AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Index Lipoprotein(a) and Polygenic Risk Score Identify Coronary Artery Disease Risk in First-Degree Relatives
By Annink M, Beverloo C, Weijs B et al.
Background: Lipoprotein(a) [Lp(a)] and coronary artery disease (CAD) polygenic risk scores (PRS) are heritable markers of CAD risk. Lp(a) measurement is recommended at least once in adulthood, and PRS is increasingly being considered for clinical risk refinement. However, when elevated Lp(a) or PRS is found in an index individual, it remains unclear whether this information can be used to identify first-degree relatives (FDRs) at higher clinical risk and support cascade screening for these common inherited risk markers. Research Question: We assessed whether Lp(a) and PRS measured in index individuals independently and jointly predict incident CAD in FDRs. Methods: We studied 26,914 genetically inferred FDR pairs from the UK Biobank. Index participants and FDRs were free of atherosclerotic cardiovascular disease at baseline. Exposures were Lp(a) and GPS Mult , a CAD PRS, measured in index participants. Incident FDR CAD was defined as myocardial infarction or coronary revascularization. Cox models with robust standard errors clustered by FDR identifier estimated associations between index exposures and FDR CAD, adjusted for established cardiovascular risk factors. Results: The cohort included 25,106 unique index participants and 24,706 unique FDRs; 78.4% of pairs were siblings. During 13.9 years median follow-up, 973 CAD events occurred in FDRs. Index and FDR values correlated for Lp(a) and GPS Mult , with correlation coefficients of 0.37 and 0.52, respectively. Compared with FDRs whose index participant had Lp(a) in the bottom half of the distribution, FDRs whose index participant had Lp(a) in the highest decile had higher CAD risk (adjusted hazard ratio [aHR], 1.42; 95% CI, 1.15-1.74; P trend =0.002). Each SD increase in index Lp(a) was associated with higher FDR CAD risk (aHR, 1.10; 95% CI, 1.04-1.18; P=0.002), with no evidence of nonlinearity (P nonlinearity =0.48). Compared with FDRs whose index participant had GPS Mult in the bottom half, FDRs whose index participant had GPS Mult in the highest decile had an aHR of 1.69 (95% CI, 1.38-2.08; P trend <0.001). Each SD increase in index GPS Mult was associated with higher FDR CAD risk (aHR, 1.23; 95% CI, 1.15-1.32; P<0.001), with no evidence of nonlinearity (P nonlinearity =0.68). Associations did not differ by FDR sex for either marker. Index GPS Mult was more strongly associated with CAD in younger versus older FDRs (P interaction =0.01). FDRs whose index participant had both Lp(a) and GPS Mult at or above the 80 th percentile had more than double the CAD risk compared with FDRs whose index participant had both markers below the 80 th percentile (aHR, 2.34; 95% CI, 1.56-3.52). Conclusions: Lp(a) and PRS measured in index individuals independently and additively predict CAD in their FDRs. These findings support extending cascade screening beyond monogenic lipid disorders to include common inherited cardiovascular risk markers.