Year: 2025
Type: editorial
Abstract: This editorial refers to ‘Elevated lipoprotein(a) is not linked to coronary artery calcification incidence or progression’, by D.-E. Sung et al., https://doi.org/10.1093/eurjpc/zwaf088. Lipoprotein(a) [Lp(a)] is a LDL-like particle connected via a disulfide bond to apolipoprotein(a) [apo(a)]. Due to its structure and ability to carry oxidized phospholipids, Lp(a) confers a unique atherosclerotic cardiovascular disease (ASCVD) risk profile involving atherogenesis, anti-fibrinolysis, and inflammation.1 While sub-clinical atherosclerotic burden may be one mediator for the heightened risk attributable to Lp(a), there has been an inconsistent association between Lp(a) and coronary artery calcium (CAC) in observational cohort studies.2–4 Beyond understanding the mechanisms underlying Lp(a) and sub-clinical atherosclerosis, assessment of their potential independent pathways may be important to guide personalized risk assessment and treatment among individuals without clinical ASCVD. In this issue of the European Journal of Preventive Cardiology, Sung et al.5 evaluated the association between baseline Lp(a) and incident CAC and CAC progression among nearly 42 000 statin naïve young adults in the Kangbuk Samsung Health Study of South Korea. Lipoprotein(a) was evaluated across quintiles (Q1: 2–3 mg/dL, Q2: 3–5 mg/dL, Q3: 6–10 mg/dL, Q4: 10–21 mg/dL, and Q5: 22–337 mg/dL), across clinical thresholds (<30, 30–49, 50–99, and ≥100 mg/dL), and continuously. Over a median 4 years of follow-up, there was a similar crude rate of incident CAC across baseline Lp(a) quintiles and clinical thresholds of Lp(a), which was between 2 and 3 events per 1000 person-years. In multivariable modelling, Lp(a) was not significantly associated with incident CAC or CAC progression across Lp(a) quintiles or clinically relevant thresholds. Sensitivity analyses including those on statin therapy yielded similar results. The study by Sung et al.5 further underlines the complex relationship of Lp(a) with CAC and broader sub-clinical atherosclerosis burden. Strengths of the analysis include measurement of Lp(a) and CAC among nearly 42 000 participants and robust statistical analyses, including assessment of Lp(a) across several different modelling strategies. Furthermore, CAC progression was evaluated in multiple ways among the ∼9600 individuals with baseline prevalent CAC, which provides excellent statistical power. While the focus on younger adults helps contribute data for this demographic group, this may limit broader generalizability to middle-aged and older adults with a higher burden of risk factors that may interact with Lp(a) to increase risk for CAC development. Less than 15% of participants in the current study had hypertension, whereas the prevalence of hypertension is considerably higher in the US and European countries, and prior work suggests that hypertension modifies the association between Lp(a) and ASCVD.6 Additional limitations of the analysis to consider include a single-centre design in South Korea and a study sample consisting of 85% men, which may limit generalizability to other race/ethnicities and women, respectively. Prior meta-analyses including a mix of cross-sectional and prospective studies suggest that there is a positive association between Lp(a) and CAC in primary prevention; however, there has been considerable heterogeneity across all studies with I2 values ranging from 76 to 91%.2–4 Such heterogeneity may be attributable to several factors, including the Lp(a) threshold, assay used, race/ethnicity studied, lipid-lowering therapy, and differences in study follow-up time. While Lp(a)-mediated ASCVD risk is generally similar across race/ethnicity, population mean Lp(a) levels differ (Black ∼75 nmol/L, South Asian ∼30 nmol/L, White ∼25 nmol/L, and Latino ∼15 nmol/L).7 Such heterogeneity may be further complicated by assay differences (mass based, mg/dL, vs. particle based, nmol/L) and lipid-lowering therapies that may affect Lp(a) values, most commonly statins (10–20% potential increase in Lp(a) values). Last, there has been a mix of studies that include cross-sectional and prospective assessment of CAC, which may complicate interpretation of Lp(a) as a potential contributor to the development of calcified plaque. Prior studies that have only included those with baseline CAC = 0 may be affected by healthy participant bias. Beyond heterogeneity in Lp(a) measurement and statistical methods, there are important pathophysiological considerations when evaluating the association between Lp(a) and CAC. Prior work from the Multi-Ethnic Study of Atherosclerosis (MESA) demonstrated that Lp(a) and CAC were independent and additive for ASCVD risk.8 In general, Lp(a) has been more strongly associated with non-calcified plaque and high-risk plaque features (positive remodelling, spotty calcification, low-attenuation, and napkin ring) as opposed to calcified plaque alone. Among ∼1800 asymptomatic adults in the Miami Heart Study, individuals with Lp(a) ≥ 125 nmol/L were four times more likely to have presence of high-risk plaque compared with those with Lp(a) < 125 nmol/L, independent of traditional risk factors. Among those without CAC, those with high Lp(a) were significantly more likely to have any plaque (24.2 vs. 14.2%).9 In another study from MESA, Lp(a) was less strongly associated with CAC when compared with other lipid biomarkers.10 The modest association between Lp(a) and atherosclerosis may also be explained by several non-atherosclerotic mechanisms by which Lp(a) increases ASCVD risk, including potential pro-thrombotic and pro-platelet effects as well as vascular inflammation contributed to by oxidized phospholipids. Oxidized phospholipids carried by the apo(a) moiety may be particularly important contributors to the development of non-calcified, high-risk plaque and acute CHD events.7 Given their potentially independent contributing pathways for ASCVD risk, Lp(a) and CAC may provide complimentary information for risk assessment and defining eligibility for preventive therapies among individuals without clinical ASCVD. Thus, there is a continuum of risk that may be captured with the concurrent measurement of Lp(a) and CAC, which may help guide personalization in statin and non-statin lipid-lowering therapy, LDL-cholesterol goals, and aspirin in those without clinical ASCVD.11 While the majority of participants in ongoing Phase 3 outcome trials evaluating Lp(a)-lowering therapies have a history of clinical ASCVD, prior work suggests that individuals with advanced sub-clinical atherosclerosis (e.g. CAC ≥300) have similar risk.12 Thus, if such Lp(a)-lowering trials are indeed positive and show significant ASCVD risk reduction benefit, measurement of CAC may help identify the highest risk individuals to facilitate earlier risk reduction with Lp(a)-lowering therapies prior to an index event. In summary, the association between Lp(a) and CAC remains complex, and further work will be required to define their degree of association across standardized lab assays, uniform follow-up time, and demographically diverse samples. The totality of evidence suggests an inconsistent association between Lp(a) and CAC (Table 1), suggesting unique risk pathways, which may be able to be harnessed in routine ASCVD risk assessment to guide personalization in preventive lifestyle and pharmacotherapies across the spectrum of sub-clinical atherosclerosis burden. Contributions to inconsistent association between lipoprotein(a) and coronary artery calcium Emphasize reporting of continuous Lp(a) and standardized Lp(a) thresholds Genetic and epidemiological data suggest that ASCVD risk begins as low as ≥30 mg/dL or ≥75 nmol/L While Lp(a)-mediated risk is generally similar across race/ethnicity, population mean Lp(a) levels differ: Black ∼75 nmol/L, South Asian ∼30 nmol/L, White ∼25 nmol/L, Latino ∼20 nmol/L, and East Asian ∼15 nmol/L Prioritize ancestral diversity and sufficiently power analyses to identify potential race/ethnicity differences Mass based (mg/dL) vs. particle based (nmol/L) Transition to universal, standardized particle-based assay for enhanced precision Mix of cross-sectional and prospective studies, prospective studies with different lengths of follow-up Emphasize prospective follow-up Lp(a) may interact with several adjacent risk factors to increase risk of sub-clinical atherosclerosis Consider assessment across specific risk factors (e.g. diabetes, obesity, and inflammation) and similar covariable adjustment across studies Healthy participant bias for studies including only those with CAC = 0 at baseline Risk factors for CAC initiation vs. progression may differ Uniform modelling strategies for assessing CAC progression Lp(a) is more strongly associated with non-calcified plaque and high-risk plaque features as opposed to calcified plaque Consideration of pro-thrombotic and pro-inflammatory effects of oxidized phospholipids carried by apo(a) moiety of Lp(a) No relevant funding sources for the current manuscript. No new data associated with this article.
Language: English
Subfield: Surgery
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4409229096