Targeting Lipoprotein(a): Can RNA Therapeutics Provide the Next Step in the Prevention of Cardiovascular Disease?

Abstract

Numerous genetic and epidemiologic studies have demonstrated an association between elevated levels of lipoprotein(a) (Lp[a]) and cardiovascular disease. As a result, lowering Lp(a) levels is widely recognized as a promising strategy for reducing the risk of new-onset coronary heart disease, stroke, and heart failure. Lp(a) consists of a low-density lipoprotein-like particle with covalently linked apolipoprotein A (apo[a]) and apolipoprotein B-100, which explains its pro-thrombotic, pro-inflammatory, and pro-atherogenic properties. Lp(a) serum concentrations are genetically determined by the apo(a) isoform, with shorter isoforms having a higher rate of particle synthesis. To date, there are no approved pharmacological therapies that effectively reduce Lp(a) levels. Promising treatment approaches targeting apo(a) expression include RNA-based drugs such as pelacarsen, olpasiran, SLN360, and lepodisiran, which are currently in clinical trials. In this comprehensive review, we provide a detailed overview of RNA-based therapeutic approaches and discuss the recent advances and challenges of RNA therapeutics specifically designed to reduce Lp(a) levels and thus the risk of cardiovascular disease.

Key Summary Points

Elevated lipoprotein(a) (Lp[a]) levels are an independent risk factor for cardiovascular disease.

Currently, there are no standard lipid-lowering therapies that sufficiently reduce Lp(a).

Novel RNA-based drugs that lower serum Lp(a) concentrations, such as pelacarsen, olpasiran, SLN360, and lepodisiran, are in clinical trials.

Long-term risks of Lp(a) depletion are difficult to assess due to the largely unknown physiological role of Lp(a).

Introduction

The number of drugs that interfere with RNA transcription, maturation, and translation is rapidly growing, including antisense oligonucleotides (ASOs), small interfering RNAs (siRNAs), messenger RNAs (mRNAs), and microRNAs (miRNAs) [1]. Compared with established drugs such as recombinant proteins, RNA molecules have the advantage of recognizing a wide range of ligands [2, 3]. This property allows RNA-based therapeutics to reach previously inaccessible targets and enables them to act in a variety of ways, such as inhibiting protein translation or inducing protein coding [4]. However, challenges such as specific and efficient delivery to their intended target site remain to be resolved [5]. Nevertheless, nine ASOs and five siRNAs are currently approved by the Food and Drug Administration (FDA) for treatment of various diseases such as Duchenne muscular dystrophy and hereditary transthyretin-mediated amyloidosis [6–8]. In addition, two mRNAs have been approved for vaccination against the severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) virus [9].

Notably, none of the approved RNA therapeutics directly target cardiovascular disease (CVD), but rather CVD-associated conditions such as familial hypercholesterolemia [10]. Given that CVD remains the leading cause of death worldwide, the development of new therapies and the discovery of novel approaches to disease prevention remain particularly important [11].

Among others, elevated serum levels of lipoprotein(a) (Lp[a]) are an established independent risk factor for CVD [12, 13]. Specifically, individuals with Lp(a) levels above 50 mg/dL have an increased risk of CVD and cardiovascular events including stroke, myocardial infarction, and cardiovascular death [14, 15]. Specifically, concentrations of Lp(a) within the range of 30–50 mg/dL indicate moderate risk, 51–180 mg/dL indicate high risk, and concentrations above 180 mg/dL are associated with very high cardiovascular risk [16]. Lp(a) levels are not influenced by lifestyle changes, but are genetically determined by the LPA gene, which encodes apolipoprotein(a) (apo[a]) [17, 18]. Therefore, modulation of apo(a) expression via LPA transcripts represents a promising target for RNA-based therapies, with numerous publications reporting on their beneficial effects in reducing Lp(a) levels and the associated CVD risk. This review aims to outline the current opportunities and challenges with RNA-based therapeutics targeting Lp(a). The review is based on previously published studies and does not include new human or animal studies conducted by any of the authors.

RNA-Based Therapeutics

Since the discovery of RNA in 1967 and its crucial role in translating genetic information, RNA-based therapeutics have emerged as an important approach to treat human diseases [19]. In general, four different classes of RNA drugs have evolved over time, which include ASOs, siRNAs, mRNAs, and miRNAs (Fig. 1).

Fig. 1.

Overview of gene-silencing mechanisms of RNA-based therapeutics. a ASOs bind to target mRNAs, resulting in RNase H1 cleavage or steric blockade of translation initiation and protein expression. Steric-blocking ASOs can also be used to express a different protein isoform by initiating alternative splicing. b The guide strand of the siRNA enters the RISC, and the activated complex cleaves the target mRNA. c Therapeutics based on mRNA are expressed by the cell's own synthesis machinery and encode a target protein. d MiRNA drugs act by binding and cleaving the target mRNA, thereby inhibiting protein translation, or by competing with endogenous miRNAs, preventing mRNA cleavage by blocking the RISC and allowing protein expression. ASOs antisense oligonucleotides, siRNA small interfering RNA, mRNA messenger RNA, miRNA microRNA, RISC RNA-induced silencing complex. Created with BioRender.com

ASOs

ASOs are 18–30 nucleotides long, single-stranded nucleic acids with a molecular weight of 6–9 kDa [5, 20]. They can be divided into two classes: RNase H-competent and steric-blocking ASOs [5]. RNase H-competent ASOs mostly use the gapmer approach, where a DNA-based gap is framed by RNA-based sequences for target mRNA binding [5]. Upon binding, RNase H1 is recruited, which recognizes the DNA–RNA duplex formed by the aforementioned DNA gap and the mRNA to cleave the bound target [5, 20, 21]. To date, four ASOs using this mechanism have been approved by the FDA: mipomersen, fomivirsen, inotersen, and tofersen [5, 22]. Inotersen and tofersen have also been approved by the European Medicines Agency (EMA) [23–25]. Table 1 lists FDA-approved ASOs and phase III RNA therapeutics for CVD. In contrast, steric-blocking ASOs bind their target mRNA without inducing degradation due to a lack of RNase H1 competence [5, 21]. Specifically, they work via two different mechanisms. First, they can promote and mediate alternative splicing by interacting with pre-mRNAs and thus interfering with the splicing complex [5, 20]. This can lead to an increase in protein expression rather than a silencing effect [20]. Due to the role of alternative splicing in the diversity of the proteome, ASOs could also be used to promote the switch to a different protein isoform and prevent the expression of pathological proteins [5]. There are currently five FDA-approved ASOs that use the splice-switching approach: eteplirsen, golodirsen, casimersen, viltolarsen, and nusinersen [5, 8, 22]. In addition, viltolarsen and nusinersen have been approved by the EMA [26, 27]. Second, ASOs can bind to the codon responsible for translation initiation, thereby inhibiting translation initiation and attenuating target protein expression [5, 20].

Table 1.

FDA-approved or phase III RNA therapeutics for CVD

RNA class	Drug	Indication	Target gene	Side effects	Status	References
ASO	Inotersen	Hereditary ATTR Amyloidosis (hATTR)	Hepatic transthyretin	Thrombocytopenia, glomerulonephritis	FDA-approved; EMA-approved	[5, 23, 87–90]
	Olezarsen	Hyperlipoproteinemia type I (III); hypertriglyceridemia (III)	ApoC-III	Mild erythema at the injection site	Phase III	[86, 87, 91]
	Eplontersen	Amyloidosis; transthyretin-related hereditary amyloidosis; transthyretin-mediated amyloid cardiomyopathy	Transthyretin	Diarrhea, urinary tract infections, vitamin A deficiency	Phase III	[86, 87, 92]
	Volanesorsen	Hyperlipoproteinemia type I; hypertriglyceridemia; lipodystrophy (III)	ApoC-III	Injection site reactions, upper respiratory tract infection, thrombocytopenia	Phase III	[86, 87, 93]
	Mipomersen	Hyperlipoproteinemia type IIa; atherosclerosis	ApoB	Injection site reaction, flu-like symptoms, hepatotoxicity/transaminitis	FDA-approved (discontinued)/refused by EMA	[5, 87, 88, 94, 95]
siRNA	Inclisiran	Hypercholesterolemia (approved); atherosclerotic cardiovascular disease (III)	PCSK9	Injection site reactions, musculoskeletal pain, headache, cough, and back pain along with acute nasopharyngitis or hiccups	FDA-approved; EMA-approved; phase III	[30, 36, 86–88, 96]
	Patisiran	Amyloid polyneuropathy (approved); amyloidosis; hATTR with cardiomyopathy	Transthyretin	Upper respiratory tract infections, injection-related reactions	FDA-approved; EMA-approved; phase III	[30, 33, 87, 88, 97]
	Vutrisiran	hATTR amyloid polyneuropathy (approved); hATTR with cardiomyopathy (III)	Transthyretin	Pain in extremities, arthralgia, vitamin A deficiency	FDA-approved; EMA-approved; phase III	[30–32, 87, 88, 98]
	Fitusiran	Hemophilia A; hemophilia B	Anti-thrombin III	Injection site reactions, arthralgia, increase in liver aminotransferase	Phase III	[87, 88, 99]
	ARO APOC3	Dyslipidemias; hypertriglyceridemia (II); hyperlipoproteinemia type I (III)	ApoC-III	Headache, upper respiratory tract infections, injection site reactions,	Phase III; phase II	[87, 100]

ApoC-III apolipoprotein C-III, ApoB apolipoprotein B, PCSK9 proprotein convertase subtilisin/kexin type 9, ATTR amyloid-transthyretin, hATTR hereditary ATTR amyloidosis, ASO antisense oligonucleotides, siRNA small interfering RNA, FDA Food and Drug Administration, EMA European Medicines Agency, CVD cardiovascular disease; modified from Robinson et al. [86], Zhu et al. [88], and Bejar et al. [87]

In summary, ASOs have significant therapeutic potential due to their ability to bind to mRNAs or pre-mRNAs, resulting in target cleavage, alteration of alternative splicing, or the steric inhibition of mRNA translation. This versatility has contributed to the development of several FDA-approved ASO drugs that provide effective treatments for a range of diseases.

SiRNAs

A second category of RNA-based drugs is siRNAs, which serve as effector molecules of RNA interference (RNAi) and have a very distinct and defined structure. They consist of double-stranded RNA of 20–25 base pairs in length, with the addition of two 3′-overhanging nucleotides [28]. An siRNA is composed of a guide strand complementary to the target mRNA sequence and a passenger strand [5]. Once inside the cell, the siRNA is incorporated into the RNA-induced silencing complex (RISC). The passenger strand is then removed and the activated complex cleaves the target mRNA upon binding of the guide strand to the target sequence [28, 29]. Current FDA- and EMA-approved siRNAs for the treatment of various diseases include vutrisiran, patisiran, givosiran, lumasiran, and inclisiran [30–36]. Table 1 lists FDA-approved siRNAs and phase III RNA therapeutics for CVD.

mRNAs

The third class of RNA-based therapeutics consists of mRNAs. These take advantage of the fact that exogenous mRNA can be translated into functional proteins [19]. Therapies using mRNAs can be divided into two classes: (i) exogenous mRNAs, which replace or supplement endogenous proteins, and (ii) vaccines against infectious diseases or cancer, where the mRNA encodes, for example, a viral protein [19]. Compared with DNA-based drugs, mRNAs have higher transfection efficiency and lower toxicity, since they function without cell nucleus entry and can be degraded in regular cell processes, thus eliminating the potential for infections or insertional mutagenesis [37, 38]. The two vaccines developed for SARS-CoV-2, BNT162b2 and mRNA-1273, are currently the only mRNAs approved by the FDA and the EMA [30, 39, 40].

MiRNAs

The fourth category of potential RNA-based drugs is miRNAs. These single-stranded RNAs are first expressed as longer transcripts and subsequently undergo modification and processing, resulting in the formation of a mature miRNA approximately 22 nucleotides in length [41, 42]. This mature miRNA then acts by specifically binding to a complementary sequence of the target mRNA [43]. In contrast to ASOs, which make the target mRNA recognizable for RNase H1, miRNAs recruit the RISC similarly to siRNAs [5, 44]. The main difference between siRNA and miRNA is that miRNAs target multiple mRNAs and thereby regulate their expression, whereas siRNAs target one specific mRNA [44]. In contrast to siRNAs, miRNAs require only partial pairing to recognize their target mRNA, enabling regulation of multiple mRNAs [44–46]. Consequently, miRNAs regulate mRNA silencing through mechanisms such as translational repression, decapping, deadenylation, or exonuclease activity [44]. Although miRNAs represent a promising field of drug development, miRNA drugs have not yet been approved by the FDA or the EMA [47, 48]. Currently, there are two main categories of miRNA drugs in clinical trials: (i) miRNA mimics and (ii) miRNA repressors [47]. MiRNA mimics are double-stranded RNA molecules designed to replicate the structure of mature miRNA duplexes, effectively mimicking endogenous miRNAs duplexes [42]. In contrast, miRNA repressors are RNA molecules that contain the reverse complement of a specific miRNA, allowing them to bind to endogenous miRNAs and consequently suppress their activity [42, 49].

Limitations of RNA-Based Therapeutics and Potential Solutions

Although RNA-based therapeutics hold immense potential for gene-silencing therapy approaches, there are multiple challenges to their application:

• (i) Stability. Short, unmodified oligonucleotides are rapidly degraded due to their phosphate-sugar backbone, which is recognized and cleaved by nucleases [50–52]. To counteract this, various chemical modifications are introduced to increase their resistance to enzymatic degradation [50, 52]. Early modifications include the substitution of an oxygen atom with a sulfur atom in the oligonucleotide backbone, known as phosphorothioate (PS) modifications [50, 52]. Additional changes such as the introduction of modified sugar molecules further improve oligonucleotide properties [29, 53] by preventing cellular defense mechanisms and degradation [50, 54]. For ASOs, siRNAs, and miRNAs, these sugar modifications include the conversion of the 2′-hydroxyl group to 2′-O-methyl (2′-O-Me), 2′-O-methoxyethyl (2′-MOE), and 2′-fluoro (2′-F) groups [54, 55]. In contrast, for mRNAs, modifying the cap and poly(a) structures of mRNA is utilized to improve the stability of mRNA therapeutics [56].
• (ii) Potency. Stability-improving modifications also contribute to potency, as the longer the oligonucleotides remain in their target tissue, the more pronounced their effect becomes [57]. PS modifications, for example, enhance stability but also potency by improving the cellular delivery of ASOs [51]. However, they can also reduce the binding affinity to the target nucleic acid [55, 58]. 2′-Sugar modifications can improve the effects of PS modifications on target binding affinity [55, 59] or even increase RNA binding affinity of ASOs [57]. Compared with ASOs, siRNAs have a certain inherent stability due to their double-stranded structure, but often exhibit reduced activity without appropriate modifications [60, 61]. To enhance the potency of siRNAs, the passenger strand is often heavily modified to prevent it from entering the RISC and to promote loading of the guide strand into the enzyme complex [28]. Since unintended cooperation of the passenger strand instead of the designated guide strand can lead to unwanted off-target effects and a reduction in target-directed potency [62], the 5′-end of the guide strand should have a relatively low pairing energy to ensure its selection as the strand loaded into the RISC [61, 63]. For mRNAs, changing the cap structure increases protein synthesis, while modifying the poly(a) tail enhances translation efficiency [64]. To increase the binding affinity of miRNA repressors similar to ASOs and siRNAs, 2′-MOE and 2′-F modifications are introduced [65]. In contrast, the potency of miRNA mimics is transient and largely limited by low transfection efficiency [66]. To overcome this, similar to siRNAs, nucleotides in the passenger strand are modified to enhance miRNA incorporation into the RISC [42].
• (iii) Toxicity. In general, the PS backbone is the most common cause of toxicity [50, 67], which tends to induce immune responses due to nonspecific protein binding of the oligonucleotide and subsequent complement activation [57, 58, 68]. In addition to the immunostimulatory effects of introduced modifications, the basic molecule itself can trigger the immune system. By binding double-stranded RNA (dsRNA)-activated protein kinase (PKR) or toll-like receptors (TLRs), dsRNA induces interferon responses that result in nonspecific degradation of mRNA and apoptosis [44, 63]. TLRs also recognize siRNAs, therefore triggering defense mechanisms and inducing inflammatory cytokine production [61, 69]. Depending on the cell type, limiting the siRNA sequence to less than 30 nucleotides can partially mitigate these effects [63]. Similar to siRNAs, mRNA drugs can trigger an immune response by TLRs, making cell delivery more challenging than with smaller therapeutic components [19, 37]. Since both innate and adaptive immune responses are required for effective immunization, which must induce a long-term stimulatory effect on effector and memory cell production, the immunostimulatory properties of native mRNA hold potential for adjuvant vaccine activity [56, 70]. To increase the safety of the mRNA by modulating this inherent immunogenicity and to enhance translatability, modifications of nucleotides, the 5′-cap, and the poly(a) tail are introduced [56, 71, 72]. In addition, off-target silencing can lead to an observable and toxic phenotype [73] affecting cell viability [63, 73], with off-target effects observed on numerous genes in siRNA-treated cells [63, 74]. Similar to siRNAs, ASOs induce toxicity by two mechanisms: RNase H1-dependent silencing of or hybridization to off-target nucleotides due to complete or partial complementarity [67, 75]. The functional groups of ASOs allow them to interact with proteins or other small-molecule drugs, potentially leading to targeting of unintended organs [75]. In addition, ASOs can be toxic through their accumulation in cytoplasmic granules and aptamer binding [67]. Similarly, the broad binding characteristics of miRNAs and their ability to regulate a wide range of mRNAs raise concerns that they may bind to yet unknown off-targets [76].
• (iv) Delivery. Emerging as an early transport approach, nanoparticles were used to protect siRNAs against degradation by RNases and to mask the charge of siRNA and to facilitate escape of the siRNA into the cytoplasm [54]. Today, N-acetylgalactosamine (GalNAc) modifications are a prototypical siRNA conjugate for delivery into hepatocytes [54]. This approach takes advantage of the presence of the asialoglycoprotein receptor (ASGPR) on hepatocytes, with approximately 500,000 receptor copies per cell [54, 77]. Upon binding of GalNAc by ASGPRs and internalization into endosomes, the ligand is released and the receptor undergoes rapid recycling [54, 77, 78]. The majority of FDA- and EMA-approved siRNAs utilize this mechanism of glycoprotein uptake to achieve liver-specific delivery [77, 79]. For miRNAs, several delivery approaches have been developed to facilitate cell membrane penetration, including liposomes, polymers, or exosomes [80]. Similarly, lipid nanoparticles, cationic nano-emulsions, or cationic peptides and polymers can be used for mRNA delivery [9].
• (v) Side effects. Common side effects of siRNA treatment include injection site reactions, infections of the upper respiratory tract, and in rare cases, more target organ-specific reactions such as elevated liver enzymes [81]. Similarly, ASO administration is commonly associated with injection site reactions, headache, fever, or nausea [75]. Common side effects of mRNA therapeutics that have been extensively studied in the context of SARS-CoV-2 vaccination include pain at the injection site, fatigue, and headache [82]. More serious complications include thrombosis, stroke, and myocarditis [83]. Most of the side effects are likely caused by the injection itself and the immune responses induced by the drug [75]. However, in some cases, more serious adverse effects such as hepatic or renal toxicity are reported, which may be explained by accumulation of the drug in the target tissue [75]. While mild effects such as headache or flu-like symptoms can be easily treated with nonprescription medications [75], more severe side effects require thorough investigation before approval and clinical use.

In summary, the major challenges in the development of RNA-based therapeutics include ensuring stability, efficacy in terms of silencing and induction of immune responses after administration, and specific delivery to the target site. While mild side effects, such as flu-like symptoms, can be easily managed, more serious side effects, such as liver toxicity, may pose the greatest challenge to clinical translation. However, the introduction of certain modifications and the development of new delivery systems are improving the properties of the oligonucleotide and helping to bring RNA-based drugs a step closer to clinical implementation.

RNA-Based Therapeutics and CVD

As CVD remains the leading cause of death worldwide, with an estimated 19 million deaths in 2020 [84], the potential of RNA-based therapeutics to prevent cardiovascular events such as myocardial infarction and to reduce potential risk factors associated with genetic predisposition has gained traction. CVD risk factors include high blood pressure, smoking, diabetes, and high blood cholesterol levels as well as age, family history, or sex [85]. Novel drugs marketed for the treatment of CVD have been predominantly limited to combinations of already existing and approved agents, including small molecules such as statins or beta-blockers [86]. However, recent advances, particularly in the field of nucleic acid-based therapies, have made it possible to target pathways, including those associated with CVD, for example through miRNA regulation [86]. The development and approval of RNA-based therapeutics for the prevention of CVD is mainly focused on hypertension and heart failure, angiogenesis, amyloidosis, and lipid management [86] (Table 1).

Elevated levels of low-density lipoprotein cholesterol (LDL-C) have been shown to be directly associated with CVD [41, 85]. After a meta-analysis demonstrated a 23% relative risk reduction of major cardiovascular events for every 38.7 mg/dL reduction in LDL-C [101], therapeutic lowering of LDL-C levels became a cornerstone for secondary CVD prevention [102]. While lifestyle changes can reduce high LDL-C levels, a subset of individuals require pharmacological interventions to achieve adequate reduction of lipoprotein concentrations [41]. In such cases, LDL-C levels are typically lowered with statins, which are hydroxymethylglutaryl-CoA reductase inhibitors [103] and the recommended drug for lowering LDL-C levels in patients with atherosclerotic cardiovascular disease (ASCVD) [104]. Nevertheless, certain patient populations such as those with familial hypercholesterolemia require additional or alternative treatment approaches to statins to achieve the predefined target [41, 105]. According to the European Society of Cardiology guidelines, the treatment goal for individuals with familial hypercholesterolemia and no concomitant risk factors is to achieve a reduction in LDL-C of ≥ 50% from baseline, along with LDL-C levels below 70 mg/dL [105, 106]. In the LDL metabolism, proprotein convertase subtilisin/kexin type 9 (PCKS9) plays a critical role by accelerating LDL receptor degradation and limiting receptor recycling [107]. In patients with an inadequate or no response to statin treatment, PCSK9 inhibitors offer a viable option [41]. These inhibitors increase the number of LDL receptors that bind LDL-C and subsequently facilitate its degradation [108] and include antibodies such as alirocumab or RNA-based drugs such as inclisiran, an FDA-approved siRNA targeting LDL-C for the treatment of ASCVD and familial hypercholesterolemia [109].

In addition to targeting elevated LDL-C levels, RNA-based drugs have been developed in the context of CVD-associated conditions such as hypertriglyceridemia, which has been associated with increased risk of CVD [110]. Advanced therapy medicinal products (ATMPs) developed and studied for their efficacy in hypertriglyceridemia include olezarsen, volanesorsen, and ARO-APO C3 [111–113]. Another cause of cardiac diseases is cardiac amyloidosis, which is caused by extracellular deposition of incorrectly folded proteins in the heart [114]. ATMPs that have been developed and studied for the treatment of amyloidosis, particularly hereditary transthyretin (hATTR) amyloidosis-associated polyneuropathy, include eplontersen, inotersen, patisiran, and vutrisiran [92, 115–117]. An additional CVD-related condition that has been the focus of RNA-based drug development is hemophilia, where CVD diagnoses and CVD-associated deaths in patients with hemophilia have increased in recent years [118, 119]. Fitusiran has been studied for its preventive effect on the rate of untreated bleeding in patients with hemophilia [120].

In summary, while LDL-C remains the primary target for CVD prevention, some individuals require additional treatment beyond standard statin therapy to effectively reduce CVD risk, such as targeting PCKS9, which has become a stepping stone in the development and clinical application of RNA-based drugs.

Lipoprotein(a) as an Independent Risk Factor for CVD

Despite receiving therapeutic interventions for lipid management, some patients may continue to experience cardiovascular events, suggesting the presence of additional risk factors beyond elevated LDL-C levels [121]. Additional risk factors include elevated levels of Lp(a), an LDL-like particle that contributes significantly to LDL-C levels [122]. Specifically, elevated Lp(a) serum concentrations have been described as a contributing factor to coronary artery disease and atherosclerosis-related disorders affecting overall life expectancy [12–14, 123–126]. Intriguingly, elevated Lp(a) levels are not uncommon, affecting an estimated 1.4 billion people, or approximately 20–30% of the general population [125, 126]. Serum concentrations are not influenced by lifestyle, but are primarily determined by genetic factors [12, 14]—for example, the dietary pattern recommended for blood lipid control had only a slight effect on Lp(a) levels [127].

Individuals with elevated Lp(a) levels have a 31% higher risk of CVD and are 41.5% more likely to have an ASCVD event [123]. Lp(a) values above 180 mg/dL are even associated with a CVD risk comparable to that observed in patients with familial hypercholesterolemia [12, 14, 126]. A randomized Mendelian study showed no difference between men and women with elevated Lp(a) levels in terms of associated CVD risk [124].

Structure, Genetics, and Function of Lp(a)

The Lp(a) particle, primarily synthesized in the liver, consists of covalently linked apolipoprotein A (apo[a]) and apolipoprotein B-100 (apo[b]) attached to an LDL-like particle [123, 125, 126] (Fig. 2). Almost 90% of the Lp(a) serum concentration can be explained by the LPA gene [12]. This gene encodes for apo(a), whereas apo(a) length is determined by the repetition of specific sequences called kringle IV factors (KIV) [12]. Individuals with fewer KIV repeats express a smaller apo(a) isoform and vice versa [12]. In addition to apo(a) isoform size, studies indicate other genetic variants that regulate Lp(a) serum levels [12], such as single-nucleotide polymorphisms rs10455872 or rs3798220 [128]. A comprehensive analysis of three placebo groups from the IONIS-APO(a)_Rx and IONIS-APO(a)-L_Rx clinical trials revealed significant changes of up to −13.1 to 21.6% in Lp(a) levels within one individual without a significant change in mean absolute Lp(a) concentration in any group. This indicates a certain intra-individual variability and implies the existence of factors other than genetic influences [129]. Fluctuations in Lp(a) concentrations can also be attributed to metabolic and structural factors related to very low-density lipoprotein triglycerides, factors associated with thrombus formation, and indicators of acute-phase responses as reflected by changes in the erythrocyte sedimentation rate [130].

Fig. 2.

Structure of LDL-C and Lp(a) particle. a LDL-C consists of a lipid particle, associated with apoB-100. b Lp(a) consists of an LDL-like particle with apo(a). The length of the apo(a) isoform depends on the number of kringle IV (KIV) repeats. The shorter the apo(a) isoform, the higher the particle number. LDL-C low-density lipoprotein cholesterol, Lp(a) lipoprotein(a), Apo(a) apolipoprotein(a), ApoB-100 apolipoprotein B-100. Created with BioRender.com

In contrast to the well-investigated pathological effects, the physiological functions of Lp(a) are not well understood [12, 131]. Studies suggest a potential role in wound healing. Immunohistochemical analysis has revealed positive staining for apo(a)/apo(b) in wounds at various stages of the healing process, suggesting the involvement of Lp(a) in mediating wound healing [131, 132]. In addition, a proteomics study has identified a correlation between Lp(a) and many proteins involved in wound healing [131, 133].

Pathological Mechanism

The underlying pathological mechanisms of Lp(a) are well understood, supported by several studies demonstrating its pro-atherogenic and pro-thrombotic nature [134] (Fig. 3).

Fig. 3.

Pathological mechanism of Lp(a). Lp(a) has pro-thrombotic, pro-atherogenic, and pro-inflammatory properties. Lp(a) interferes with plasminogen activation and fibrin degradation by promoting prothrombin activation. OxPL on the particle surface makes Lp(a) pro-inflammatory and induces monocyte activation and transmigration. Lp(a) is associated with atherosclerosis by promoting foam cell formation, necrotic core formation, endothelial cell binding, and smooth muscle cell proliferation. As a result of plaque formation and rupture, individuals with elevated serum concentrations may experience myocardial infarction and ischemic stroke. Turbulent blood flow and calcification can lead to atherosclerotic stenosis and aortic valve stenosis [163, 206, 207]; SMC smooth muscle cell, EC endothelial cell, Lp(a) lipoprotein(a), Apo(a) apolipoprotein(a), ApoB-100 apolipoprotein B-100, OxPL oxidized phospholipids. Created with BioRender.com

A key factor is the structural resemblance of apo(a) to plasminogen, a pro-enzyme of the fibrinolytic system [125, 135], which is converted to its active protease plasmin, and plays a critical role in the degradation of fibrin and subsequent clots [136, 137]. Although the loop structures of Lp(a) allow binding of plasminogen activators, these regulators are unable to activate Lp(a), unlike plasminogen [138]. Consequently, Lp(a) has the potential to disrupt fibrinolysis and mediate intravascular thrombosis [135] by interfering with conversion of plasminogen to plasmin [139, 140] and by competing with plasminogen for fibrin binding, further reducing fibrinolysis [140]. These effects are more pronounced for smaller apo(a) isoforms due to their increased fibrin binding affinity [140].

Lp(a) also promotes prothrombin activation by binding to free or bound tissue factor pathway inhibitor [125, 141]. Upon activation, prothrombin is converted to thrombin, inducing the formation of insoluble fibrin [141], which can serve as a polymeric scaffold and potentially obstruct blood vessels, leading to pathological thrombosis [142]. In addition, Lp(a) impairs the fibrinolytic process by stimulating the expression of plasminogen activator inhibitor-1 (PAI-1), the primary inhibitor of fibrinolysis [140].

Lp(a) also plays a multifaceted role in the development of atherosclerotic plaques. The greater susceptibility of Lp(a) to oxidation than LDL potentially paves the way for Lp(a) uptake by macrophages, which then transform into foam cells and serve as precursors to atherosclerosis [135, 143]. This induces the expression of inflammatory cytokines and promotes the upregulation of endothelial cell adhesion molecules, both of which contribute to the formation and progression of atherosclerotic lesions [140].

Lp(a) is the most important carrier of oxidized phospholipids (OxPL), which serve as ligands for innate immune cell receptors, and, upon binding, initiate a cascade of diverse pro-inflammatory processes [144]. Activated monocytes play a particularly important role in these inflammatory processes that drive atherosclerosis and thus contribute significantly to its development [145]. Studies have shown increased arterial wall inflammation and mononuclear cell recruitment in individuals with elevated Lp(a) levels [144]. Moreover, Lp(a) was able to induce a pro-inflammatory response in monocytes from healthy individuals. However, when OxPLs are inactivated on the surface of Lp(a) particles, the pro-inflammatory effect is significantly attenuated, highlighting the important role of OxPLs in the pathophysiology of Lp(a), particularly through monocyte activation and consequent increase in CVD risk [144, 146].

In summary, based on the high sequence homology between apo(a) and plasminogen, which leads to the promotion of atherosclerosis and thrombosis, as well as the inflammatory properties of OxPL on the particle surface, Lp(a) may represent a functional link between thrombosis and atherosclerosis [12, 125, 143].

Lp(a) in the Context of Lipid-Lowering Approaches

Given the pathological effects of Lp(a) and its association with CVD, the reduction of elevated Lp(a) levels has emerged as a therapeutic target to reduce cardiovascular events [123]. Previously, it was estimated that a reduction in Lp(a) of more than 100 mg/dL was required to achieve a reduction in the associated risk of CVD [147]. However, more recent analyses have suggested a significantly lower value of 65.7 mg/dL to achieve a similar reduction in CVD risk as would be obtained by lowering LDL-C by 38.67 mg/dL [148]. The previous overestimation of the required reduction can be attributed to suboptimal standardization of Lp(a) measurement assays [148]. Apo(a) size polymorphism poses a major challenge to current assays. Other factors that may affect measurement accuracy include the difference in molar particle mass, poor purification properties of pure Lp(a), and mixed aggregates of Lp(a) and LDL, which cannot be fully dissociated [149].

Currently, there are no therapies specifically designed to target elevated Lp(a) serum concentrations [12, 14, 126, 150, 151]. The only method available to significantly reduce Lp(a) concentrations is a procedure known as lipoprotein apheresis [13, 152]. In this process, lipoprotein particles are physically removed from the blood [153], resulting in a significant reduction in both LDL and Lp(a) levels by 60–70% directly after the procedure [12]. However, Lp(a) levels tend to return to pre-therapy levels within approximately 1 week, necessitating weekly treatment to maintain low Lp(a) levels [154]. Lipoprotein apheresis has several additional drawbacks, including high cost, low capacity, poor accessibility, and its chronic nature [13, 152]. In addition, it is semi-invasive, time-consuming, and often associated with a decrease in quality of life and depression [152]. Furthermore, studies of lipoprotein apheresis have been criticized for poor randomization and blinding [12, 155], which may call into question the reported clinical benefit.

While Lp(a) has been recognized as a CVD risk factor, LDL-C–lowering therapies by statins remain the standard approach to reducing atherosclerotic cardiovascular risk in primary and secondary care populations [156]. In contrast to their efficacy in reducing LDL-C levels, studies have suggested that statins actually increase Lp(a) levels, a phenomenon first described in 1989 with lovastatin, which showed a dose-dependent increase in Lp(a) [157]. Meta-analyses have further confirmed this trend, reporting Lp(a) increases from 8.5 to 19.6% with statin treatment, compared with decreases of −0.4 to −2.3% in the placebo group. Therefore, it has been debated whether the increase in Lp(a) levels counteracts the risk reduction achieved by lowering LDL-C with statin therapy [12]. A meta-analysis showed a 47% increased risk of cardiovascular events in statin-treated patients with elevated Lp(a) levels relative to those with Lp(a) levels below 50 mg/dL (hazard ratio [HR] = 1.48, 95% CI 1.23–1.78). In the control group, subjects with Lp(a) levels above 50 mg/dL had an increased risk of 23% (HR = 1.23, 95% CI 1.04–1.45%) [12, 158]. This suggests that the Lp(a)-dependent risk of CVD is even more pronounced when LDL is reduced by statins [12, 158]. However, this study did not use either group as a reference, making direct risk comparisons difficult [12]. Nevertheless, studies by Tsimikas et al. show significant Lp(a) elevation under statin treatment, with variations among patient [156, 159]. Counteracting this increase by combining statins with Lp(a)-lowering therapeutics may hold great potential for maximizing statin-mediated CVD risk reduction.

While PCKS9 inhibitors primarily target LDL-C, they can also reduce Lp(a) levels by 25–30% [125]. The FDA-approved siRNA inclisiran, a PCKS9 inhibitor, showed no significant change but a trend toward Lp(a) reduction in 80% of patients at high risk for CVD due to elevated LDL-C levels [160]. In addition, two antibody-based PCKS9 inhibitors, evolocumab and alirocumab, have been shown to reduce Lp(a) levels [126]. In the FOURIER trial, evolocumab led to a 12 mg/dL reduction in Lp(a) levels, which was associated with a 15% relative risk reduction for CVD [161, 162]. Similarly, the ODYSSEY study showed a 5 mg/dL reduction in Lp(a) with alirocumab treatment, although these results were influenced by patients’ baseline serum levels [163]. Niacin, another PCKS9 inhibitor, showed a 30% reduction in Lp(a), but no CVD benefit [123, 126, 151]. Although PCSK9 inhibitors can lead to modest reductions in Lp(a) levels by 20% on average [160], they have no effect on arterial wall inflammation [146].

In summary, while established LDL-C-lowering statins appear to increase Lp(a) levels and the associated CVD risk in patients with elevated Lp(a) levels, PCSK9 inhibitors may moderately reduce both. Although lipid apheresis significantly reduces serum Lp(a) concentrations, its various drawbacks make its standardized use highly unlikely. Therefore, other approaches to directly target and reduce Lp(a) need to be developed and investigated.

Reduction of Lp(a) by RNA-Based Therapeutics

Given the limitations of existing drugs discussed above, there is an urgent need for innovative approaches for the reduction of Lp(a), and thus the risk of CVD warrants novel options. Due to its tight genetic control, Lp(a) has been a focal point for the development of RNA-based therapeutics. Most RNA-based approaches currently under development follow a general strategy of targeting the LPA gene in hepatocytes, ultimately leading to a reduction in apo(a) production and consequently Lp(a) particles [14, 123–125]. Another approach is to target apo(b), which is present not only in Lp(a), but also in LDL-C and other lipoproteins. Trials with the ASO mipomersen, which targets apo(b), resulted in moderate reductions in Lp(a) levels, but accompanied by hepatic steatosis during follow-up of up to 6 months. Thus, targeting apo(b) is less likely to be adopted for clinical practice [164]. Currently, there are only four RNA therapeutics in clinical trials that directly aim to prevent CVD by lowering Lp(a) levels: pelacarsen, olpasiran, SLN360, and lepodisiran (Table 2).

Table 2.

Overview of RNA-based Lp(a)-lowering therapeutics under clinical development

Drug	Type	Current clinical trial	Administration	Efficacy	Safety
Pelacarsen	ASO	Phase 3	Subcutaneous	Phase 2 trial in adults with established CVD and elevated Lp(a) levels resulted in Lp(a) reduction of up to 80% (20 mg, weekly) [165] Phase 3 trial currently in progress (NCT04023552) [166]	90% (214/239 total) of patients receiving pelacarsen reported adverse events versus 83% (39/47 total) in the placebo group. The most common side effects were injection site reactions [165]
Olpasiran	SiRNA	Phase 3	Subcutaneous	Phase 1 trial in patients with elevated Lp(a) showed a sustained > 90% reduction in Lp(a)level at doses greater than 9 mg/kg [123] Phase 2 trial in individuals with established ASCVD and Lp(a) levels above 150 nmol/L resulted in a maintained reduction of > 95% relative to placebo when administered every 12 weeks [167] Phase 3 trial currently in progress (NCT05581303) [168]	No major adverse events were reported [123]
SLN360	SiRNA	Phase 2	Subcutaneous	Phase 1 trial included treatment of healthy individuals with elevated Lp(a) levels. Results showed a Lp(a) reduction of more than 80% with the highest dose of 600 mg [14] Phase 2 trial currently in progress (NCT05537571) [169]	Generally well tolerated in individuals with elevated liver enzymes, possibly associated with SARS-CoV-2 vaccination [14]
Lepodisiran	siRNA	Phase 2	Subcutaneous	Phase 1 trial included treatment of individuals without CVD and Lp(a) levels ≥ 75 nmol/L. Treatment with various doses led to a Lp(a) reduction of 97% with the highest dose (608 mg) [170] Phase 2 trial currently in progress (NCT05565742) [171] Phase 1 trial in patients with impaired renal function in progress (NCT05841277) [172]	No drug-related serious adverse events. Generally well tolerated, with 2 participants with elevated amino transaminase, 2 participants with elevated aminotransferases, and 3 participants with elevated creatine kinase levels. Levels were transient and returned to baseline [170]

Lp(a) lipoprotein(a), ASO antisense oligonucleotides, siRNA small interfering RNA, CVD cardiovascular disease, ASCVD atherosclerotic cardiovascular disease, NA no data available

Pelacarsen

Pelacarsen [TQJ230], also known as AKCEA-APO[a]L_Rx, is an ASO and the most advanced RNA-based therapeutic designed to efficiently lower Lp(a) levels [173, 174]. Its predecessor, AKCEA-APO[a]R_x, is a second-generation ASO with 2′-O-(2-methoxyethyl) modifications and was initially tested in a phase 1 study [175]. In a single-dose regimen of AKCEA-APO[a]R_x administered to healthy male subjects with elevated Lp(a) levels, no reduction in Lp(a) or other lipid parameters was observed after 30 days [175]. However, with multiple doses of AKCEA-APO[a]R_x, there was a significant and sustained reduction in Lp(a), OxPL, and apo(b) levels when assessed at 84 days [175].

The phase 2 trial consisted of a randomized, double-blinded trial with placebo-controls. The primary objective of this study was to assess the reduction of Lp(a) levels in a cohort of 64 individuals with elevated plasma Lp(a) levels at day 85 or 99, respectively [176]. Cohort A included patients with Lp(a) concentrations between 125 and 437 nmol/L (62% men vs. 38% women); cohort B included patients with higher Lp(a) levels above 438 nmol/L (18% men vs. 82% women). Treatment with AKCEA-APO[a]R_x resulted in a mean reduction in Lp(a) levels of 66.8% in cohort A and 71.6% in cohort B, significantly different from the placebo control group. Similar beneficial effects were found for other parameters, including apo(b) and LDL-C levels [176].

As part of the same study, a phase 1/2a trial was conducted with a GalNAc-modified version of AKCEA-APO[a]R_x, referred to as AKCEA-APO[a]L_Rx, later known as pelacarsen. Fifty-eight healthy individuals were treated with either single or multiple doses of AKCEA-APO[a]L_Rx (single-dose: 71% men vs. 29% women; multi-dose: 47% men vs. 53% women). The single-dose regimen led to an 85.3% reduction at the highest dose of 80 mg. This reduction was sustained up to day 90, with a remaining reduction of 44%. In the multiple-dose cohort, a mean reduction of 92% was observed at day 36 for the 40-mg group. All participants successfully completed the study with no reported adverse events, validating the benefits of GalNAc modifications [176]. Moreover, the GalNAc modification improved the potency of AKCEA-APO[a]L_Rx, surpassing its predecessor by more than 30-fold [176].

A phase 2 study of AKCEA-APO[a]L_Rx recruited 286 individuals (66% men vs. 34% women) with established CVD and Lp(a) concentrations of at least 60 mg/dL [165]. Administration regimens included 20, 40, or 60 mg every 4 weeks, 20 mg every 2 weeks, and 20 mg weekly [165]. The primary endpoint of the study was the percent change in Lp(a) levels observed at 6 months after treatment initiation. All treatment regimens administered resulted in a significant reduction of Lp(a) levels at 6 months. The most significant reduction, reaching up to 80%, was observed when AKCEA-APO[a]L_Rx was administered weekly at a dose of 20 mg [165].

A phase 3 study is currently underway to determine the impact of Lp(a) lowering with pelacarsen on cardiovascular events in patients with established CVD and elevated Lp(a) levels (ClinicalTrials.gov: NCT04023552); the estimated study completion date is May 2025 [166].

In summary, phase 1 and 2 studies have shown promising results, with pelacarsen reducing Lp(a) concentrations by more than 80% after multiple administrations. A phase 3 study is now evaluating the Lp(a)-lowering potential of pelacarsen in CVD risk reduction.

Olpasiran

Another promising RNA-based drug currently in development is olpasiran, previously called AMG 890. Olpasiran is a GalNAc-modified siRNA designed to target the LPA gene and use RNAi to induce degradation of LPA mRNA, ultimately reducing Lp(a) levels [177]. Previous in vivo studies in transgenic mice and cynomolgus monkeys showed a reduction of more than 80% at a dose of 1 mg/kg, which was sustained for 5–8 weeks after administration [123, 177].

The primary outcome of a phase 1 trial of olpasiran was to assess its safety and tolerability. Secondary endpoints included assessment of Lp(a) reduction and pharmacokinetic analysis [123]. The study enrolled 64 individuals (cohorts 1–5: 70% men; cohorts 6–7 58% men) with elevated Lp(a) levels who were treated with one subcutaneous dose of olpasiran ranging from 3 to 225 mg. Olpasiran was generally well tolerated [123, 178]. Even 1 month after a single dose of olpasiran, Lp(a) levels remained below baseline. Treatment resulted in a maximum mean change of −71 to −97% in individuals with Lp(a) levels of 70–199 nM, and −76 to −91% in cohorts 6 (9 mg olpasiran, single dose) and 7 (75 mg olpasiran, single dose), which included participants with Lp(a) levels above 200 nM [123].

Based on these positive outcomes, a double-blind, placebo-controlled, randomized, dose-finding phase 2 study was conducted. The trial enrolled 281 patients diagnosed with established ASCVD and Lp(a) concentrations greater than 150 nmol/L [167, 179]. The primary objective of this study was to evaluate the efficacy of olpasiran treatment over 36 weeks using subcutaneous injections at doses of 10 mg, 75 mg, and 225 mg administered every 12 weeks or 225 mg every 24 weeks [167, 177]. The secondary endpoint was the percentage change in Lp(a) and other lipid parameters, at both 36 and 48 weeks [167, 179]. In the placebo group, Lp(a) concentration increased by 3.6% at 36 weeks (95% confidence interval [CI], −0.1 to 7.3), whereas Lp(a) concentration decreased significantly under olpasiran treatment [167]. Notably, Lp(a) concentrations were reduced in a dose-dependent manner by up to −101% (95% CI, −105.8 to −96.5) in subjects receiving the highest dose of olpasiran every 12 weeks [167]. At 48 weeks, Lp(a) concentrations were reduced by up to −100.9% (95% CI, −106.7 to −95.0) with the 225-mg dose every 12 weeks [167].

A phase 3 study evaluating the impact of olpasiran treatment on cardiovascular events including coronary heart disease death, myocardial infarction, and coronary revascularization in patients with established ASCVD and elevated Lp(a) levels is ongoing (ClinicalTrials.gov: NCT05581303), with an estimated end date of December 2026 [180].

In conclusion, a phase 2 study evaluating the effect of multiple doses of olpasiran in patients with ASCVD and elevated Lp(a) levels showed a strong decrease in Lp(a) concentrations of up to −101% at the highest dose administration. An ongoing phase 3 study is evaluating the impact of olpasiran on cardiovascular events.

SLN360

A more recent addition to the field of Lp(a)-targeting RNA therapeutics is SLN360. This innovative siRNA targets the LPA mRNA and consequently lowers apo(a) expression [14, 181]. It consists of a 19-mer siRNA modified with GalNAc [14, 181, 182].

Previous in vitro toxicology assessment studies have already shown a reduction of LPA expression in primary human hepatocytes without relevant off-target effects [182]. In preclinical studies, subcutaneous injections in rats resulted in specific delivery of SLN360 to the liver and kidneys, a promising sign for targeted therapy. Importantly, SLN360 administration was generally well tolerated in both rats and nonhuman primates. Liver LPA mRNA and serum Lp(a) levels were significantly reduced at 1 and 2 months after injection [181, 182]. Based on these promising results, this siRNA entered the first phase 1 trial. The primary objective of the phase 1 APOLLO study was to evaluate the tolerability of SLN360 after a single dose and to measure Lp(a) levels at a maximum of 150 days after administration [183]. The study enrolled 32 adults over the age of 18 with no history of ASCVD and Lp(a) levels greater than 60 mg/dL. Overall, SLN360 was well tolerated [14]. Although Lp(a) levels gradually increased after nadir, they did not return to baseline levels even 150 days after administration. Notably, median Lp(a) concentrations were reduced by 70% and > 80% after injection of the highest doses of 300 mg and 600 mg, respectively. Furthermore, SLN360 induced a dose-dependent 18% reduction in total cholesterol and a 26% reduction in LDL-C [14].

A multicenter, randomized, double-blind, placebo-controlled phase 2 trial will evaluate the efficacy, safety, and tolerability of SLN360 in 160 patients with elevated Lp(a) levels and at high risk for ASCVD [169]. The primary endpoint is the mean change in Lp(a) at 36 weeks. Secondary endpoints include assessment of Lp(a) levels at 48 and 60 weeks and changes in other lipid parameters such as LDL-C and apoB [169]. The study is expected to be completed in June 2024 [169].

In summary, a phase 1 trial demonstrated that treatment with the siRNA SLN360 reduced Lp(a) levels by more than 80% after single administration of 600 mg in healthy individuals with elevated Lp(a) levels. Based on these results, a phase 2 study is ongoing to evaluate the effect of SLN360 on Lp(a) levels in patients at high risk for ASCVD.

Lepodisiran (LY3819469)

The latest advances in Lp(a)-lowering therapies is the GalNAc conjugated and 2-O-Me, 2′-F, and unmodified Dicer siRNA lepodisiran or LY3819469, which targets the mRNA encoding apo(a) [184]. A single-ascending-dose, placebo-controlled phase 1 study enrolled 48 subjects without cardiovascular disease and with serum lipoprotein(a) concentrations ≥ 75 nmol/L (65% men vs. 35% women) [170]. The primary outcome was the assessment of tolerability and safety, pharmacokinetics, and effects on Lp(a) concentrations after administration of a single dose of lepodisiran [170]. Secondary outcomes included lepodisiran plasma levels 168 days after dosing and changes in serum Lp(a) concentrations after 336 days of follow-up (48 weeks) [170]. The maximum median change in Lp(a) concentration was −97% in the highest dose group (608 mg), which was maintained at day 337 post-injection (−94%) [170]. A placebo-controlled phase 2 study is ongoing to evaluate the efficacy of siRNAs in an estimated 254 participants with Lp(a) levels above 175 nmol/L over 20 months. The current study end date is October 18, 2024 [171]. In April 2023, a new phase 1 study focused on pharmacokinetics, safety, and tolerability was initiated in 28 patients with normal and impaired renal function [172].

Challenges in RNA-Based Therapies for Lp(a) Reduction

Due to its tight genetic control and various insights into its pathological mechanism, Lp(a) has been a target for the development of RNA therapeutics. However, as its physiological function remains unclear, it is uncertain whether the physiological or pathological role of Lp(a) is more significant in the human body. Therefore, achieving > 90% reduction in Lp(a) levels, already possible using ASO and siRNAs, may have unknown consequences, supported by the observation that very low Lp(a) levels are associated with increased risk of type 2 diabetes mellitus [185, 186]. In fact, a case–control study with 143,087 participants and a Mendelian randomized analysis revealed a causal relationship between very low Lp(a) levels (< 1.5 mg/dL) and type 2 diabetes mellitus [186]. In addition, depletion of apolipoprotein A1, a component of high-density lipoprotein particles, is associated with the development of atherosclerosis [187]. This illustrates how the reduction of an essential metabolic and structural component of lipid particles or excessive lowering of serum Lp(a) levels may induce new pathological conditions.

Cholesterol management is an established focus in CVD, for example by statin therapy. Cholesterol plays an important role in the physiology of the central nervous system, with particular emphasis on members of the LDL receptor family [188–190]. Apart from effects of statins on the mevalonate pathway, which is responsible for the majority of cholesterol production, emerging evidence suggests that statins may exert effects in tissues beyond the liver due to the importance of the mevalonate pathway in various cell types [190]. While some studies suggest a beneficial effect of statins on neurodegenerative diseases, a study in female mice showed an increase in plaques and β-amyloid production in the hippocampus. Therefore, low cholesterol levels may be a risk factor for Alzheimer’s disease in women [191, 192]. In addition, LDL-C deficiency has been associated with an increased risk of neurological disease [193]. However, other studies suggest that the responsible gene variants have no causal effects on Alzheimer’s disease risk or other neurological disorders [194]. A systematic review of randomized clinical trials and observational studies found no evidence of adverse cognitive effects with statin use [195]. The PROSPER study reported no difference in cognitive impairment in patients treated with pravastatin versus placebo [196, 197]. Although literature on the effect of low LDL-C levels on neurological disease is inconclusive, depletion of an LDL-like particle may have similar effects on the brain or other organs.

Additional safety concerns arise from the observed effect of immune responses induced by antisense therapy [28], such as the induction of systemic immune responses and flu-like symptoms after administration of an antisense phosphorothioate DNA oligonucleotide directed against the human immunodeficiency virus 1 in the mid-1990s [28, 198]. Since Lp(a) plays a crucial role in CVD initiation and development, Lp(a)-lowering drugs are likely to be used as a preventive measure for CVD. This implies that both patients with established CVD and healthy individuals would receive treatment for elevated Lp(a) levels. Although these concerns have been addressed in healthy donors during phase 1 trials, the potential for long-term complications remains, especially considering that phase 1 trials for pelacarsen were completed in 2016 [199]. Therefore, long-term effects that manifest later cannot yet be estimated.

In light of possible side effects, it is essential to carefully assess the safety and toxicity of newly developed RNA therapies. While phase 1 trials of RNA therapies targeting Lp(a) have generally reported positive results, adverse effects such as elevated liver enzymes have been reported following SLN360 administration [14]. Although the increase in liver enzymes was likely caused by SARS-CoV-2 vaccination, the possibility of elevated liver enzymes as a result of hepatic adverse events could not be definitively ruled out [14]. In phase 2 studies for pelacarsen, 90% of patients reported mild or moderate adverse effects, compared with 83% of patients in the placebo group [165]. Ten percent of patients treated with pelacarsen experienced severe adverse events, resulting in a total of 5% of patients discontinuing the trial due to arthralgia, myalgia, or malaise [165]. The most common adverse event was injection site reaction, reported in 27% of patients treated with pelacarsen [165]. Other common adverse events affecting more than 10% of patients were urinary tract infections, myalgia, and headaches [165]. In the phase 2 study of olpasiran, two patients receiving olpasiran (10 mg every 12 weeks, 225 mg every 24 weeks) experienced a cardiovascular event and underwent percutaneous coronary intervention [167]. Six percent of patients in the placebo group experienced cardiovascular events, including ischemic stroke and unstable angina, and one patient underwent coronary artery bypass grafting [167]. Twelve patients receiving olpasiran 225 mg every 12 weeks experienced injection site reactions compared with six patients in the placebo group [167]. Pain at the injection site led to discontinuation of the study in three cases in which other dermatologic conditions were described previously [167]. This underscores the fact that even when a drug is considered safe and well tolerated, certain adverse events, particularly those related to the injection site, may still occur.

Even after official approval, adverse events may still be found, as in the case of mipomersen, an ASO targeting apo(b) [41, 200]. Studies have shown that mipomersen can achieve a 30% reduction in LDL-C levels when administered weekly at a dose of 200 mg in various patient populations [41, 201, 202]. Therefore, mipomersen received FDA approval in 2013 for the treatment of heterozygous familial hypercholesterolemia [203]. In addition, mipomersen has been shown to significantly reduce Lp(a) levels in individuals with various lipid disorders and cardiovascular risk [204]. However, post-approval reports revealed several adverse effects including injection site reactions, elevations in hepatic alanine aminotransferases, and hepatic steatosis [41, 201]. Liver toxicity was generally reported after 6 months of treatment [41]. Cardiac events such as coronary artery disease and angina pectoris were also observed, with 12 events in the mipomersen group (n = 39) versus only one in the placebo group (n = 19) [41, 201]. These findings prompted the FDA to issue a boxed warning for mipomersen [41, 203], while the EMA had rejected the initial application for approval and a re-evaluation in 2013 [95]. Mipomersen is a compelling example of a drug that initially demonstrated the desired therapeutic effect, but later showed serious side effects, some of which occurred after a considerable time of treatment. This underscores the importance of conducting thorough studies of target protein reduction using siRNA or ASOs in a manner that allows for the detection of all potential adverse effects. However, the propensity for adverse effects can vary significantly between drugs and individuals, making them difficult to predict and estimate in advance.

Finally, it is important to consider the frequency of administration when evaluating the potential for clinical implementation of these drugs. Patients receiving the first approved siRNA drug patisiran are treated every 3 weeks. After pre-medication with antihistamines and corticosteroids to prevent infusion site reactions, the actual drug administration takes approximately 80 min; 60 min of pre-medication and 80 min of treatment results in a single treatment cycle of more than 2 h [205]. The phase 3 APOLLO study of patisiran showed adverse effects in 97% of patients, most of which were mild or moderate in severity [205]. Notably, the patisiran and placebo groups had similar rates of the most commonly reported adverse events [205]. In these cases, treatment with an RNA therapeutic involved multiple, prolonged treatment cycles with potentially serious adverse events for patients, perhaps for life. In particular in the context of Lp(a) reduction, which may serve as a preventive measure rather than a direct treatment for CVD, it is important to carefully weigh the benefits and burdens for the patient. In addition to considering the patient's perspective, particularly the need for inpatient care, RNA-based drug development requires an assessment of whether the potential benefits of the drug outweigh the need for repeated treatment cycles, typically every 3 to 4 weeks. It is important to determine whether the drug is designed for preventive use or primarily for the treatment of individuals with established disease.

Conclusions

RNA-based drugs represent an emerging class of drugs with potential applications in the treatment of CVD. An established risk factor associated with CVD and cardiovascular events is Lp(a) and its serum concentration. With both its pro-atherosclerotic and pro-thrombotic effects, Lp(a) contributes to the development and progression of cardiovascular disease. Based on its tightly regulated genetic control, Lp(a) has become an attractive target for the development of novel RNA-based therapeutics. To date, there are no standard lipid-lowering agents that exclusively reduce Lp(a). However, three promising RNA drugs (pelacarsen, olpasiran, and SLN360) are currently in development, each of which has demonstrated the ability to effectively reduce elevated serum Lp(a) levels. Another candidate, lepodisiran, is currently being investigated for its effect on Lp(a) serum levels. However, the physiological mechanisms of Lp(a) and the long-term consequences of Lp(a) reduction are still largely unknown. Although preliminary clinical data suggest that RNA-based therapeutics may offer a safe and effective option for achieving sustained Lp(a) reduction, several critical questions remain, including how they can be translated into clinical procedures and how effective the treatment is in terms of CVD risk reduction. In addition, the question remains as to whether excessive Lp(a) reduction is beneficial and what impact it would have on other organs and other pathologies, such as neurodegenerative diseases.

Authors’ Contributions

All authors contributed to the conceptualization of this review. The first draft was written by Henriette Thau. Figures and tables were created by Henriette Thau. All versions were reviewed and edited by Sebastian Neuber, Maximilian Y. Emmert and Timo Z. Nazari-Shafti. All authors read and approved the final version of the manuscript.

Funding

No funding or sponsorship was received for this study or publication of this article.

Data Availability

Data sharing is not applicable to this article, as no original datasets were generated or analyzed.

Declarations

Conflict of Interest

Henriette Thau, Sebastian Neuber, Maximilian Y. Emmert and Timo Z. Nazari-Shafti declare no conflict of interest and no commercial or financial relationships.

Ethical Approval

This review is based on previously published studies and does not include new human or animal studies conducted by any of the authors.