Lipoprotein(a)

INTRODUCTION — Elevated serum lipoprotein(a), also referred to as Lp(a), is a risk factor for atherosclerotic cardiovascular disease (ASCVD). There is likely a causal relationship between high Lp(a) and the development of ASCVD and aortic valve stenosis.

This topic will review the association between Lp(a) and ASCVD and its clinical implications. The association between Lp(a) and aortic valve sclerosis is discussed separately. (See "Aortic valve sclerosis and pathogenesis of calcific aortic stenosis", section on 'Pathogenesis of calcific aortic valve disease'.)

STRUCTURE AND FUNCTION — Lp(a) is a type of low-density lipoprotein in which a large glycoprotein, apolipoprotein(a) (apo(a)), is covalently bound to apolipoprotein B by a disulfide bridge [1]. The apo(a) chain contains five cysteine-rich domains known as "kringles" [2]. The fourth kringle is homologous with the fibrin-binding domain of plasminogen, a plasma protein that dissolves blood clots when activated.

The absolute concentration of Lp(a) particles, rather than apo(a) isoform size, mediates the relationship between Lp(a) and atherosclerotic cardiovascular disease risk [3].

MECHANISMS AFFECTING ATHEROSCLEROTIC CARDIOVASCULAR DISEASE RISK — High Lp(a) may promote atherosclerosis as well as thrombosis by affecting fibrinolysis, inflammation, endothelial function, and oxidative stress:

●Interfering with fibrinolysis – Due to apolipoprotein(a) (apo(a))'s structural similarity to plasminogen, apo(a) may interfere with fibrinolysis by competing with plasminogen binding to molecules and cells [4,5]. However, this property has not been established with Lp(a), suggesting impaired fibrinolysis may not be a clinically relevant property of Lp(a). In addition, therapeutic lowering of Lp(a) with an mRNA inhibitor (pelacarsen) did not alter fibrinolysis [4,6-9]. Lp(a) also increases tissue-factor-mediated thrombosis and inhibition of clot fibrinolysis [10].

●Binding to macrophages – Lp(a) binds to macrophages by a high-affinity receptor that promotes foam cell formation and deposition of cholesterol in atherosclerotic plaques [11]. The macrophage very low-density lipoprotein receptor can bind to and mediate catabolism of Lp(a) by endocytosis, leading to its degradation within lysosomes [12] and intracellular accumulation of lipid within macrophages. Supporting this hypothesis is the observation that Lp(a) is ubiquitous in human coronary atheroma, colocalizes with plaque macrophages, and is detected in larger amounts in tissue from culprit lesions in patients with unstable compared with stable coronary disease [13].

●Binding to the endothelium and components of the extracellular matrix [14]. This may interfere with normal endothelial function [15].

●Increasing expression of intercellular adhesion molecule-1, resulting in the recruitment of monocytes to the arterial wall and binding to macrophages [16], promoting foam cell formation and the localization of Lp(a) in atherosclerotic plaques [11,17,18].

●Oxidized phospholipids on Lp(a) trigger inflammation by toll-like receptor 2-mediated pathway [19] and increase arterial wall inflammation [20,21]. The cardiovascular risk associated with Lp(a) and oxPL:apoB is modified by inflammatory interleukin-1 haplotype, thereby reinforcing the mechanistic link between Lp(a) and immune dysregulation [22].

GENETICS — Serum Lp(a) levels are genetically determined [23]; however, levels can increase in response to interleukin-6 [24]. In families of European descent without familial hypercholesterolemia, greater than 90 percent of the variability in Lp(a) levels can be explained by polymorphisms at the apolipoprotein(a) (apo(a)) gene locus (isoforms), also referred to as the LPA gene (Online Mendelian Inheritance in Man 152200) [25]. One important LPA polymorphism is the kringle IV type 2 size polymorphism, which results in a large number of differently sized isoforms of apo(a) [26]. There is a strong inverse relationship between the size of the apo(a) isoforms and the Lp(a) concentrations [25,27-29]. A significant proportion (30 to 60 percent) of the population variation in Lp(a) levels is determined by this polymorphism [30]. However, the molar concentration of Lp(a), not the apo(a) size, determines risk for coronary artery disease, as discussed below [3].

The distribution of serum Lp(a) concentrations is highly skewed toward lower values among most racial/ethnic groups; however, Black Americans, Africans, and individuals from India have a more normal distribution, centered at a higher mean Lp(a) level [29,31].

MEASUREMENT OF SERUM LP(A) CONCENTRATION — Measurement of Lp(a) concentrations has been challenging due to the complexity of this lipoprotein, the variation of lipid composition of Lp(a) within and between individuals, and the large variation in the size of the apo(a) protein moiety, also called isoform, between individuals [32]. Different laboratories may use different methods for measuring Lp(a) concentration.

It is our recommendation that clinicians choose a laboratory which has been certified by a reputable independent reference laboratory. A recommendation from a lipid or laboratory medicine expert in the clinician’s region of practice may be helpful in selecting a laboratory to utilize for the reliable measurement of Lp(a) levels.

Preferred assays report the Lp(a) particle number in molar concentration units, typically nmol/L, rather than in mass units (mg/dL). Lp(a) molar concentration by enzyme-linked immunosorbent assay (ELISA) better predicts atherosclerotic cardiovascular disease events than measures using nephelometry (turbidity) or those that measure Lp(a) cholesterol content [32]. Measurements cannot be converted accurately from mg/dL to nmol/L and vice versa. One conversion commonly used is to multiply the Lp(a) mass by 2.4 to provide a very rough estimate of Lp(a) molar concentration but the variability is substantial and is not recommended for clinical use [33-35].

Low-density lipoprotein cholesterol (LDL-C) measurements include the cholesterol component of Lp(a). In an occasional patient, a substantial fraction of LDL-C may be carried in Lp(a) particles rather than in normal LDL. In these cases, treatment with a statin may lead to a lesser reduction in LDL-C than expected, as statins do not effectively reduce Lp(a) levels. This would likely be notable only in patients with very high Lp(a) levels, very low LDL-C levels, or both. Estimating the cholesterol content of Lp(a) from the Lp(a) mass measurement is not a very reliable technique. Conventionally, the cholesterol content of the Lp(a) mass is estimated at 30 to 35 percent.

Lp(a) levels are stable in healthy individuals over time [36], though age-related increases in Lp(a) concentrations due to sex steroid deficiency, acute and chronic inflammation, and reductions in renal function can be observed. Repeat measurements of Lp(a) typically are not necessary, though if a clinician wishes to evaluate the impact of an intervention on Lp(a) concentrations, the same assay should be used whenever two values are being compared.

DISEASE ASSOCIATIONS

Cardiovascular events — Lp(a) is an independent risk factor for coronary heart disease, atherosclerotic cardiovascular disease (ASCVD), cerebrovascular disease, and aortic stenosis.

Although some evidence suggests that measurement of apolipoprotein(a) isoforms may improve ASCVD risk prediction, its contribution is considered to be small [26,33-36]. The risk associated with high Lp(a) may be greater in the context of systemic inflammation [37,38]. There is evidence that Lp(a) molar concentration is associated with ASCVD independent of Lp(a) mass but not the converse [3]. For this reason, we suggest relying more on recent evidence that uses the preferred molar concentration method for Lp(a) rather than older studies that used the Lp(a) mass measurement.

●ASCVD – Genetic and experimental data suggest Lp(a) contributes to the pathogenesis of ASCVD [26,39]. In the United Kingdom Biobank Study of 460,000 persons, Lp(a) levels were associated with higher ASCVD in a log-linear fashion for levels above the median (20 nmol/L) [40] The standardized risk for ASCVD was 11 percent higher for each increment of 50 nmol/L (hazard ratio 1.11 per 50 nmol/L, 95% CI 1.10-1.12). This association did not differ by race-ethnicity (White, Black, South Asian, Chinese). Similarly, in an analysis of statin-treated cardiovascular disease patients (with low-density lipoprotein cholesterol levels <70 mg/dL) enrolled in the FOURIER randomized controlled trial of PCSK9-inhibitor evolocumab versus placebo, patients with baseline Lp(a) in the highest quartile (>165 nmol/L) had a higher risk of coronary heart disease death, myocardial infarction, or urgent revascularization compared with those in the lowest quartile (hazard ratio 1.22, 95% CI 1.01-1.48) [41].

Older studies suggested Lp(a) is associated with ASCVD regardless of statin treatment [42,43]. Use of the older assay (using mg/dL) limits the interpretation and generalizability of these older studies.

●Cerebrovascular disease – Lp(a) is associated with cerebrovascular disease. In a systematic review including six studies [44], the relative risk for ischemic stroke related to high Lp(a) was 2.14 (95% CI 1.85-2.97) [45]. The association between Lp(a) and cerebrovascular disease may be stronger in men than in women [46-50].

●Aortic stenosis – Specific genotypic variants in LPA (the gene encoding a significant portion of Lp(a) have been associated with aortic valve calcium and aortic stenosis in population studies) [51]. Plasma levels of Lp(a) vary by rs10455872 genotype between 20 and 50 mg/dL per minor allele. In a meta-analysis of five studies with >131,000 people, clinically significant aortic stenosis was significantly associated with a specific genetic variant within LPA (rs10455872; risk ratio 1.65, 95% CI 1.43-1.90).

RELATIONSHIP WITH FAMILIAL HYPERCHOLESTEROLEMIA — The prevalence of high Lp(a) is higher in people with heterozygous familial hypercholesterolemia (FH) than those without it. In FH patients, elevated Lp(a) further increases atherosclerotic cardiovascular disease risk [52,53]. The cause(s) of the association between FH and high Lp(a) is not known. (See "Familial hypercholesterolemia in adults: Overview".)

A study of nearly 40,000 individuals with FH came to the following conclusions [53]:

●The increase in prevalence of Lp(a) in patients with FH was explained by an increased frequency of a single-risk allele (18.8 versus 8.8 percent; p<0.05).

●Since Lp(a) contains low-density lipoprotein cholesterol (LDL-C), FH individuals with a risk allele have higher LDL-C levels than those without and are thus more likely to come to clinical attention.

SCREENING — The indications for screening are not entirely agreed on, and there is some variation in the practices of our experts. For the most part, our experts practice in a manner similar to that recommended by the National Lipid Association as presented below.

The European Society of Cardiology recommends that Lp(a) measurement should be considered at least once in each adult person's lifetime to identify those with very high inherited Lp(a) levels >180 mg/dL (>430 nmol/L) who may have a very high lifetime risk of atherosclerotic cardiovascular disease (ASCVD) similar to those with heterozygous familial hypercholesterolemia (FH) [54].

The American Heart Association Consensus Statement on Lipoprotein(a) recommends that people have at least one Lp(a) measurement during their lifetime [55].

The American Heart Association/American College of Cardiology cholesterol guidelines do not comment on who to screen but state that Lp(a) ≥50 mg/dL or ≥125 nmol/L is an ASCVD "risk-enhancing factor" that in patients 40 to 75 years old without diabetes mellitus but with 10-year ASCVD risk 7.5 to 19.9 percent would favor initiation of statin therapy [56].

The National Lipid Association stated that Lp(a) testing is reasonable to refine ASCVD risk in adults with [57]:

●First-degree relatives with premature ASCVD (<55 years of age in men; <65 years of age in women).

●A personal history of premature ASCVD.

●Primary severe hypercholesterolemia (low-density lipoprotein cholesterol [LDL-C] ≥190 mg/dL) or suspected FH. (See 'Relationship with familial hypercholesterolemia' above.)

We agree with the National Lipid Association. In addition, Lp(a) testing may be reasonable in adults:

●To aid in the clinician-patient discussion about whether to prescribe a statin in those aged 40 to 75 with borderline (5 to 7.4 percent) 10-year ASCVD risk.

●To identify a possible cause for a less-than-anticipated LDL-C lowering.

●To use in cascade screening of family members with severe hypercholesterolemia.

●To identify those at risk for progressive ASCVD.

We agree with the AHA Scientific Statement on Lp(a) that recommends verifying the Lp(a) with an additional test [55], particularly in patients with systemic inflammatory conditions due to the upregulation in Lp(a) that is induced by interleukin-6.

High Lp(a) or presence of certain Lp(a) polymorphisms are not by themselves indications for use of aspirin for primary ASCVD prevention. (See "Aspirin in the primary prevention of cardiovascular disease and cancer", section on 'Our approach'.)

MANAGEMENT — There is limited evidence indicating that Lp(a) lowering reduces atherosclerotic cardiovascular disease (ASCVD) risk (see 'Disease associations' above). Thus, except in very rare cases, we do not target Lp(a) with any therapy known to lower Lp(a). (See 'Next steps' below.)

Initial approach — Our initial approach to reducing ASCVD risk in patients with elevated Lp(a) is to reduce low-density lipoprotein cholesterol (LDL-C) to its target. (See "Management of low density lipoprotein cholesterol (LDL-C) in the secondary prevention of cardiovascular disease" and "Low-density lipoprotein cholesterol-lowering therapy in the primary prevention of cardiovascular disease".)

Usually this involves treatment with a statin, with or without ezetimibe. Statins increase Lp(a) levels [58]. However, the impact of this increase on cardiovascular events is not known and is felt to be small [42,59]. Ezetimibe does not lower Lp(a).

Some patients who cannot achieve an optimal LDL-C with statin plus ezetimibe are treated with a PCSK9 inhibitor. In the FOURIER trial, Lp(a) reduction observed with PCSK9 inhibitor therapy was associated with ASCVD risk reduction independent of LDL-C lowering, as discussed below [41]. Thus, very high-risk patients with high Lp(a) levels may benefit disproportionately from a lipid-lowering strategy that includes a PCSK9 inhibitor. In a post-hoc analysis of ODYSSEY outcomes, participants with LDL-C levels between 55 mg/dL and <70 mg/dL with a Lp(a) level above the median of 13.6 mg/dL had fewer cardiovascular disease events on treatment with alirocumab [60].

Next steps — For patients with elevated Lp(a) who have reached their LDL-C target or who have received all recommended therapies to lower LDL-C, we await definitive evidence from ongoing cardiovascular outcomes trials with selective Lp(a)-lowering therapies and do not have any broader recommendations on Lp(a) lowering.

The use of these therapies in this setting has not been proven in prospectively designed clinical outcome trials. In addition, there are costs to these therapies.

Some of us may consider PCSK9 inhibitors as second-line therapy for people with LDL-C or non-HDL-C above target who have Lp(a) excess.

PCSK9 inhibitors — In a meta-analysis of study-level data from 6566 patients in 12 randomized trials that compared treatment with PCSK9 antibody to no antibody, Lp(a) was lowered by approximately 26 percent [61]. In an exploratory analysis from the FOURIER trial, which randomly assigned statin-treated cardiovascular disease patients to evolocumab or placebo, evolocumab reduced the risk of coronary heart disease death, myocardial infarction, or urgent coronary revascularization [41]. After adjustment for change in LDL-C, each 25 nmol/L reduction in Lp(a) from evolocumab led to a 15 percent relative risk reduction (95% CI 2.0-26.0 percent). Evolocumab reduced the risk of coronary heart disease death, myocardial infarction, or urgent revascularization by 23 percent in patients with a baseline Lp(a) >median (37 nmol/L; hazard ratio 0.77, 95% CI 0.67-0.88) and in those with a baseline Lp(a) ≤median by 7 percent (hazard ratio 0.93, 95% CI 0.80-1.08; p interaction = 0.07).

In a prespecified analysis of the ODYSSEY Outcomes trial of patients with an acute coronary syndrome, alirocumab reduced Lp(a) by 5 mg/dL and reduced the risk of major adverse cardiovascular evens (hazard ratio 0.85, 95% CI 0.78-0.93) [44]. The beneficial impact of Lp(a) lowering by alirocumab was independent of its lowering of LDL-C. In this study, a 1 mg/dL reduction in Lp(a) mass was associated with an hazard ratio of 0.994 (95% CI 0.990-0.999) for ASCVD events. A subsequent post-hoc analysis of ODYSSEY Outcomes demonstrated that a higher (>13.7 mg/dL) versus lower Lp(a) (<13.7 mg/dL) in patients a recent acute coronary syndrome was associated with a higher cardiovascular event rate regardless of whether the LDL-C levels were <70 mg/dL or >70 mg/dL. Additionally, there was a reduction in cardiovascular events with alirocumab in the LDL-C subgroup <70 mg/dL among patients with a higher Lp(a) level (>13.7 mg/dL) [60].

Rarely used drugs — We rarely use nicotinic acid and we do not use estrogen replacement therapy for Lp(a) lowering:

●Nicotinic acid (2 to 4 g/day) can lower Lp(a) levels [62,63] by as much as 38 percent [64,65]. In addition, nicotinic acid lowers LDL-C, apoB-100, small LDL, and triglycerides and raises high-density lipoprotein cholesterol (HDL-C) levels. However, nicotinic acid has many side effects, and no prospective studies have demonstrated the effects of Lp(a) lowering with niacin in patients with high Lp(a). Indeed, in subgroup analyses of trials designed to investigate the efficacy of niacin on increasing HDL-C, niacin-mediated reductions in Lp(a) were not accompanied by improved clinical outcomes. Nicotinic acid is used infrequently by some of our contributors; others have stopped using it in their practices.

●Although estrogen replacement therapy reduces Lp(a) levels, it is not recommended for cardiovascular disease risk reduction [66,67]. (See "Menopausal hormone therapy and cardiovascular risk".)

Lipoprotein apheresis — Lipoprotein apheresis temporarily lowers Lp(a) by as much as 75 percent [65,68-71]; however, it is not clear if this strategy reduces ASCVD risk. Some observational studies in patients receiving maximally tolerated lipid-lowering therapy have suggested a clinical benefit [69]. In a study of 20 patients with refractory angina, Lp(a) levels >500 mg/L, and who were randomly assigned to three months of weekly apheresis or sham intervention, there was a significant improvement in the active treatment group in myocardial perfusion reserve, atherosclerotic burden measures, exercise capacity, symptoms, and quality of life [70]. (See "Treatment of drug-resistant hypercholesterolemia", section on 'LDL apheresis'.)

Antisense therapy — An antisense oligonucleotide that reduces Lp(a) by inhibiting the production of apolipoprotein(a) (apo(a)) is in late-stage clinical development. (See "Overview of gene therapy, gene editing, and gene silencing".)

In a dose-ranging study of individuals with ASCVD and Lp(a) levels >60 mg/dL (150 nmol/L), one antisense oligonucleotide therapy (pelacarsen, also known as apo(a)-L_Rx) lowered Lp(a) in a dose-dependent fashion by as much as 80 percent [72]. A separate dose-finding study of patients with ASCVD and similar Lp(a) levels as the prior study showed that high doses of olpasiran 225 mg daily (a small interfering RNA molecule) could lower placebo-adjusted Lp(a) by over 100 percent [73]. The most common adverse events in both studies were injection-site reactions [72,73]. Additional studies are ongoing.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Lipid disorders in adults" and "Society guideline links: Primary prevention of cardiovascular disease" and "Society guideline links: Secondary prevention of cardiovascular disease".)

SUMMARY AND RECOMMENDATIONS

●Background – Lipoprotein(a) (Lp(a)) is likely a causal independent risk factor for atherosclerotic cardiovascular disease (ASCVD) events. (See 'Disease associations' above.)

The risk of ASCVD events increases linearly with Lp(a) concentrations. ASCVD risk increases are clinically relevant as Lp(a) concentrations exceed 30 to 50 mg/dL (125 nmol/L) and especially when they exceed 180 mg/dL (430 nmol/L).

●Screening – There is no broad agreement on when to screen for high Lp(a). Based on the available evidence, we perform targeted screening. Specific examples are presented above. (See 'Screening' above.)

●Approach – No clinical trials have adequately tested the hypothesis that Lp(a) reduction reduces the incidence of future ASCVD events. . Thus, except in very rare cases, we do not target Lp(a) with any therapy known to lower Lp(a). (See 'Initial approach' above.)

•Initial approach – In patients with high Lp(a) (≥50 mg/dL or ≥125 nmol/L), more aggressive low-density lipoprotein cholesterol (LDL-C) targets than those recommended for the general population may be considered. This is based on the knowledge that there may be excess residual cardiovascular disease risk associated with high Lp(a) levels. (See "Management of low density lipoprotein cholesterol (LDL-C) in the secondary prevention of cardiovascular disease".)

•Our initial approach to reducing ASCVD risk in patients with elevated Lp(a) is to reduce LDL-C to its target. We consider the use of PCSK9 monoclonal antibodies as preferential to ezetimibe as second-line lipid-lowering therapy. (See 'Initial approach' above.)

•Patients at LDL-C goal – For patients with elevated Lp(a) who have reached their LDL-C target or who have attempted all recommended therapies to lower LDL-C, we await definitive evidence from ongoing cardiovascular outcomes trials with selective Lp(a)-lowering therapies and do not yet have any broader recommendations on Lp(a) lowering. (See 'Next steps' above.)