Adding Lp(a) to standard risk models improves coronary disease reclassification by up to 42% in patients without modifiable risk factors, BioHEART-CT study of 1,718 finds (Eur J Prev Cardiol 2025)

Aims: Lipoprotein(a) [Lp(a)] has regained attention as an independent cardiovascular risk factor, particularly given emerging therapies entering late-phase clinical trials. Here, we aim to examine the association of Lp(a) with CAD and the potential of Lp(a) as an enrichment criterion for identifying individuals more likely to benefit from screening for subclinical CAD with CT imaging.

Methods: We analysed data from 1,718 adults undergoing CTCA for suspected CAD enrolled in the BioHEART study. Lp(a) levels were measured, and CAD burden was assessed using coronary artery calcium score (CACS) and Gensini scores. Plaque morphology for the most stenotic plaque of each Gensini segment was classified as calcified, non-calcified or mixed. Youden's index with 10,000 bootstraps was used to identify the optimal threshold for increased risk of clinically actionable CAD.

Results: Lp(a) was strongly associated with all CTCA measures of CAD examined. Elevated Lp(a) above 22 nmol/L was linked to more advanced multi-segment (ordinal OR = 1.14 [1.03-1.25]) and multivessel disease (ordinal OR = 1.11 [1.02-1.20]), with a 2.6% increased risk of a CACS >100 for every 10 nmol/L increment. Lp(a) was most strongly associated with mixed plaque burden even after adjusting for traditional risk factors (β = 4.75, p=0.001), but not with non-calcified or calcified plaque. Adding Lp(a) to standard risk models resulted in an overall NRI of 16% [0.06-0.27] and 42% [0.16-0.70] in patients without standard modifiable risk factors.

Conclusion: Our findings suggest Lp(a)'s role in a new clinical pathway: screening patients considered low or intermediate risk, particularly those without standard modifiable risk factors for non-invasive imaging to detect subclinicalCAD.