IntroductionElevated circulating lipoprotein(a) (Lp(a)) is associated with an increased risk of first and recurrent cardiovascular disease (CVD). Circulating levels of Lp(a) are primarily determined via the LPA gene locus [1Human Genetics and the Causal Role of Lipoprotein(a) for Various Diseases.]. Epidemiological [2Nordestgaard B.G. Langsted A. Lipoprotein (a) as a cause of cardiovascular disease: insights from epidemiology, genetics, and biology.] and mendelian randomisation studies strongly support Lp(a) as a direct cause of coronary artery disease [3Genetic Variants Associated with Lp(a) Lipoprotein Level and Coronary Disease.], valve stenosis [4Genetic associations with valvular calcification and aortic stenosis.], heart failure [5Kamstrup P.R. Nordestgaard B.G. Elevated Lipoprotein(a) Levels, LPA Risk Genotypes, and Increased Risk of Heart Failure in the General Population.] and peripheral atherosclerosis [6Phenotypic Characterization of Genetically Lowered Human Lipoprotein(a) Levels.]. Recently, both European and Canadian guidelines for the management of dyslipidaemias have recognised the utility of measuring Lp(a) within the general population, recommending the assessment of Lp(a) for all individuals at least once in their lifetime, underscoring the importance of Lp(a) on cardiovascular health [7Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement From the American Heart Association., 82021 Canadian Cardiovascular Society Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in Adults., 92019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk.]. Conversely, American guidelines recommend that Lp(a) be measured in only select individuals, namely woman with hypercholesterolemia [7Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement From the American Heart Association.]. However, conventional lipid panels do not measure Lp(a), and it is still not routinely assessed in the clinic [7Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement From the American Heart Association.], thus highlighting the need for robust evidence towards its impact in CVD. Due to a global aging population, decreasing risk in acute coronary syndrome (ACS) treatment is a priority for clinicians and researchers alike and finding risk factors which may predict outcome is of the highest importance. Currently, percutaneous coronary intervention (PCI) is commonly used for the treatment of ACS however it carries significantly higher risk rates and worse six-month outcomes in older patients [10Coronary artery stenting in the elderly: short-term outcome and long-term angiographic and clinical follow-up., 11A Novel Multiple Risk Score Model for Prediction of Long-Term Ischemic Risk in Patients With Coronary Artery Disease Undergoing Percutaneous Coronary Intervention: Insights From the I-LOVE-IT 2 Trial.]. Hence, a better understanding of the contribution of Lp(a) to ACS risk in these at-risk patients is needed to identify which patients may benefit most from Lp(a) therapy. This study set out to investigate elevated Lp(a) levels as a potential predictor of ACS over 8 years in the STRAMBO study, a prospective cohort study of 755 community dwelling older men (> age 60 years), who are not on any specific Lp(a) lowering therapy.ResultsPatient Population and Baseline CharacteristicsBaseline characteristics of the study population are presented in Table 1. Based on clinically relevant cut offs [5Kamstrup P.R. Nordestgaard B.G. Elevated Lipoprotein(a) Levels, LPA Risk Genotypes, and Increased Risk of Heart Failure in the General Population.], participants were stratified into three Lp(a) categories; low;50 mg/dL (n=91). The mean age of the study population was 71.5 ±7.3 years, with no difference in age across the Lp(a) groups (p=0.19). Of the 755 participants within the cohort, 42 reported a previous AMI, with a further 70 being diagnosed with IHD prior to recruitment. Body mass index, blood pressure, smoking status and diabetes mellitus status were comparable across the groups. Participants with high Lp(a) were more likely to have of a history of IHD (including undergoing a previous coronary artery bypass graft) (p=0.04) but not AMI. Distribution of Lp(a) within participants ranged from 0.2 mg/dL to 120 mg/dL (Figure 2).

Table 1Baseline characteristics of study population.

Data presented as mean±SD, n (%), or median (IQR). BMI indicates body mass index; BP blood pressure; LDL-C low density lipoprotein-C; IHD Ischemic heart disease; AMI acute myocardial infarction; ACEi angiotensin-converting-enzyme inhibitor; PTH parathyroid hormone; OPG osteoprotegerin; hsCRP high sensitivity c-reactive protein. Ca calcium; P phosphate, CKD-EPI GFR Glomerular filtration rate-chronic kidney disease epidemiology collaboration; BP blood pressure.

Figure 1Study Design and Incidence Outcomes. Of the 1149 male participants enrolled in the STRAMBO study, 755 over the age of 60 were assessed for lipoprotein(a) levels at baseline. Kaplan-Meier cumulative event rate for acute coronary syndrome was assessed in three subgroups.

Figure 2Lipoprotein(a) Frequency Distribution

Absolute Risk of Acute Coronary Syndrome is Associated with Elevated Lp(a) LevelsMedian follow-up time to event was 4.5 years (IQR-2.5-6.5 years), during which 49 individuals had an ACS event. 16% of participants without an event did not complete the 8-year follow up. The incident rate (number of events per 1000 person-years at risk) of ACS was greatest in those with high (>50 mg/dL) levels of Lp(a) (p=0.04), however, there was no difference between low (Figure 3). Individuals with elevated Lp(a) levels had an incidence rate of 8 per 1000 person-years for ACS, whilst the incidence rate for those with high (>50 mg/dL) Lp(a) levels was 17. In participants with Lp(a) greater than 30 mg/dL at baseline, total incidence of ACS was 8.5% of the population. Within the unadjusted absolute incidence rate ratio with low (50 mg/dL) Lp(a) levels.

Figure 3Absolute risk of acute coronary syndrome stratified via circulating lipoprotein(a) levels. Data based on 755 individuals from the STRAMBO cohort. Incidence rates and incidence rate ratios (IRR) with 95% confidence intervals.

Figure 4Kaplan-Meier curve corresponding with low, elevated and high lipoprotein(a) levels for eight-year follow up of acute coronary syndrome. Low Lp(a) corresponds with a value under 30 mg/dL, elevated Lp(a) corresponds with a value between 30 to 50 mg/dL and high Lp(a) is levels over 50 mg/dL.

Within a sub-analysis of participants with prior AMI or IHD diagnosis (n=112), absolute incidence rate of ACS was 1.54 (0.4-5.6) in the high (>50 mg/dL) Lp(a) group, and further increased to 3.3 (0.7-14.9) in the elevated (30-50 mg/dL) Lp(a) group when compared to low (<30 mg/dL) reference group. Of the 112 participants with elevated or high levels of Lp(a), 4.5% of the participants had an ACS event within the 8-year follow-up (Supplemental Figure I). Notably, the elevated Lp(a) subpopulation had a greater incidence rate than the high Lp(a) cohort, which may be in part, due to the high levels of Lp(a) leading to early death and confounding this aspect of analysis, however this not confirmed. Comparatively, in a sub analysis of participants without prior AMI or IHD diagnosis (n=558), absolute incidence ratio was 1.7 (0.4-7.4) in the elevated Lp(a) when compared to low (<30 mg/dL) reference group (p=0.45). Within the high Lp(a) group, absolute risk ratio was significantly higher at 3.2 (1.4-7.6, p=0.004) when compared to control, with the incidence rate was 27 per 1000 person years. Notably, in participants without previous AMI or IHD, Lp(a) shows a positive correlation with absolute risk compared to those with previous disease, in which Lp(a) over 30 mg/dL appears to increase incidence risk. However, neither sub-analyses showed a significant equality of cause-specific cumulative incidence risk (Supplemental Figure 2, 4).

Elevated Lp(a) Levels are Associated with Acute Coronary Syndrome in 8-year Follow-upKaplan-Meier cumulative event rate for ACS after 8-year follow up were 5.9% in the low Lp(a) group (50 mg/dL) Lp(a) group (Figure 3). Within a Cox regression model, hazard ratios were higher in those with high (>50 mg/dL) Lp(a) levels with and without the adjustment for several covariates. Univariate analysis hazard ratios were 1.1 (0.4-2.7) in elevated Lp(a) (30-50 mg/dL) and 2.0 (1.0-4.1) (p=0.04) in the high (>50 mg/dL) Lp(a) group, with low Lp(a) (Figure 5).

Figure 5Cox regression analysis corresponding with low, elevated and high lipoprotein(a) levels for eight year follow up of acute coronary syndrome. Low Lp(a) levels <30 mg/dL, elevated Lp(a) levels <30-50 mg/dL, and high Lp(a) levels >50 mg/dL. Unadjusted hazard ratios (white), adjusted hazard ratios for age (black), adjusted hazard ratios for age and BMI (yellow), adjusted hazard ratios for age, BMI, cholesterol, LDL-C (blue), adjusted hazard ratios for smoking, age, BMI (green), adjusted hazard ratios for smoking, age, BMI cholesterol, LDL-C (red) and adjusted hazard ratios for smoking, age, BMI cholesterol, LDL-C. diabetes mellitus (brown).

In analyses stratified by Lp(a) level, adjusted for age and BMI hazard ratios were 1.1 (0.4-2.8) in the elevated (30-50 mg/dL) Lp(a) group and 2.0 (1.0-4.1) (p=0.04) in the high (>50 mg/dL) Lp(a) group. When adjusting for age, BMI, cholesterol and LDL-C hazard ratios were 1.1 (0.4-2.8) in the elevated (30-50 mg/dL) Lp(a) group and 2.1 (1.0-4.2) (p=0.05) in the high Lp(a) group. Adjusting for smoking, age and BMI hazard ratios were 1.1 (0.4-2.7) in elevated (30-50 mg/dL) Lp(a) group and 2.1 (1.0-4.2) (p=0.06) in high (>50 mg/dL) Lp(a) group.

When adjusting for age, BMI, smoking, cholesterol and LDL-C hazard ratios were 1.0 (0.4-2.6) in the elevated (30-50 mg/dL) Lp(a) group and 2.1 (1.0-4.3) (p=0.06) in high Lp(a) group. Finally, when adjusting for all covariates; age, BMI, DM, smoking, cholesterol, and LDL-C, hazard ratios were 1.0 (0.4-2.7) in the elevated (30-50 mg/dL) Lp(a) group and 2.1 (1.0-4.2) (p=0.05) in high Lp(a) group thus demonstrating a limited impact on other traditional risk factors on the impact of Lp(a) in ACS development.

DiscussionThis single centre prospective cohort study investigated the association of Lp(a) levels with ACS events in older community-dwelling men (median follow up time = 4.5 years). Serum levels of Lp(a) greater than 50 mg/dL associated with increased risk of ACS after adjustment for multiple covariates, including LDL-C, cholesterol, age and BMI. No difference, however, was observed between elevated (30-50 mg/dL) and low levels (202016 ESC/EAS Guidelines for the Management of Dyslipidaemias., 23Lp(a) (Lipoprotein[a]) Concentrations and Incident Atherosclerotic Cardiovascular Disease.]. Recently, two large placebo-controlled trials of cholesterol-lowering PCSK9 (proprotein convertase subtilisin kexin 9) inhibitors have demonstrated a secondary effect of reducing circulating Lp(a) [24Lipoprotein(a) lowering by alirocumab reduces the total burden of cardiovascular events independent of low-density lipoprotein cholesterol lowering: ODYSSEY OUTCOMES trial.]. Reductions of Lp(a) by alirocumab were associated with a reduction in primary peripheral arterial events, venous thromboembolism and major cardiovascular events, independent of low-density lipoprotein cholesterol reduction, supporting the concept that circulating Lp(a) is modifiable CVD risk factor.Recent advances of RNA technology have led to the development of Lp(a)-specific oligonucleotide-based therapies, such as mipomersen; which target ApoB [25The Effect of Mipomersen in the Management of Patients with Familial Hypercholesterolemia: A Systematic Review and Meta-Analysis of Clinical Trials.] and pelacarsen; which targets Apo(a) [26Karwatowska-Prokopczuk E. et al.Prevalence and influence of LPA gene variants and isoform size on the Lp(a)-lowering effect of pelacarsen.]. Mipomersen however, is more potent for LDL than Lp(a), demonstrating best results in hypercholesterolemia patients [27Apolipoprotein B Synthesis Inhibition With Mipomersen in Heterozygous Familial Hypercholesterolemia.], whereas the antisense oligonucleotide; pelacarsen, directly targets the cholesterol portion of Lp(a); Lp(a)-C, reducing levels up to 80% [28Effect of Pelacarsen on Lipoprotein(a) Cholesterol and Corrected Low-Density Lipoprotein Cholesterol.]. These agents operate independently of isoform size and genetic variant and have shown promising results in clinical trials, with further therapies such as Amgen’s RNAi ARC-LPa, targeting of LPA mRNA in early development [29Treatment and prevention of lipoprotein(a)-mediated cardiovascular disease: the emerging potential of RNA interference therapeutics.]. More recently, siRNA targeting LPA mRNA, was trialled in a small study (n=32). A single dose of siRNA SLN360 was given to increase its selective uptake and concentration within hepatocytes, binding and degrading the apo(a) mRNA and reducing overall Lp(a) plasma levels for 150 days [30Single Ascending Dose Study of a Short Interfering RNA Targeting Lipoprotein(a) Production in Individuals With Elevated Plasma Lipoprotein(a) Levels.]. With studies such as these highlighting the urgent need for Lp(a) specific therapies, the current guidelines must still focus on managing and mitigating the risk associated with high Lp(a) levels prior to the need for intervention until a specific therapy is widely available [7Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement From the American Heart Association., 82021 Canadian Cardiovascular Society Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in Adults., 312018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.]. Within the current study, this data demonstrates that integration of classical risk factors into HR models do not alter the risk for ACS, further evidencing the critical importance of incorporating Lp(a) levels in patient treatment plans. These findings support the latest guidelines from both the European and Canadian guiding bodies, which recommend a one-time Lp(a) screening for all, irrespective of CVD risk and other risk factors [7Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement From the American Heart Association., 82021 Canadian Cardiovascular Society Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in Adults., 312018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.].Despite strong clinical evidence that reductions in Lp(a) could meaningfully reduce CVD risk [32Lipoprotein(a) and cardiovascular disease: prediction, attributable risk fraction, and estimating benefits from novel interventions.], no specific and effective therapeutic agents targeting Lp(a) are currently available [33A Test in Context: Lipoprotein(a): Diagnosis, Prognosis, Controversies, and Emerging Therapies.]. Recent studies have demonstrated that Lp(a) remains a risk factor in individuals who are currently undergoing treatment with statins [34Baseline and on-statin treatment lipoprotein(a) levels for prediction of cardiovascular events: individual patient-data meta-analysis of statin outcome trials.], and in those with prior cardiovascular events on secondary prevention therapies [35Lipoprotein(a)-Lowering by 50 mg/dL (105 nmol/L) May Be Needed to Reduce Cardiovascular Disease 20% in Secondary Prevention: A Population-Based Study.]. Statins have been shown to modestly increase Lp(a) when compared to placebo [36Statin therapy increases lipoprotein(a) levels.] thus demonstrating the urgent need for targeted Lp(a) treatment. Despite the increase in Lp(a) levels, statins remain at the frontline of treatment for those with high cholesterol. Evidence is available from mechanistic, observational, and mendelian randomization studies to support a causal role of Lp(a) in the development of a multitude of CVD subtypes [32Lipoprotein(a) and cardiovascular disease: prediction, attributable risk fraction, and estimating benefits from novel interventions., 37Lipoprotein(a) levels are associated with aortic valve calcification in asymptomatic patients with familial hypercholesterolaemia., 38Oxidized Phospholipids and Risk of Calcific Aortic Valve Disease: The Copenhagen General Population Study., 39Association of Long-term Exposure to Elevated Lipoprotein(a) Levels With Parental Life Span, Chronic Disease-Free Survival, and Mortality Risk: A Mendelian Randomization Analysis.], within both a primary [40Cardiovascular disease risk associated with elevated lipoprotein(a) attenuates at low low-density lipoprotein cholesterol levels in a primary prevention setting., 41Clinical Utility of Lipoprotein(a) and LPA Genetic Risk Score in Risk Prediction of Incident Atherosclerotic Cardiovascular Disease.] and secondary [35Lipoprotein(a)-Lowering by 50 mg/dL (105 nmol/L) May Be Needed to Reduce Cardiovascular Disease 20% in Secondary Prevention: A Population-Based Study.] setting, but few studies aside from The Copenhagen City Heart Study [42

Kamstrup, P.R., et al., Extreme lipoprotein(a) levels and risk of myocardial infarction in the general population: the Copenhagen City Heart Study. (1524-4539 (Electronic)).

, 43Association of LPA Variants With Risk of Coronary Disease and the Implications for Lipoprotein(a)-Lowering Therapies: A Mendelian Randomization Analysis.], have undertaken such evaluations over such a long time period. While underpowered, these data may suggest that elevated Lp(a) could play a more important role in those patients who have already undergone a previous AMI or IHD (supplemental figure 1 & 2), in comparison to those with no previous CVD (supplemental figure 3 & 4). However, these finding would need to be assessed in a larger cohort to confirm. Notably, within the whole population, when adjusting for covariates in a cox hazard proportional model, Lp(a) is significantly influenced by LDL-C in those patients with previous CVD. Conversely, LDL-C has no significant impact in participants with previous AMI and IHD, despite no significant difference in LDL-C levels within the groups in both the total population assessed and the prior CVD cohort.Utilizing absolute cut offs of Lp(a), this study identified an increased risk of ACS in those with Lp(a) values >50 mg/dL. Previous studies with quintile or quartile stratification of Lp(a) analysis have identified this increased risk [3Genetic Variants Associated with Lp(a) Lipoprotein Level and Coronary Disease.]; however, the application of absolute cut-off values of Lp(a) is useful to support future clinical integration of this measurement. Whilst absolute cut offs may be beneficial for decision making within the clinic, a linear increase in atherosclerotic cardiovascular disease risk has been observed with Lp(a) concentration, suggesting that even a modest increase in Lp(a) may be determinantal [23Lp(a) (Lipoprotein[a]) Concentrations and Incident Atherosclerotic Cardiovascular Disease.]. Notably, within this study, the observation that no difference was found between low (44Elevated lipoprotein(a) and risk of coronary heart disease according to different lipid profiles in the general Chinese community population: the CHCN-BTH study.] and a linear relationship with both fatal and non-fatal cardiovascular risk [45Independent Association of Lipoprotein(a) and Coronary Artery Calcification With Atherosclerotic Cardiovascular Risk., 46Lipoprotein(a), Immune Cells and Cardiovascular Outcomes in Patients with Premature Coronary Heart Disease.]. This may, in part, be due to the disparity in race between the two studies. The aforementioned study focused on a Chinese cohort. Chinese populations are demonstrated to have lower Lp(a) levels than Caucasians [47

Banerjee, D., et al., Racial and Ethnic Variation in Lipoprotein (a) Levels among Asian Indian and Chinese Patients. (2090-3049 (Electronic)).

], and thus an elevation of Lp(a) of over 30 mg/dL in a Chinese population may be comparable to a 50 mg/dL or greater Lp(a) level in Caucasians, therefore providing a potential explanation for the discrepancy in risk.Moreover, within this study, the absolute risk of developing ACS with 8 years of baseline more than doubled within participants with Lp(a) over 50 mg/dL. This risk is greater than seen in previous studies that have integrated absolute cut-offs of Lp(a) and CVD risk. Previous observations have demonstrated the risk of Lp(a) in a short-term follow up studies (predominantly 5 years or less) and in a secondary prevention setting, but only a few reports have assessed Lp(a) as a long-term predictive measure in mixed primary and secondary prevention community based cohorts in an over 60 population [48Ariyo A.A. Thach C. Tracy R. Lp(a) Lipoprotein, Vascular Disease, and Mortality in the Elderly.]. A meta-analysis of placebo-controlled randomised statin trials, which combined both primary and secondary prevention populations, demonstrated that individuals with high Lp(a) levels (>50 mg/dL) had an age and sex adjusted hazard ratio of 1.31 in CVD, compared to those with Lp(a) levels under 15 mg/dL. Similarly, in a large observational Danish prospective cohort study, those with Lp(a) values between 50-100 mg/dL had a hazard ratio of ∼1.5 compared to those with Lp(a) Strengths of this analysis involve the length of follow up. In the current study, the population had an average age of 71.5, however subsequent ACS occurred on average of 5 years later. Therefore, this report demonstrates the importance of studies with long follow-up periods, as current short-term studies may miss the true impact of Lp(a) on ACS outcomes. In addition, this study used the same commercially available standardized assay with minimal isoform bias measured over a short time period (5-day) for all samples, facilitating low inter-assay variability. Measurement of Lp(a) is currently not uniformly calibrated and standardized, and studies which utilize different methods over long periods of time will result in variable absolute measurements. Conversely, this study is not without limitations. The present investigation is limited in that only Caucasian males residing in France were recruited in the study population. However, these results are consistent with reports that included more diverse ethnic backgrounds and those assigned female at birth. There are currently conflicting results in literature regarding sex differences in the predictive value of Lp(a), showing both no association of Lp(a) in large prospective cohorts in a female-only subset analysis [48Ariyo A.A. Thach C. Tracy R. Lp(a) Lipoprotein, Vascular Disease, and Mortality in the Elderly.] and also contrary findings that females have a stronger relative risk in regards to Lp(a) then men in a secondary prevention setting [49Gender differences in plasma levels of lipoprotein (a) in patients with angiographically proven coronary artery disease., 50Gender difference in lipoprotein(a) concentration as a predictor of coronary revascularization in patients with known coronary artery disease., 51Association of sex-specific differences in lipoprotein(a) concentrations with cardiovascular mortality in individuals with type 2 diabetes mellitus.]. Secondly, non-fasting blood draws were used for Lp(a) measurement which could influence values, however the use of non-fasting lipid profiles is now widely endorsed in risk assessment.Article InfoPublication History

Accepted: June 12, 2022

Received in revised form: June 7, 2022

Received: May 13, 2022

Publication stageIn Press Accepted ManuscriptFootnotes

Sources of Funding

This work was supported by Agence Nationale de la Recherche grants ANR-07-PHYSIO-023-01 and ANR-10-BLAN-1137 to PS and RC. National Institutes of Health grants R01HL147095, R01HL141917, and R01HL136431 to EA. Boehringer Ingelheim Fonds MD fellowship to JZ.

Disclosures

The authors state no disclosures.

CRediT author statement

Francesca Bartoli-Leonard: Conceptualization, methodology, software, analysis, investigation, data curation, writing, editing, reviewing, visualization, Mandy E. Turner: Analysis, writing, editing, reviewing, Jonas Zimmer: Conceptualization, methodology, writing, Roland Chapurlat: Resources, data curation, funding acquisition,Tan Pham: Investigation, data curation, Masanori Aikawa: Funding acquisition, Aruna D. Pradhan: Writing, editing, reviewing, Pawel Szulc: Conceptualization, methodology, resources, data curation, writing, editing, reviewing, supervision, funding acquisition, Elena Aikawa: Supervision, funding acquisition, writing, editing, reviewing.

Identification

DOI: https://doi.org/10.1016/j.jlr.2022.100242

Copyright

© 2022 THE AUTHORS. Published by Elsevier Inc on behalf of American Society for Biochemistry and Molecular Biology.

User License Creative Commons Attribution (CC BY 4.0) | ScienceDirectAccess this article on ScienceDirect