Lipoprotein(a) levels and cholesterol efflux capacity: inverse association and impact on atherosclerosis in familial hypercholesterolemia patients on lipid-lowering therapy

Abstract Backgrounds Familial hypercholesterolemia (FH) patients have higher serum lipoprotein(a) (Lp(a)) levels than non-FH patients, and high Lp(a) levels raise atherosclerotic cardiovascular disease (ASCVD) risk 1). We previously reported that decreased cholesterol efflux capacity (CEC) is a risk marker for ASCVD in patients with statin-treated FH 2). Others have reported that Lp(a) interferes with the enhancement of ABCA1-mediated cholesterol efflux by plasminogen and that ABCA1-mediated CEC is lower in patients with Lp(a) >50 mg/dL. However, the association of CEC, not necessarily ABCA1-mediated, with Lp(a) is not clear and has not been studied in patients with FH. It is also unclear whether high Lp(a) and low CEC in patients with FH are associated with the severity of atherosclerosis. Purpose We examined whether Lp(a) levels are negatively associated with ABCA1-independent CEC in patients with FH receiving lipid-lowering therapy. We also examined the impact of the combination of high Lp(a) and low CEC on the presence of corneal arcus, Achilles tendon thickness, carotid intima-media thickness (IMT), and pre-existing ASCVD. Methods This cross-sectional study included 371 patients clinically diagnosed with FH receiving lipid-lowering therapy. CEC was measured in apolipoprotein B-depleted plasma and 3H-cholesterol-labeled J774.1 cells without cAMP-stimulated ABCA1 overexpression. Lp(a) was measured by enzymatic methods (Sekisui Medical, Tokyo, Japan) using an automated analyser. Achilles tendon thickness was measured by X-ray and carotid IMT by ultrasound. Results Mean age was 55±16 years, and 191 (51%) were women. The median serum Lp(a) level was 18.7 mg/dL, higher than the previously reported median level (13 mg/dL) in a Japanese population. One hundred nineteen patients (32%) had serum Lp(a) levels of 30 mg/dL or higher, and they had significantly lower CEC (P=0.0015). This negative association persisted after adjusting for age, sex, BMI, smoking status, serum LDL-cholesterol, serum triglycerides, and eGFR in a multiple linear regression model (beta-coefficient: -0.047, 95% CI: -0.075 - -0.019, P=0.0011). In addition, patients with both Lp(a) > 30 mg/dL and CEC lower than the median had a higher percentage of corneal arcus, thicker Achilles tendons, greater carotid IMT, and a higher prevalence of pre-existing ASCVD compared to patients with either or neither. Conclusions We found an inverse relationship between Lp(a) levels and CEC in FH patients receiving lipid-lowering therapy. Our results suggest that Lp(a) may also inhibit ABCA1-independent CEC, including passive cholesterol diffusion from peripheral cells. Furthermore, the combination of high Lp(a) and low CEC may be a residual risk for patients with FH receiving lipid-lowering therapy. Future studies are needed to determine whether agents that decrease Lp(a) levels will increase CEC and thereby resolve part of the residual ASCVD risk in FH.