Ancestry, Lipoprotein(a), and Cardiovascular Risk Thresholds

JACC Review Topic of the Week

Sotirios Tsimikas, MD; Santica M. Marcovina, PHD, SCD

J Am Coll Cardiol. 2022;80(9):934-946.

Abstract and Introduction

Abstract

This study reviews ancestral differences in the genetics of the LPA gene, risk categories of elevated lipoprotein(a) [Lp(a)] as defined by guidelines, ancestry-specific Lp(a) risk, absolute and proportional risk, predictive value of risk thresholds among different ancestries, and differences between laboratory vs clinical accuracy in Lp(a) assays. For clinical decision-making, the preponderance of evidence suggests that the predictive value of Lp(a) does not vary sufficiently to mandate the use of ancestry-specific risk thresholds. This paper interprets the literature on Lp(a) and ancestral risk to support: 1) clinicians on understanding cardiovascular disease risk in different ancestral groups; 2) trialists for the design of clinical trials to ensure adequate ancestral diversity to support broad conclusions of drug effects; 3) regulators in the evaluation of the design and interpretation of results of Lp(a)-lowering trials with different Lp(a) inclusion thresholds; and 4) clinical laboratories to measure Lp(a) by assays that discriminate risk thresholds appropriately.

Introduction

With a rapidly accumulating evidence base over the last decade, lipoprotein(a) [Lp(a)] is now generally accepted as a genetic, independent risk factor for cardiovascular disease (CVD) and aortic stenosis.^[1,2] This is reflected by at least 7 national and international societies recommending the measurement of Lp(a) levels in either all adults at least once or in subjects in intermediate- or high-risk categories based on clinical characteristics. The development of new therapeutic modalities to inhibit translation of LPA messenger RNA in the hepatocyte^[3] has allowed the testing of the "Lp(a) hypothesis," namely that lowering Lp(a) levels will lead to a reduction in CVD events. A major unresolved issue in the field is the appropriate interpretation of current data of risk thresholds according to ancestry. The relevance of this issue is reflected by the clinically significant differences in population mean Lp(a) levels.^[4] Elevated Lp(a) may influence absolute risk across populations with differences in population mean Lp(a) levels and also influence proportional risk at the individual level irrespective of the population mean prevalence.

This review summarizes and interprets the literature on ancestral differences in Lp(a) levels, their measurement with current techniques, and their potential impact in the emerging field of Lp(a) in clinical trials, regulatory implications, clinical investigation, and clinical care.^[5]

1 2 3 4 5 6 7 8 9 10

Next Section

Table 1. Guideline and Consensus Statements on Measuring Elevated Lp(a)
	Lp(a) Risk Thresholds	Major Recommendation for Measuring Lp(a)
American Society of Apheresis, 2017⁵⁶	>30 mg/dL (>45 nmol/L)	Subjects at intermediate or high risk, premature CVD, FH, family history of premature CVD without elevated LDL-C, or recurrent CVD despite statin treatment
ACC/AHA, 2018⁵⁷	>50 mg/dL (>125 nmol/L)	Risk-enhancing factor at levels >50 mg/dL or >125 nmol/L. Family history of premature CVD or personal history of CVD not explained by major risk factors
ESC/EAS, 2019⁵⁸	>50 mg/dL >180 mg/dL (>430 nmol/L) equivalent to heterozygous FH	At least once in each adult person's lifetime
NLA, 2019⁴⁶	>50 mg/dL or >100 nmol/L (based on >80th percentile in Caucasian subjects)	Adults with personal history of premature CVD, or first-degree relatives with premature CVD, or LDL-C >190 mg/dL or suspected FH
HEART UK, 2019⁴²	Risk thresholds: 32–90 nmol/L, minor 90–200 nmol/L, moderate 200–400 nmol/L, high >400 nmol/L, very high	Adults with a personal or family history of premature ASCVD; first-degree relatives who have Lp(a) levels >200 nmol/L patients with FH; patients with calcific aortic valve stenosis and those with borderline (but <15%) 10-year risk of a cardiovascular event
AACE/ACE, 2020⁵⁹	>50 mg/dL	All patients with clinical ASCVD, premature or recurrent ASCVD despite LDL-C lowering; family history of premature ASCVD and/or increased Lp(a); South Asian or African ancestry, 10-year ASCVD risk ≥10% (primary prevention setting), personal or family history of aortic valve stenosis; patients with refractory elevations of LDL-C despite aggressive LDL-C–lowering therapy (ie, statin resistance)
Canadian Cardiovascular Society, 2021⁶⁰	>50 mg/dL; >100 nmol/L in primary prevention	Once in a person's lifetime as a part of the initial lipid screening

AACE/ACE = American Association of Clinical Endocrinologists/American College of Endocrinology; ACC/AHA = American College of Cardiology/American Heart Association; ASCVD = atherosclerotic cardiovascular disease; CVD = cardiovascular disease (coronary heart disease plus ischemic stroke); ESC/EAS = European Society of Cardiology/European Atherosclerosis Society; FH = familial hypercholesterolemia; Lp(a) = lipoprotein(a); NLA = National Lipid Association.

Table 2. Studies Reporting CVD Outcomes with Comparison Ancestry Data According to Plasma Lp(a) Levels
	Events/Subjects	Year	Follow-Up	Lp(a), Median (IQR)	OR/HR (95% CI)
ERFC³⁰		2009	1.3M pt y	mg/dL	Per 3.5-X (~1 SD) usual Lp(a)
Black	261/4,546			12.6 (4.9–32.1)	CHD: 1.05 (0.90–1.23)
White	7,540/95,753			12.6 (4.9–32.1)	CHD: 1.14 (1.09–1.19)
ARIC²³		2012	20 y	mg/dL	Per 1 SD	HR Q5 vs Q1
US Black	676/3,647			12.8 (7.1–21.7)	CVD: 1.13 (1.04–1.23) CHD: 1.11 (1.00–1.22) Stroke: 1.21 (1.06–1.39)	CVD: 1.35 (1.06–1.74)
White	1,821/9,851			4.3 (1.7–9.5)	CVD: 1.09 (1.04–1.15) CHD: 1.10 (1.05–1.16) Stroke: 1.07 (0.97–1.19)	CVD: 1.27 (1.10–1.47)
MESA²²		2015	8.5 y	mg/dL	CHD: Per 1 SD	Threshold ≥50 vs <50 nmol/L
U.S. Black	66/1,323			35.1 (20.4–61.6)	1.49 (1.09–2.04)	1.69 (1.03–2.76)
U.S. White	102/1,677			12.9 (5.8–29.6)	1.22 (1.02–1.80)	1.82 (1.15–2.88)
U.S. Chinese	18/548			12.9 (7.7–23.4)	1.08 (0.65–1.80)	1.04 (0.22–4.98)
U.S. Hispanic	49/1,044			13.1 (6.3–28.8)	1.14 (0.86–1.50)	2.37 (1.17–4.78)
DHS²⁵		2019	9.5 y	nmol/L		MACE ^a HR Q4 vs Q1
U.S. Black	146/1,792			79.0 (43–132)		3.01 (1.27–7.11)
U.S. White	52/1,030			26.9 (10–69)		1.83 (0.88–3.80)
U.S. Hispanic	13/597			21.3 (9–46)		5.43 (0.79–37.2)
INTERHEART²¹		2019	NA	mg/dL (5%-95%)		MI OR Q4 vs Q1
African	294/775			27.1 (4.1–110.6)		0.92 (0.65–1.31)
Arab	528/1,352			9.8 (2.3–53.4)		1.13 (0.80–1.59)
Chinese	234/443			18.1 (2.0–82.6)		1.62 (1.20–2.19)
European	951/856			11.5 (2.0–99.1)		1.36 (1.05–1.76)
Latin American	731/1,469			14.7 (2.0–100.0)		1.67 (1.25–2.22)
South Asian	948/1,829			18.9 (3.2–88.2)		2.14 (1.59–2.89)
SE Asian	507/1,221			12.9 (2.4–74.2)		1.83 (1.17–2.88)
UK Biobank²⁰		2021	11.2 y	nmol/L	ASCVD ^b Per Lp(a) 50 nmol/L	≥90th vs <90th Percentile
U.K. White	18,764/41,7687			18.7 (7.4–72.7)	1.11 (1.10–1.12)	1.52 (1.46–1.59)
U.K. S. Asian	558/8,402			31.3 (11.9–69.4)	1.10 (1.04–1.16)	1.35 (1.03–1.78)
U.K. Black	242/7,013			74.8 (43.8–133.9)	1.07 (1.00–1.15)	1.51 (1.05–2.18)
REGARDS¹⁹		2022	~10 y	nmol/L	CHD ^c per SD	CHD Q4 vs Q1
Black	967/967			100.1 (47.5–185.8)	1.26 (1.02–1.56)	1.45 (0.78–2.69)
White	981/981			23.4 (8.2–112.9)	1.16 (1.02–1.31)	1.36 (0.93–1.98)
						CHD/Stroke Race-Specific Q4 vs Q1 1.68 (1.12–2.52) 1.34 (0.92–1.94)

DHS Quartiles (Q) for Lp(a): Q1 <19.6, Q2 >19.6 to <49.9, Q3 >49.9 to <110.5, and Q4 >110.5 nmol/L. Atherosclerosis Risk in Communities (ARIC): Q1–5 Black individuals, 0.1 to 6.1, >6.1 to 10.3, >10.3 to 15.8, >15.8 to 24, and >24 to 81.7. ARIC: Q1–5 White individuals, 0.1 to 1.5, >1.5 to 3.1, >3.1 to 6.0, >6.0 to 13.5, and >13.5 to 80.3. ^aAdjusted for sex, hypertension, diabetes mellitus, smoking, age in deciles, body mass index, low-density lipoprotein cholesterol per 25 mg/dL, high-density lipoprotein cholesterol per 10 mg/dL, and log2 triglyceride. ^bComposite of coronary artery disease (myocardial infarction and its acute complications, coronary artery bypass graft surgery, or percutaneous angioplasty/stent placement) and ischemic stroke (cerebral infarction due to thrombosis or cerebral atherosclerosis or cerebrovascular syndromes). ^cCHD events include a myocardial infarction hospitalization or CHD death (ie, a death suspected to be CHD related without evidence of a noncoronary cause).
CHD = coronary heart disease; CVD = cardiovascular disease (coronary heart disease plus ischemic stroke); DHS = Dallas Heart Study; ERFC = Emerging Risk Factors Collaboration; Lp(a) = lipoprotein(a); MACE = cardiac death, nonfatal myocardial infarction, stroke/transient ischemic attack, unstable angina requiring hospitalization, arterial revascularization; MESA = Multi-Ethnic Study of Atherosclerosis; MI = myocardial infarction; pt = patient; REGARDS = Reasons for Geographic and Racial Differences in Stroke.