1 INTRODUCTION

Oxidative stress and inflammation lead to the oxidative modification of phospholipids (PLs) in cellular membranes and lipoproteins and generation of oxidized phospholipids (OxPLs). OxPLs are present in a variety of diseases including atherosclerosis and exert either proinflammatory or anti-inflammatory effects.1, 2 Lipoprotein(a) [Lp(a)], a lipoprotein particle found in plasma, is also susceptible to oxidative stress, leading to the formation of proinflammatory and proatherogenic OxPLs.3-6 Lp(a) exerts inflammatory, thrombotic and atherogenic properties, representing an independent cardiovascular risk factor.3-5, 7 Among others, it mediates atherogenesis through mechanisms linked to its associated OxPLs, as Lp(a) is their major lipoprotein carrier.6 Elevated Lp(a) plasma concentrations and OxPLs of Lp(a) (OxPL-apoB) predict the presence and progression of coronary, femoral and carotid artery disease.8 Furthermore, they are increased following acute coronary syndrome and percutaneous coronary intervention, as well as predict death, myocardial infarction, stroke and need for revascularization in unselected populations.8 They are also correlated with an amplified risk of aortic valve stenosis and peripheral artery disease.8 Notably, there are currently no available treatments for potent reduction in high Lp(a) or OxPL-apoB. In this narrative review, we present an update on OxPLs, focusing mainly on the OxPLs of Lp(a) and their biological role, relationship with cardiovascular risk and the effect of current and future medical interventions on their levels.

2 METHODS

Relevant studies were identified by searching MEDLINE, EMBASE and CENTRAL databases up to 21 September 2021 using the following terms: oxidized phospholipids, lipoprotein(a), Lp-PLA2, cardiovascular risk, hypolipidaemic treatment, statins, ezetimibe, proprotein convertase subtilisin/kexin 9 inhibitors, lipoprotein apheresis and antisense oligonucleotides. Reference lists from these articles were also scrutinized. Reporting of the study conforms to broad EQUATOR guidelines.9

2.1 OxPLs: What are they and how do they form?

Under conditions of oxidative stress or during inflammation, different biomolecules, including proteins, lipids and nucleic acids, can be oxidatively modified. Phospholipids in cellular membranes or lipoproteins are among such biomolecules. This oxidative modification is observed when there is an imbalance between the reactive oxygen species (ROS) generation and the protective capacity of cellular antioxidant mechanisms.1, 2 OxPLs generally arise from the oxidation of the polyunsaturated fatty acid residues (PUFAs) of phospholipids, which are usually esterified at the sn-2 position of phospholipids.10 PUFAs are highly susceptible to radical-mediated oxidation. Their oxidation is initiated either enzymatically by lipoxygenases or ROS under conditions of oxidative stress and propagates via the classical mechanism of lipid peroxidation chain reaction. This implies that the production of OxPLs cannot be regulated by adjusting the amount or activity of enzymes; hence, there is an uncontrolled generation of OxPLs during oxidative stress.10 The number of oxidation products from each PUFA is likely to be at least 50, and the effects of all these products are unknown.11 The oxidative modification of membrane phospholipids can produce two major categories of OxPLs. OxPLs of the first category are characterized by a shortened oxidized fatty acid at the sn-2 position that contains hydroxyl or carbonyl groups. These OxPLs are generated due to a fragmentation of the oxidized fatty acid. Highly reactive short-chain aldehydes, such as 4-hydroxynonenal or malondialdehyde, are also released. These short-chain aldehydes could potentially modify proteins, lipids and nucleic acids. OxPLs belonging to the second category have more complex structures. They are the result of intramolecular cyclization, rearrangements and further oxidation. They may contain isoprostanes, isothromboxanes or isofurans at the sn-2 position. Examples of both categories have been detected ex vivo in the context of atherosclerosis and characterized as synthetic compounds. It has been revealed that several OxPLs share common effects, but there are also distinct bioactivities for structurally diverse OxPL species.1, 2

The presence of OxPLs has been documented in a variety of diseases, including acute and chronic microbial infections, metabolic disorders such as hepatic steatosis, degenerative diseases, atherosclerosis, calcific aortic valve stenosis and acute lung injury.1, 2 They are ubiquitously formed in many inflammatory tissues and contribute to the pathogenesis of inflammation in general, either by induction or by resolution. The pathophysiological relevance of OxPLs to these conditions appears to be directly linked to their ability to modulate the underlying biological and inflammatory processes.1, 2 During the oxidative modification of phospholipids, different ‘oxidation-specific epitopes’ (OSEs) are formed. OSEs include the phosphorylcholine (PC) moiety revealed on OxPLs, as well as different types of OxPL-protein adducts that are generated during lipid peroxidation. Different OxPLs and OSEs are present on cellular membranes and lipoproteins as ‘danger-associated molecular patterns’ (DAMPs) that perpetuate local inflammation, apoptosis and tissue disintegration.12, 13 It has been suggested that OxPLs can integrate into lipid membranes of cells and lipoproteins; they then can either act as ligands or may cause local membrane disruption.11 They are, also, recognized by a set of pattern-recognition receptors (PRRs) that trigger innate immunity.12, 13 By triggering innate immunity, they elicit signalling pathways that evoke various proinflammatory and plaque-destabilizing processes.12, 13 Several receptors, involved in the initial interaction of OxPLs with cells, have been identified, including CD36, CD14, scavenger receptor class B type 1, prostaglandin E2 receptor 2, vascular endothelial growth factor receptor 2, Toll-like receptor 2/1 and 6 and the platelet-activating factor receptor. An interaction with Toll-like receptor 4 has also been suggested.11 OxPLs can interact with soluble PRRs as well, such as C-reactive protein, complement components and germ line–encoded natural antibodies.1, 2 In turn, a wide variety of responses is generated.11 The direct recognition of oxidized fatty acids of OxPLs from PRRs and the recognition of the newly formed OSEs (previously described) mediate the proinflammatory activity of OxPLs. As a result, they trigger the secretion of diverse inflammatory mediators, including vasoactive substances, proinflammatory cytokines, chemokines and phospholipid-derived lipid mediators, which orchestrate the resulting innate immune response. However, certain OxPLs, formed by radical-mediated lipid peroxidation, and lipid mediators, generated by enzymatic reactions, have been shown to exert anti-inflammatory effects by negatively influencing the Toll-like receptor activation in response to microbial ligands in vitro and in vivo and interfering with proinflammatory nuclear factor kappa-light-chain-enhancer of activated B cell (NFkB) signalling. These inhibitory effects of OxPLs on Toll-like receptor activation are not restricted to a single Toll-like receptor (Figures 1 and 2).1, 2 Another factor determining whether OxPLs exert proinflammatory or anti-inflammatory effects may be their local concentration. It seems that individual OxPL species that are detectable in low concentrations in blood exert anti-inflammatory effects, while significantly higher concentrations are required to elicit proinflammatory cytokine secretion from endothelial cells in vitro and in vivo.1, 2 Moreover, recent studies suggest that the anti-inflammatory bioactivity of OxPLs is predominantly mediated by OxPLs that have a cyclopentenone as their defining structural component. Indeed, they were found to promote the resolution of inflammation in vivo. These OxPLs exhibit high functional and structural similarity to the endogenous proresolving prostaglandins and mimic their bioactivity.2

Formation of OxPLs. OxPLs generally arise from the oxidation of the PUFAs of phospholipids, which are usually esterified at the sn-2 position of phospholipids. PUFAs are highly susceptible to radical-mediated oxidation. Their oxidation is initiated either enzymatically by lipoxygenases or ROS under conditions of oxidative stress and propagates via the classical mechanism of lipid peroxidation chain reaction. The number of oxidation products from each PUFA is likely to be at least 50, and effects of all these products are unknown. During the oxidative modification of phospholipids, different OSEs are formed, including the PC moiety revealed on OxPLs, as well as different types of OxPL-protein adducts that are generated during lipid peroxidation. Different OxPLs and OSEs are present on cellular membranes and lipoproteins as DAMPs, which are recognized by a set of PRRs, and a wide variety of proinflammatory and anti-inflammatory effects is generated. PUFAs, polyunsaturated fatty acid residues; ROS, reactive oxygen species; OxPL-apoB, OxPLs of Lp(a); OSEs, oxidation-specific epitopes; PC, phosphorylcholine; OxPL, oxidized phospholipid; DAMPs, danger-associated molecular patterns; PRRs, pattern-recognition receptors

Bioactivity of OxPLs. Under conditions of oxidative stress or during inflammation, phospholipids in cellular membranes or lipoproteins can be oxidatively modified. This oxidative modification is observed when there is an imbalance between the ROS generation and the protective capacity of cellular antioxidant mechanisms. During the oxidative modification of phospholipids, different OSEs are formed. OSEs include the phosphorylcholine moiety revealed on OxPLs, as well as different types of OxPL-protein adducts that are generated during lipid peroxidation. Different OxPLs and OSEs are present on cellular membranes and lipoproteins as DAMPs. OxPLs are recognized by a set of PRRs, and a wide variety of responses is generated. The direct recognition of oxidized fatty acids of OxPLs from PRRs and the recognition of the newly formed OSEs mediate the proinflammatory activity of OxPLs. As a result, they trigger the secretion of diverse inflammatory mediators, including vasoactive substances, proinflammatory cytokines, chemokines and phospholipid-derived lipid mediators, which orchestrate the resulting innate immune response. However, certain OxPLs, formed by radical-mediated lipid peroxidation, and lipid mediators, generated by enzymatic reactions, have been shown to exert anti-inflammatory effects by negatively influencing the TLR activation in response to microbial ligands in vitro and in vivo and interfering with proinflammatory NFkB signalling. These inhibitory effects of OxPLs on TLR activation are not restricted to a single TLR. OxPL, oxidized phospholipid; ROS, reactive oxygen species; PRRs, pattern-recognition receptors; OSEs, oxidation-specific epitopes; TLR, Toll-like receptors; NFkB, nuclear factor kappa-light-chain-enhancer of activated B cells

2.2 Preferential association of OxPLs with Lp(a) in human plasma

Most of circulating OxPLs are associated with Lp(a) with only small amounts found on other apolipoprotein B-100 (apoB-100) particles [eg low-density lipoprotein (LDL)] and high-density lipoprotein (HDL).6 OxPLs are also associated with plasminogen, even favouring its fibrinolytic action.14, 15

The levels of OxPLs are measured in a chemiluminescent immunoassay using the murine monoclonal antibody MB47, which binds human apoB-100 in plasma. Then, the natural murine IgM monoclonal antibody, E06, that recognizes the PC head group on oxidized but not on native phospholipids is added to determine the amount of OxPLs present on apoB-100.8, 16-18 Therefore, OxPL plasma levels, mainly associated with Lp(a), are expressed as OxPL-apoB.

Lp(a) consists of an LDL-like particle, in which the apolipoprotein B-100 is linked by a single disulphide bridge to a unique plasminogen-like glycoprotein, known as apolipoprotein(a) [apo(a)].3-5 Approximately 90% of circulating Lp(a) levels are inherited and strongly determined by the LPA gene.19 The LPA gene and therefore apo(a) contain 10 copies of kringle IV (KIV), KV and a catalytically inactive protease domain. KIV contains 1 copy of KIV1 and KIV3-10, but variable copies of KIV2, ranging from 1 to >40 on each allele.20, 21 There is an inverse relationship between the number of KIV2 repeats of apo(a) and the levels of Lp(a) in plasma.22 Due to the high structural heterogenicity of Lp(a), it remains a challenge to standardize a biochemical quantification for clinical practice. Since now, none of the available commercial assays for Lp(a) quantification is completely isoform-insensitive. The only method (gold standard) that is independent of apo(a) size polymorphism uses a monoclonal antibody that does not interact with epitopes in KIV2, but does interact with an epitope in KIV9. Antibodies binding to epitopes in the KIV2 region will generate multiple antibody-antigen complexes dependent on the size of apo(a). However, this ELISA method may be affected by protein conformational diversity.23 The use of LC-MS/MS analysis seems to be independent of apo(a) size polymorphism, as well, by selecting specific quantification peptides not present in the KIV2 region. Indeed, the results obtained by this method are in excellent agreement with those obtained by the current gold standard ELISA.23 Despite all the attempts to harmonize commercial assays, the conversion between mg/dl of Lp(a) mass and nmol/L of apo(a) is not accurate and should be avoided in clinical practice.23 In general, traditional thresholds for elevated Lp(a) are suggested to be >30–50 mg/dL (>75–125 nmol/L), although differences among guidelines exist.3, 24, 25 Different studies support the role of Lp(a) as a risk factor for coronary heart disease, ischaemic stroke, peripheral artery disease, heart failure, venous thromboembolism, calcific aortic valve stenosis and retinopathy in diabetic patients, and as an independent cardiovascular risk factor as well.3, 7, 26-28 Also, Lp(a) levels seem to be a marker of restenosis after percutaneous transluminal coronary angioplasty, saphenous vein bypass graft atherosclerosis and accelerated coronary atherosclerosis of cardiac transplantation.29 Moreover, when coexisting with thrombotic factors, such as fibrinogen, d-dimer, plasmin-antiplasmin complex and factor VIII, elevated Lp(a) levels are associated with even higher risk of atherosclerotic cardiovascular events.15 In the era of COVID-19 pandemic, elevated Lp(a) levels do not seem to increase susceptibility to SARS-CoV-2 infections nor to thromboembolic events. However, SARS-CoV-2 infections may enforce the association between elevated Lp(a) levels and ischaemic heart disease.30

OxPLs can be formed on oxLDL, apoptotic cells, oxidized cell membranes and sites of inflammation.6, 16 Afterwards, OxPLs are released into the circulation from these sources and are preferentially transferred to Lp(a).6, 16 Up to 90% of all OxPLs found in human lipoproteins are carried on Lp(a), which is not necessarily oxidized. Thus, OxPLs may impart additional and potent proinflammatory properties to Lp(a) and play a key role in Lp(a) functionality.6, 12 An interesting observation has been made in patients undergoing percutaneous coronary intervention.31 Before percutaneous coronary intervention, most OxPLs in plasma were associated with Lp(a). Immediately after percutaneous coronary intervention, an acute increase was noted in both plasma OxPL-apoB and Lp(a). However, only 50% of OxPLs were associated with Lp(a), whereas the remaining were present on non-Lp(a) apoB-100 particles. However, by 6 h, >90% of OxPLs were again present on Lp(a), suggesting their transfer and selective binding on Lp(a).31 This observation strongly supports a physiological function of Lp(a) in preferentially binding OxPLs compared with other apoB-100-containing particles and suggests a transfer of OxPLs to Lp(a).

A study, though, suggested that the linkage of OxPLs to apo(a) is a cellular event and that this linkage probably occurs in the hepatocyte, at the time of apo(a) synthesis/secretion, particularly if there is enhanced hepatocyte oxidative stress.32 Indeed, Lp(a) is susceptible to oxidative modifications as well, leading to the formation of proinflammatory and proatherogenic OxPLs, oxysterols and oxidized lipid-protein adducts.12, 13 Moreover, in vitro transfer studies demonstrate that OxPLs from oxLDL are preferentially transferred to Lp(a) rather than LDL in a time-/temperature-dependent fashion.6 The pathways and molecules associated with binding of OxPLs to Lp(a) are not fully determined but may involve covalent binding to apo(a) and distribution in the lipid phase of the LDL of Lp(a).16 Of OxPLs, 40–70% are actually present in the lipid phase of Lp(a), whereas the remaining are on apo(a).33 OxPLs are thought to bind to apo(a) by forming a covalent bond with the KIV10 of the apo(a) fragment of Lp(a).16 KV of apo(a) has, also, the ability to covalently bind OxPLs.34 In general, candidate amino acids for OxPL binding are cysteines, lysines and histidines. However, all 6 cysteines are occupied in disulphide bonds and there are no lysines in KIV10 of human apo(a). Therefore, it appears that the 3 histidines present in KIV10 are likely the sites in KIV10 that bind OxPLs.12 As for the KV, OxPLs can be directly bound to apo(a) by forming covalent bonds with the active lysines of this domain.34 A role in the association of OxPLs with apo(a) may be also played by β-2 glycoprotein I, which can bind to apo(a),35 as well as to anionic phospholipids and OxPLs.36, 37Other studies have also shown strong associations between OxPL and Lp(a).31, 34, 38-41 OxPL-apoB levels correlate with Lp(a) concentration broadly, but this association is strongest in patients with elevated Lp(a) levels and concomitantly small apo(a) isoforms or with the presence of LPA single nucleotide polymorphisms (SNPs) rs3798220 and rs10455872, which are also associated with elevated Lp(a) levels.42, 43 The stronger association of OxPLs with small Lp(a) particles may at least partially explain their enhanced atherogenicity and association with higher cardiovascular disease risk as compared to the large ones.10 As Lp(a) is a lipoprotein carrier of OxPLs, it is likely that the OxPL-apoB levels are also genetically determined to some extent. In this context, studies in monozygotic and dizygotic twins examined the strength of the heritability of this relationship. It was demonstrated that OxPL-apoB levels and other biomarkers of oxidized lipoproteins are genetically determined in most subjects in a manner parallel to Lp(a), highly reflecting apo(a) isoform size.43 In a recent study, OxPL-apoB levels were positively associated with allele-specific Lp(a) levels carried on small apo(a) isoforms in both Black and White participants. A similar finding was seen for large apo(a) isoforms only in the Black group, while a significant heritability for OxPL-apoB levels was noted only in Black participants.41

2.3 Lp-PLA2: association between Lp(a) and OxPLs of Lp(a)

Once OxPLs are bound on Lp(a), they become accessible to endogenous phospholipase A2 (PLA2).16 PLA2 is present in soluble form (sPLA2) and in lipoprotein-associated form (Lp-PLA2). Lp-PLA2 and sPLA2 catalyse the degradation of OxPLs at the sn-2 position into lysophosphatidylcholine and either oxidized free fatty acids (in the case of free OxPLs) or a remnant of the oxidized fatty acyl chain covalently linked to a protein side chain (in the case of OxPLs covalently linked to a protein).10 Like OxPLs, both products of Lp-PLA2 activity manifest proinflammatory and proatherogenic effects. Circulating Lp-PLA2 is mostly associated with LDL, while a smaller amount, usually less than 5%, is associated with HDL.44, 45 Lp-PLA2 is also bound to Lp(a).44, 45 In fact, Lp(a) contains 1.5- to 2.0-fold higher mass and 7.0-fold greater Lp-PLA2 activity compared with LDL, when assayed at equimolar protein concentrations.46, 47 As Lp(a) might be a scavenger of OxPLs, OxPLs are sequestered on Lp(a) and subjected to degradation by Lp-PLA2.10 Apo(a) size may influence the association of Lp-PLA2 with Lp(a), with the small isoforms being less active compared with the large ones.10 Apo(a) does not bind the enzyme itself, though. The major role in the attachment of Lp-PLA2 on Lp(a) is played by its apoB-100 moiety (Figure 3). The low catalytic efficiency of the Lp-PLA2 associated with small apo(a) isoforms could be one of the factors that favour the sequestration of plasma OxPLs on these isoforms, and consequently the strong correlation between small apo(a) isoforms and high OxPL-apoB levels in plasma.10 Thus, when present at low plasma concentrations, Lp(a) may actually be anti-inflammatory and antiatherogenic via its ability to bind OxPLs in the circulation, which are then degraded by Lp(a)-Lp-PLA2.19 Under these conditions, the proinflammatory products, deriving from the degradation of OxPLs, could be transferred in the circulation from Lp(a) to albumin. Albumin represents the major carrier and inhibitor of lysophosphatidylcholine in plasma.10 On the contrary, this effect may become harmful, when Lp(a) exceeds its normal concentration.19 However, in the artery wall, Lp-PLA2 can release lysophosphatidylcholine from the accumulated Lp(a). This way, it may potentiate the proinflammatory effects of OxPLs.10

Structure of lipoprotein(a). Lipoprotein(a) consists of a low-density lipoprotein-like particle, in which the apolipoprotein B-100 is linked by a single disulphide bridge to a glycoprotein, known as apolipoprotein(a). Apolipoprotein(a) contains 10 copies of kringle IV (KIV), KV and a catalytically inactive protease domain. KIV contains 1 copy of KIV1 and KIV3-10, but variable copies of KIV2, ranging from 1 to >40 on each allele. Up to 90% of all oxidized phospholipids found in human lipoproteins are carried on lipoprotein(a) and subjected to degradation by lipoprotein-associated phospholipase A2

Tsironis et al showed that patients with coronary artery disease have a lower catalytic efficiency of Lp(a)-Lp-PLA2, compared with patients without coronary artery disease.47 Interestingly, the removal of apo(a) from the Lp(a) particle resulted in a significant increase in the Lp-PLA2 activity.47 This suggests that the apo(a) moiety diminishes the enzyme activity expressed by Lp(a) in coronary artery disease patients. Again, the lower catalytic efficiency of the enzyme may result in higher OxPLs levels on Lp(a).47

Of note, even though many studies showed that the total plasma Lp-PLA2 (primarily LDL-associated Lp-PLA2) is an independent risk factor for cardiovascular disease, the results of the STABILITY study were negative.48-50 The STABILITY study was a phase III clinical trial aiming to determine whether the inhibition of plasma Lp-PLA2 with the specific inhibitor darapladib has any clinical benefit in the secondary prevention of cardiovascular disease. Inhibition of Lp-PLA2 with darapladib was not associated with improvement in the primary endpoint of cardiovascular death, myocardial infarction or stroke at a median of 3.7 years in patients with established cardiovascular disease as compared to placebo. However, the need for coronary revascularization was significantly lower.50 The inhibitory effect of darapladib on Lp-PLA2 has been studied in total plasma and in LDL. Thus, whether and to what extent darapladib inhibits Lp(a)-Lp-PLA2 is currently unknown.51 So, one hypothesis to interpret the failure of the STABILITY study could be the presence of elevated Lp(a)-Lp-PLA2 levels.

2.4 Lp(a)-Associated OxPLs: data on their relationship with cardiovascular risk

OxPLs accumulate in atherosclerotic lesions and play an important role in atherosclerosis. They induce proinflammatory signalling in endothelial cells, smooth muscle cells, monocytes, macrophages, dendritic cells and platelets, all present in the vessel wall.11, 52 Moreover, OxPLs mediate plaque-destabilizing processes, as they are present in higher quantities (70-fold) in plaque than in plasma.52, 53 They are capable of stimulating proinflammatory genes leading to vascular inflammation and monocyte inflammatory responses in humans and are proapoptotic in high concentrations.11, 16, 52, 54

The Dallas Heart Study, a unique epidemiological survey, found that an association of OxPLs especially with high Lp(a) levels and small apo(a) isoforms may determine cardiovascular risk.40 In this study, OxPL-apoB levels were measured in 3481 middle-aged, asymptomatic subjects (1831 Black, 1047 White and 603 Hispanic) and correlated with age, gender, cardiovascular risk factors, and Lp(a) and apo(a) isoforms. Black participants had higher Lp(a) levels and therefore higher OxPL-apoB levels than White participants and Hispanics. OxPL-apoB levels did not correlate with age, gender or risk factors, such as inflammatory variables, markers of myocardial damage or increased left ventricular wall stress but did correlate strongly with both elevated Lp(a) levels and smaller apo(a) isoforms, irrespective of race, age and gender. This correlation was not observed in the larger apo(a) isoforms, suggesting that this relationship is related to underlying genetic differences in apo(a) size and/or number rather than race per se.40

In a series of epidemiological and interventional studies, OxPL-apoB levels were elevated in patients with coronary, carotid and femoral artery disease.16, 55, 56 They were also increased following acute coronary syndrome and percutaneous coronary intervention compared with OxPL-apoB levels exactly before the percutaneous coronary intervention, and predict death, myocardial infarction, stroke and need for revascularization in unselected populations.16, 55, 56 Myocardial ischaemia followed by reperfusion causes a burst of ROS, mainly targeting the PC-containing PLs, which compromise the cellular bilayer within the mammalian cells. Oxidation of these PLs results in the formation of OxPLs, both in vivo and ex vivo, which are generated within cardiomyocytes during reperfusion and have detrimental effects on cardiomyocyte viability, not only through necrosis and apoptosis but also through ferroptosis. Ferroptosis is an iron-mediated cell death caused by the accumulation of lipid peroxidation products.57 Inactivation of such OxPLs in vivo with E06 results in a reduction in infarct size.58 Among others, the Bruneck study, a large prospective population-based epidemiological survey of 40- to 79-year-old men and women, demonstrated that OxPL-apoB levels are significant predictors of symptomatic cardiovascular, carotid and femoral atherosclerosis.55 OxPL-apoB levels are positively associated with risk of peripheral artery disease in men and women,59 while Lp(a) is also associated with risk of peripheral artery disease. This observation reinforces the key role of OxPLs in the pathophysiology of atherosclerosis mediated by Lp(a).59 Results from the Bruneck study demonstrate that OxPL-apoB, measured at baseline in an unselected population derived from the general community, predicts cardiovascular events (death, myocardial infarction, stroke, transient ischaemic attack and revascularization) over a 10-year prospective follow-up period. This predictive value is independent of traditional risk factors and further amplified with increasing Lp-PLA2 activity.60 Furthermore, increased OxPL-apoB levels predict a 10-year occurrence of new cardiovascular disease events in previously healthy patients. This is, also, independent of traditional risk factors and Framingham risk score.60, 61 Increased OxPL-apoB levels allow reclassification of a significant number of patients in intermediate Framingham risk category into higher or lower risk categories.62 Similar findings were noted for Lp(a).60 OxPL-apoB and Lp(a) were not independent of each other but were independent of all other measured risk factors.60 In two more epidemiological studies, it was demonstrated that acute increases occur in OxPL-apoB in patients following acute coronary syndrome39 or during uncomplicated percutaneous coronary intervention.31 This suggests the generation and/or release of OxPLs into the circulation from atherosclerotic lesions. Indeed, a prospective study in patients with acute coronary syndrome demonstrated that OxPL-apoB levels rise rapidly (approximately 54%) after an acute myocardial infarction.39 Afterwards, they tend to decrease towards the levels before the acute coronary syndrome over the next 7 months.39 Another follow-up study showed that OxPL-apoB and Lp(a) levels increased rapidly (by 36% and 64%, respectively) after percutaneous coronary intervention and returned to baseline after 6 h.31

Emerging data from epidemiological studies show that increased Lp(a) and OxPL-apoB levels are causally associated with increased valve calcification in elderly patients with advanced aortic stenosis, a faster haemodynamic progression of aortic stenosis, and increased risk of aortic valve replacement and death.63, 64 This increased risk may be mediated by the OxPLs of Lp(a) through their proinflammatory and procalcific activity on valvular interstitial cells.63, 64

Another epidemiological study has demonstrated that apolipoprotein C-III (apoC-III), an important component of Lp(a), has strong immunological presence in aortic valve leaflets and colocalizes with Lp(a) and OxPLs. Although apoC-III circulates on Lp(a), there is no correlation between their levels. Thus, only a subset of Lp(a) particles carry apoC-III, and if these patients concomitantly have elevated Lp(a), OxPL-apoB or OxPL-apo(a) levels, they are at high risk of fast aortic stenosis progression, especially from mild-moderate to severe stenosis.65, 66 These findings point to a potential novel therapeutic strategy to reduce the progression of cardiovascular disease, calcific aortic valve stenosis and the need for aortic valve replacement by targeting Lp(a) and its associated OxPLs, as well as the Lp(a)/OxPL/apoC-III pathway.65, 66 To our knowledge, however, there have been no randomized clinical trials demonstrating that lowering Lp(a) reduces the risk for these events. In accordance, genetic studies have shown that SNPs in the LPA gene (rs10455872 and rs3798220) are associated with elevated Lp(a) levels, aortic valve calcification and aortic stenosis.67, 68 All things considered, the selective association of Lp(a) with atherothrombotic vascular disease at specific sites may be, at least in part, mediated by the OxPLs of Lp(a).

2.5 Effect of hypolipidaemic treatment on Lp(a) and OxPL-apoB

There are neither known nonpharmacologic methods, nor any medications specifically approved for lowering Lp(a) levels. The effects of currently used therapeutic agents on circulating levels of Lp(a) are not well understood. Below, we review the established and emerging therapeutic agents that affect Lp(a) levels (Table 1).

TABLE 1. Current and emerging Lp(a)-lowering therapies Lp(a)-lowering therapy Plasma Lp(a) levels Plasma OxPL-apoB levels Lipoprotein apheresis 25–40% reduction 50–75% reduction Statins 6.5–26.0% increase (mean 10.6%) 8–48% increase (mean 23.8%) Low-saturated-fat diets 7–79% increase (mean 19.4%) 12–54% increase (mean 21.3%) Long-term fasting Unchanged Unknown effect Bariatric surgery ~11.3%