive versus standard LDL C and blood pressure targets. Each 10 mg/dl decrease in ApoB was associated with a 10% reduction in CHD, no decrease in stroke and a 6% decrease in major CVD risk. Non HDL C decrease modestly outperformed ApoB for prediction of CHD (Bayes factor: 1.45) and CVD (Bayes factor: 2.07) risk decrease. In the 12 statin trials, ApoB and non HDL C decreases similarly predicted CVD risk and ApoB improved CHD risk prediction when added to non HDL/LDL C decrease (Bayes factor: 3.33), but did not improve stroke risk prediction when added to non HDL C/LDL C decrease. The study concluded that across all drug classes, ApoB decreases did not consistently improve risk prediction over LDL C and non HDL C decreases. In statin-treated subjects, ApoB decreases added information to LDL C and non HDL C decreases for CHD risk prediction, but not for stroke or overall cardiovascular risk decrease [33]. The Heart Protection Study was a randomized controlled trial that enrolled 20,536 men and women, aged years, with a history of CHD, cerebrovascular disease or other occlusive disease of noncoronary arteries, or Type 1 or 2 diabetes mellitus. The study also included men aged 65 years undergoing treatment for arterial hypertension. Subjects were randomly assigned to receive 40 mg simvastatin daily, matching placebo and antioxidant vitamins, or placebo. They were followed for a mean of 5.3 years for the incidence of myocardial infarction, stroke, vascular procedures and hospital admissions for other cardiac events. Associations between vascular events and baseline concentrations of lipid fractions, ApoB and A1, and lipoprotein particles were assessed by NMR. Major occlusive events were found to be equally strongly associated with lipids and atherogenic particle measures. Adjusted HR per one additional standard deviation higher with 95% CI were 1.25 ( ) for LDL C, 1.23 ( ) for non HDL C, 1.25 ( ) for ApoB and 1.25 ( ) for LDL P. The authors concluded that in this population with a 2% average coronary 74 Clin. Lipidol. (2013) 8(1) event rate per year, lipids and apolipoprotein, and LDL P had similar predictive values for incident major occlusive vascular events [34]. The use of non HDL C has a number of advantages, including: well-established cutpoints associated with increased risk, low cost of screening, rapid turnaround of results, universal availability and high-level consensus for its clinical value, as indicated by its incorporation into the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) guidelines. Despite these advantages, there remains significant physician knowledge gaps that have hampered physician acceptance and clinical utilization of non HDL C for patient management decisions [35]. The pros & cons of using ApoB While the clinical use of non HDL C is advo cated in those with hypertriglyceridemic states, patients with these disorders most often have a reduced content of cholesterol per LDL particle, with a resultant discordance between LDL P and LDL C. The identification of patients with LDL P/LDL C discordance is of prognostic importance. As was demonstrated in the Framingham Offspring Study [36] and in the Multi-Ethnic Study of Atherosclerosis cohort [37], when discordance between LDL P and LDL C is present, risk for CVD events tracks more closely with LDL P than with LDL C. A discordance analysis of ApoB and non HDL C as markers of cardiovascular risk was performed in the INTERHEART study, a standardized case control study of acute myocardial infarction examining blood samples of 9345 cases and 12,120 controls in 52 countries [38]. Concentrations of non HDL C and ApoB were expressed as percentiles. Concordance was defined as the phenotype when the percentile of non HDL C was equivalent to the percentile of ApoB. Discordance was defined as the phenotype when the percentile of non HDL C was >ApoB (cholesterol-enriched LDL particles) or <ApoB (cholesterol-depleted LDL particles). Using definitions of discordance ranging from 1 to 10% and among all ethnicities, risk was consistently higher in those with non HDL C <ApoB. These findings indicate that ApoB is a better risk marker in this population than non HDL C [38] Sniderman et al. provided four reasons why they believe that ApoB should be measured for CHD risk assessment. First, it identifies those individuals with Type 2 diabetes and the metabolic syndrome who have elevated

8 Non HDL cholesterol, ApoB & LDL particle concentration in coronary heart disease risk atherogenic particle concentration, independent of their levels of LDL C. Second, measurement of ApoB, TC and triglycerides allows the clinician to properly classify all of the atherogenic dyslipidemias, including identification of familial combined hyperlipidemia and familial dysbetalipoproteinemia, both of which are associated with a high risk of CVD. Third, the accuracy of measure ment of ApoB has become increasingly reliable relative to traditional lipid measure ment. Finally, they contend that although the on-statin trials show that non HDL C and ApoB are generally equivalent risk markers, those with persistently high levels of ApoB may be candidates for intensification of statin therapy [26]. In certain populations, the measurement of atherogenic particle concentration is of particular value. This observation is particularly true in patients with central obesity, insulin resistance, hyperglycemia and dyslipoproteinemia, in whom there is considerable variation in ApoB concentration for any given level of non HDL C [39]. Patients with Type 2 diabetes are at a high risk for initial and recurrent cardiovascular events. A study was performed in Type 2 diabetics to examine how often elevated LDL P is present, even when LDL C and non HDL C are very low [40]. The laboratory data base at Liposcience, Inc. (NC, USA) was searched for patients with the diagnosis code of Type 2 diabetes ( ) and a group of 1970 patients, mean age of 61 years, 51% of which were men, was identified who had, using standardized automated methods, LDL C <70 mg/dl (calculated using the Friedewald equation), non HDL C <100 mg/dl, triglycerides <150 mg/dl and HDL C >40 mg/dl. Lipoprotein particle concentration was calculated using NMR spectroscopy. In this group the mean lipid values in mg/dl were TC 126, triglycerides 83, HDL C 51 and LDL C 58. The mean LDL P was 907 nmol/l. While an LDL C of 58 mg/dl represents the second percentile of those described in the Multi-Ethnic Study of Atherosclerosis cohort, only 22% had a correspondingly low LDL P (<700 nmol/l) and 34.7% had LDL P values >1000 nmol/l, representing the 20th percentile of this cohort. When those with LDL C <50 mg/dl, HDL C >40 mg/dl and triglycerides <150 mg/dl were similarly examined, 16% had LDL P <500 nmol/l and 14% had LDL P levels above 1000 mg/dl. Finally, in the group with non HDL C <80 mg/dl, HDL C> 40 mg/dl and triglycerides <150 mg/dl, only 8% had LDL P <500 nmol/l and 25% had LDL P >1000 nmol/l [40]. The results of this study Review remind the clinician that many diabetics, even those with a low LDL C level, may have a higher LDL P than would be expected based upon their LDL C, a finding associated with increased CHD risk. A divergent perspective about the value of adding apolipoprotein testing to traditional risk factors was provided by another article from the Emerging Risk Factors Collaboration [41]. They performed a meta-analysis examining individual records of 165,544 participants without baseline CVD in 57 prospective cohorts recruited between 1968 and 2007, in which there were 15,126 incident fatal or nonfatal CVD outcomes (10,132 CHD and 4994 stroke outcomes) during a median follow-up of 10.4 years (interquartile range: years). Among 139,581 participants with information on apolipoproteins, their data suggested the controversial conclusion that information on non HDL C and the apolipoproteins did not add to risk prediction provided by simple measurement of TC and HDL C. They also concluded that adding such information to risk scores already reporting TC, HDL C and other conventional risk factors resulted in only slight improvement in CHD risk prediction, as determined by the C Index, but did not improve reclassification across the clinical cut-off levels currently used to make treatment decisions [41]. In the 10 to <20% 10-year predicted CVD risk population, if this type of targeted measurement were coupled with statin-treatment decisions as recommended by the NCEP ATP III guidelines, the use of ApoB and ApoA1 would help to prevent only one extra CVD outcome for every 4541 patients screened over 10 years [41]. When the clinician chooses to employ ApoB testing in clinical decision-making, there are several cautions that should be employed. First of all, there are increased costs associated with the measurement of ApoB and LDL P. In addition, the lack of universal availability of these tests serves as an impediment to their more widespread clinical utilization. While ApoB goals of <90 and <80 mg/dl were suggested in a consensus statement by the American Diabetes Association and American College of Cardiology Foundation for patients at high and highest cardiometabolic risk, respectively, and with concomitant lipoprotein abnormalities [39], there remains a general lack of consensus as to how aggressively to treat patients who have excessive atherogenic particle concentration after LDL C and non HDL C goals have been attained. No 75

9 Review Orringer study has prospectively demonstrated that a strategy based upon treatment to particle goal is superior in outcomes, equally safe, as well tolerated or more cost effective than treating to NCEP ATP III lipid management guidelines with a non HDL C goal <30 mg/dl above the LDL C goal. Conclusion CHD is a disease that is accelerated by excessive atherogenic particle concentration. An understanding of the role of ApoB particles in atherogenesis and an appreciation of the strengths and limitations of various lipid and lipoprotein measures allows the clinician to diagnose and treat patients with excessive concentration of atherogenic lipoproteins more accurately. Non HDL C, as a surrogate for atherogenic particle concentration, and ApoB or LDL P, as more direct measures, are likely to be superior to LDL C for initial CHD risk assessment. At this time, studies examining the value of these biomarkers to predict risk in those patients on treatment do not support the contention on a population basis that ApoB or LDL P measurement is superior to non HDL C, although atherogenic particle testing provides the advantage of identifying those with LDL P/LDL C discordance, a group that potentially has increased residual risk of CHD. Future perspective In an era of escalating healthcare expenditures, CHD remains a major contributor to that expense. Accurate, early and cost-effective identification of the patient at risk for coronary events is a high priority. RR for a coronary event, as measured by lipid or apolipoprotein testing, has clinical relevance only when the global risk for such an event is appreciated. Further refinement of currently available global risk scoring systems will improve the accuracy of initial risk stratification. Subclinical atherosclerosis imaging as a means of confirming or reclassifying absolute risk in intermediate-risk patients and the judicious employment of other biomarkers in appropriate patients will provide the clinician with additional clinically relevant tools. Carotid imaging, by identifying the spectrum of disease from normal, to increased intimamedia thickness, to plaque, to obstructive disease, noninvasively identifies patients at increased risk of both coronary and cerebrovascular disease, but requires high-end ultrasound instrumentation using highly standardized protocols for performance and interpretation of 76 Clin. Lipidol. (2013) 8(1) the studies, and highly trained technicians and readers. Although its use has been given a level IIa recommendation (it is reasonable to perform the test) by the American College of Cardiology Foundation/American Heart Association Guidelines for Assessment of Cardiovascular Risk in Asymptomatic Adults, the results have not been integrated with treatment. There is no current consistent insurance reimbursement for carotid intima-media thickness measurement in the USA. Coronary calcium scoring provides valuable information on future coronary risk in asymptomatic intermediate-risk patients and is a more powerful tool for the reclassification of absolute CHD risk than carotid intima-media thickness measurement. It is also given a class IIa level of recommendation in intermediate-risk patients, but it is not widely available at the point of care, entails radiation exposure, is expensive, generally not reimbursed by insurance and the results are not yet integrated with treatment. However, the ability to identify those at extremely low 5 10-year CHD risk (those with Agatston scores of zero) and those at high risk (those with Agatston scores 400 units), in addition to its ability to reclassify absolute risk, makes calcium scoring an attractive tool for future research. The robust database supporting the use of high-sensitivity CRP in selected intermediaterisk patients, particularly women, is well established. Its demonstrated value to identify those asymptomatic men aged 50 years and women aged 60 years with an LDL C <130 mg/dl who would benefit from potent statin therapy has resulted in its being given a class IIa recommendation in this population. The value of other inflammatory markers, including lipoprotein PLA-2, myeloperoxidase and others, remains to be determined. The future role of other testing, including cardiac computed tomographic imaging, magnetic resonance plaque imaging, genetic testing and genomics awaits further research. Financial & competing interests disclosure The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript.

10 Non HDL cholesterol, ApoB & LDL particle concentration in coronary heart disease risk Review Executive summary Biomarkers: definitions & standards for clinical utility A biomarker is a clinical test that is used to predict disease and/or to assist in more effective treatment of disease. It is used as an adjunct to global risk scoring to define the relative risk of a future clinical event. ApoB 100 particle synthesis, metabolism & role in atherogenesis ApoB-containing particles are lipoprotein particles that are produced in the liver and deliver fatty acids to muscle and adipose tissue, and cholesteryl ester to peripheral tissues. A total of 90% of ApoB particles are LDL particles. Atherogenesis is propagated by excessive plasma atherogenic particle concentration and endothelial dysfunction, resulting in the gradient-driven diffusion of ApoB-containing particles from the vascular lumen into the arterial intima. Atherogenic particle retention and expression of cytokines in the arterial wall promote plaque formation and rupture, resulting in acute coronary syndromes. Reliability of lipoprotein measurement for coronary heart disease risk assessment The Friedewald equation, the standard means of calculating LDL cholesterol (LDL C), loses it accuracy as a measure of the concentration of cholesterol in plasma LDL particles when triglycerides exceed 400 mg/dl. In patients with plasma triglycerides of mg/dl, non HDL cholesterol (HDL C) is a secondary target of therapy after the LDL C goal has been achieved. Non HDL C concentration is highly correlated, but may be discordant with that of ApoB. Non HDL C as a coronary heart disease risk marker & on-treatment guide to therapy An elevated concentration of non HDL C is a strong predictor of the presence of atherosclerotic vascular disease and generally outperforms LDL C in coronary heart disease (CHD) initial risk assessment. There is debate on whether non HDL C or ApoB better predicts the risk for initial coronary events. Non HDL C measurement is less costly than ApoB or LDL particle concentration. The pros & cons for using ApoB Measurement of ApoB helps to identify those patients with LDL particle concentration/ldl C discordance, a finding that is most often seen in patients with central obesity, insulin resistance, hyperglycemia and dyslipoproteinemia. In those populations in whom the percentile of concentration of ApoB exceeds that of non HDL C the risk of coronary events is increased. A meta-analysis of 57 prospective studies called into question the additive value of ApoB measurement to traditional lipid values in predicting CHD and stroke. No study has demonstrated improved outcomes, safety or cost effectiveness of using atherogenic particle goals as compared with traditional lipid goals in CHD risk management. References Papers of special note have been highlighted as: n of interest nn of considerable interest Greenland P, Alpert JS, Beller GA et al ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J. Am. Coll. Cardiol. 56(25), e50 e103 (2010). Morrow DA, de Lemos JA. Benchmarks for the assessment of novel cardiovascular biomarkers. Circulation 115, (2007). PO Kwiterovich Jr. Lipid, apolipoprotein and lipoprotein metabolism: implications for the diagnosis and treatment of dyslipidemia. In: The Johns Hopkins Textbook of Dyslipidemia. Kwiterovich PO Jr (Ed.). Lippincott Williams and Wilkins, PA, USA, 1 21 (2009). Walldius G, Jungner I. Apolipoprotein B and apolipoprotein A-I: risk indicators of coronary heart disease and targets for lipid-modifying therapy. J. Intern. Med. 255(2), (2004). 5 6 nn 7 Sniderman AD, de Graaf J, Couture P. ApoB and the atherogenic apob dyslipo proteinemias. In: The Johns Hopkins Textbook of Dyslipidemia. Kwiterovich PO Jr (Ed.). Lippincott Williams and Wilkins, PA, USA, (2009). Davidson MH, Ballantyne CM, Jacobson TA et al. Clinical utility of inflammatory markers and advanced lipoprotein testing: advice from an expert panel of lipid specialists. J. Clin. Lipidol. 5(5), (2011). 8 Tabas I. Subendotheilal lipoprotein retention as the initiating process in atherosclerosis: update and therapeutic implications. Circulation 116, (2007). 9 Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults: Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 285(19), (2001). 10 Conroy RM, Pyorala K, Fitzgerald AP et al. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur. Heart J. 24(11), (2003). 11 Cullen P, Schulte H, Assmann G. The Münster Heart Study (PROCAM). Total mortality in middle-aged men is increased at low total and LDL cholesterol concentrations in smokers but not in nonsmokers. Circulation 96, (1997). Overview by an expert panel of lipidologists on the use of biomarkers in initial coronary heart disease (CHD) risk prediction and on-treatment management decisions. Libby P, Ridker PM, Hansson GK; Leducq Transatlantic Network on Atherothrombosis. Inflammation in atherosclerosis: from pathophysiology to practice. J. Am. Coll. Cardiol. 54(23), (2009). 77

11 Review Orringer n Paul S. Bachorik. Measurement of lipids, lipoproteins and apolipoproteins. In: The Johns Hopkins Textbook of Dyslipidemia. Kwiterovich PO Jr (Ed.). Lippincott Williams & Wilkins, PA, USA, (2009). defining the role of low-density lipoprotein heterogeneity in coronary artery disease. J. Am. Coll. Cardiol. 50(18), (2007). 24 Ramjee V, Sperling LS, Jacobson TA. Non-high-density lipoprotein cholesterol versus apolipoprotein B in cardiovascular risk stratification: do the math. J. Am. Coll. Cardiol. 58(5), (2011). Sniderman AD, St-Pierre A, Cantin B. Concordance/discordance between plasma apoliporotein B levels and the cholesterol indexes of atherosclerotic risk. Am. J. Cardiol. 91, (2003). 25 nn Sniderman AD, Williams K, de Graaf J. Non HDL C equals apolipoprotein B: except when it does not! Curr. Opin. Lipidol. 21, (2010). Baruch L, Agarwahl S, Gupta B et al. Is directly measured low-density lipoprotein clinically equivalent to calculated low-density lipoprotein? J. Clin. Lipidol. 4, (2010). Cooper G, Smith S, Myers G. Estimating and minimizing effects of biologic sources of variation by relative range when measuring the mean of serum lipids and lipoproteins. Clin. Chem. 42, (1994). Schectman G, Patsches M, Sasse E. Variability in cholesterol measurements: comparison of calculated and direct LDL cholesterol determinations. Clin. Chem. 42, (1996). Bachorik P, Ross J. National Cholesterol Education Program recommendations for measurement of low-density lipoprotein cholesterol: executive summary. Clin. Chem. 41, (1995). Sniderman A, McQueen M, Contois J, Williams K, Furberg CD. Why is non-highdensity lipoprotein cholesterol a better marker of the risk of vascular disease than low-density lipoprotein cholesterol? J. Clin. Lipidol. 4(3), (2010). Contois JH, Warnick GR, Sniderman AD. Reliability of low-density lipoprotein cholesterol, non-high-density lipoprotein cholesterol, and apolipoprotein B measurement. J. Clin. Lipidol. 5, (2011). 26 nn 27 n Reviews the pitfalls in the interpretation of reported lipid and lipoprotein measurements. Otvos JD, Jeyarajah EJ, Cromwell WC. Measurement issues related to lipoprotein heterogeneity. Am. J. Cardiol. 90(8A), i22 i29 (2002). Mudd JO, Borlaug BA, Johnston PV et al. Beyond low-density lipoprotein cholesterol: Rainwater DL, McMahan CA, Malcom GT et al. Lipid and apolipoprotein predictors of atherosclerosis in youth: apolipoprotein concentrations do not materially improve prediction of arterial lesions in PDAY subjects. Arterioscler. Thromb. Vasc. Biol. 19, (1999). The Emerging Risk Factors Collaboration. Major lipids, apolipoproteins and risk of vascular disease. JAMA 302, (2009). Meta-analysis examining the value of measurement of lipids versus apolipoprotein biomarkers for CHD risk prediction, concluding that apolipoprotein measurement does not add to the information provided by lipid measurement. Sniderman AD, Williams K, Contois JH et al. A meta-analysis of low-density lipoprotein cholesterol, non-high-density lipoprotein cholesterol, and apolipoprotein B as markers of cardiovascular risk. Circ. Cardiovasc. Qual. Outcomes 4(3), (2011). Meta-analysis challenging the conclusions of the Emerging Risk Factors Collaboration Study and making a case for the importance of ApoB measurement in initial CHD risk assessment. Robinson JG, Wang S, Smith BJ, Jacobson TA. Meta-analysis of the relationship between non HDL C reduction and CHD risk. J. Am. Coll. Cardiol. 53, (2009). Meta-analysis demonstrating the relationship between non HDL cholesterol lowering and reduction in CHD risk in patients treated with a variety of pharmacologic and nonpharmacologic interventions. Cui Y, Blumenthal RS, Flaws JA et al. Non-high-density lipoprotein cholesterol as a predictor of cardiovascular disease mortality. Arch. Intern. Med. 161, (2001). Arsenault BJ, Rana JS, Stroes ES et al. Beyond low density lipoprotein cholesterol: respective contributions of non HDL C levels, triglycerides and total/hdl ratio to CHD risk in apparently healthy men and women. J. Am. Coll. Cardiol. 55, (2009). Bittner VA. Non HDL C levels predict five year outcomes in the BARI. Circulation 106, (2002). Boekholdt SM, Arsenault BJ, Mora S et al. Association of LDL cholesterol, non HDL cholesterol, and apolipoprotein B levels with risk of cardiovascular events among patients Clin. Lipidol. (2013) 8(1) treated with statins: a meta-analysis. JAMA 307(12), (2012). n Meta-analysis demonstrating the value of non HDL cholesterol measurement in the prediction of CHD events in statin-treated patients. 32 Mora S, Glynn R. On-treatment non HDL C, apob, triglycerides, and lipid ratios in relation to vascular risk after treatment with potent statin therapy. J. Am. Coll. Cardiol. 59, (2012). 33 Robinson JG, Wang S. Meta-analysis of comparison of effectiveness of lowering apolipoprotein B versus LDL C and non HDL C for cardiovascular risk reduction in randomized trials. Am. J. Cardiol. 110(10), (2012). 34 Parish S, Offer A, Clarke R et al. Lipids and lipoproteins and risk of different vascular events in the MRC/BHF Heart Protection Study. Circulation 125(20), (2012). 35 Virani SS. Barriers to non HDL cholesterol goal attainment by providers. Am. J. Med. 124, (2011). 36 Cromwell WC. LDL particle number and risk of future cardiovascular disease in the Framingham Offspring Study. J. Clin. Lipidol. 1, (2007). 37 Otvos JD, Mora S. Clinical implications of discordance between low-density lipoprotein cholesterol and particle number. J. Clin. Lipidol. 5, (2011). 38 Sniderman AD, Islam S, Yusuf S, McQueen MJ. Discordance analysis of apolipoprotein B and non-high density lipoprotein cholesterol as markers of cardiovascular risk in the INTERHEART study. Atherosclerosis 225, (2012). 39 Brunzell JD, Davidson M, Furberg CD et al. Lipoprotein management in patients with cardiometabolic risk. Diabetes Care 31, (2008). 40 Malave H, Castro M. Evaluation of LDL P distribution in patients with Type 2 diabetes with LDL C <50 mg/dl and non HDL C <80 mg/dl. Am. J. Cardiol. 110, (2012). 41 Di Angelantonio E, Gao P, Pennells L et al.; The Emerging Risk Factors Collaboration. Lipid-related markers and cardiovascular disease prediction. JAMA 307, (2012). n Meta-analysis calling into question the value of employing ApoB measurement in addition to standard risk factors for cardiovascular disease risk reclassification or for statintreatment decisions in intermediate-risk populations.