Clinical Utility of Lipoprotein(a) and LPA Genetic Risk Score in Risk Prediction of Incident Atherosclerotic Cardiovascular Disease

Abstract

Importance: Lipoprotein(a) is a highly heritable biomarker independently associated with atherosclerotic cardiovascular disease (ASCVD). It is unclear whether measured lipoprotein(a) or genetic factors associated with lipoprotein(a) can provide comparable or additional prognostic information for primary prevention.

Objective: To determine whether a genetic risk score (GRS) comprising 43 variants at the LPA gene, which encodes apolipoprotein(a), has clinical utility in assessing ASCVD risk compared with and in addition to lipoprotein(a) measurement.

Design, setting, and participants: The UK Biobank is a prospective observational study of approximately 500 000 volunteers aged 40 to 69 years who were recruited from 22 sites across the United Kingdom between 2006 and 2010. Using externally derived weights, an LPA GRS was calculated for 374 099 unrelated individuals with array-derived genotypes and lipoprotein(a) measures. Data were analyzed from April 2020 to March 2020.

Exposures: Measured lipoprotein(a) and LPA GRS.

Main outcomes and measures: We estimated the associations between measured lipoprotein(a) and LPA GRS with the incidence of ASCVD (peripheral arterial disease, coronary artery disease, myocardial infarction, ischemic stroke, and cardiovascular mortality) using Cox proportional hazards models. To determine the utility of using measured lipoprotein(a) and LPA GRS as risk enhancers for ASCVD, we assessed the potential improvement in ASCVD risk discrimination by QRISK3 and Pooled Cohort Equations among individuals with borderline to intermediate risk (n = 113 703 and 144 350, respectively).

Results: The mean age of the overall study population was 57.6 years, and 204 355 individuals were female (54.6%). During a median follow-up of 11.1 years (interquartile range, 1.4 years), 15 444 individuals developed an incident ASCVD event (5.1%). The LPA GRS explained approximately 60% of the variation in measured lipoprotein(a) for White/European individuals. Independently, both lipoprotein(a) and LPA GRS were associated with incident, composite ASCVD (hazard ratio per 120 nmol/L increase, 1.26; 95% CI, 1.23-1.28 vs hazard ratio, 1.29; 95% CI, 1.26-1.33; P < .001). The association between LPA GRS and ASCVD was substantially attenuated after adjusting for measured lipoprotein(a). Adding measured lipoprotein(a) or LPA GRS to QRISK3 provided modest improvements to the risk discrimination of incident ASCVD events (area under the receiver operating curve, 0.640; 95% CI, 0.633-0.647 vs 0.642; 95% CI, 0.635-0.649 for both; P = .005 and P = .01, respectively).

Conclusions and relevance: When indicated, cardiovascular risk assessment with lipoprotein(a) at middle-age may include direct measurement or an LPA GRS.

Figure 1.. Flow Diagram of Analyses

ASCVD indicates atherosclerotic cardiovascular disease; GRS, genetic risk score; Lp(a), lipoprotein(a); LDL-C, low-density lipoprotein cholesterol. To convert LDL-C to millimoles per liter, multiply by 0.0259; Lp(a) to milligrams per deciliter, divide by 2.4.

Figure 2.. Association of Measured Lipoprotein(a) (Lp[a]) and LPA Genetic Risk Score (GRS) With Incident Atherosclerotic Cardiovascular Disease (ASCVD) Among Individuals of White/European Ethnicity

The risk of incident ASCVD events is depicted as hazard ratios (HRs) and 95% confidence intervals for peripheral arterial disease (PAD), ischemic stroke, coronary artery disease (CAD), myocardial infarction (MI), cardiovascular disease (CVD) mortality, and composite ASCVD for 292 963 individuals not using cholesterol-lowering medication and without prevalent ASCVD at enrollment. Hazard ratios are scaled to depict a 120-nmol/L increase in measured Lp(a) or LPA genetic risk score (GRS) levels. All Cox proportional hazard models included age, sex, assessment center, genotyping batch, and the first 5 principal components of ancestry. Adj indicates adjusted.

Figure 3.. Association of Elevated LPA Genetic Risk Score (GRS) With Cardiovascular Risk When Individuals Are Matched for Measured Lipoprotein(a) (Lp[a]) Levels

A, Crude time-to-first incident, composite atherosclerotic cardiovascular disease (ASCVD) event curves are shown for individuals with and without an elevated LPA GRS after matching for Lp(a) levels (matched LPA GRS-: predicted Lp(a) <120 nmol/L; matched LPA GRS+: predicted Lp(a) ≥120 nmol/L). B, The proportion of individuals with and without an elevated LPA GRS. C, Incidence rate of composite ASCVD events per 1000 person-years are displayed for 50-nmol/L bins of measured Lp(a) levels. Data points for incidence rates are displayed with the 95% confidence interval stratified by elevated LPA GRS cutoff levels of 120 nmol/L.