First nmol/L-standardised Lp(a) distribution in Japanese patients proposes 25 and 125 nmol/L risk thresholds, LEAP study (J Atheroscler Thromb 2026)

Original title: Lipoprotein(a) Distribution and Cardiovascular Risk in Japan: Insights from a Multi-center LEAP Study

J Atheroscler Thromb · 2 May 2026 · 6

Yamada T, Tada H, Ogura M, Funabashi S, Yoshida H, Hiraishi C, Ebihara T, Hadano T, Masaki T, Takeji Y, Fukatsu K, Egawa M et al.

Multicenter LEAP study of 6,173 patients from six Japanese institutions, harmonising Lp(a) values to nmol/L across different immunoassays to characterise its distribution and cardiovascular risk association in a Japanese population. Median Lp(a) was 20.88 nmol/L with a right-skewed distribution; elevated levels were significantly associated with coronary artery disease (CAD), atherosclerotic cardiovascular disease, chronic kidney disease and familial hypercholesterolaemia, while diabetes was linked to lower Lp(a). ROC analysis identified 25 nmol/L as a screening threshold and 125 nmol/L as a high-risk boundary, with CAD prevalence rising stepwise across categories: 20.2% at 25 nmol/L or below, 30.3% between 25 and 125 nmol/L, and 39.9% above 125 nmol/L. The authors propose these as provisional Japanese-specific risk-stratification thresholds.

Original abstract

Aim: Although lipoprotein(a) [Lp(a)] is recognized as an independent risk factor for cardiovascular disease (CVD), its distribution and risk thresholds in the Japanese population remain unclear. This study aimed to characterize the distribution of Lp(a) in Japanese clinical settings and evaluate its association with CVD risk, including coronary artery disease (CAD), atherosclerotic cardiovascular disease (ASCVD), and familial hypercholesterolemia (FH).

Methods: The LEAP study is a multicenter retrospective cohort analysis of 6,173 patients from six Japanese institutions. The Lp(a) values were harmonized to nmol/L using kit-specific conversion equations, enabling a consistent analysis across different immunoassays. Associations with CAD, ASCVD, chronic kidney disease (CKD), diabetes mellitus (DM), and FH were assessed. The risk thresholds were defined using a receiver operating characteristic analysis.

Results: The median Lp(a) concentration was 20.88 nmol/L, with a right-skewed distribution. Elevated Lp(a) levels were significantly associated with CAD, ASCVD, CKD, and FH. In contrast, DM was related to lower Lp(a). An ROC analysis identified 25 nmol/L as a screening threshold and 125 nmol/L as a high-risk boundary. The CAD prevalence increased stepwise across these categories: 20.2% (≤ 25 nmol/L), 30.3% (25-125 nmol/L), and 39.9% (＞125 nmol/L).

Conclusion: This study provides the first nmol/L-standardized Lp(a) distribution in Japanese patients and proposes provisional thresholds for clinical risk stratification.

Summary written by lp-a.org from the published abstract; figures as published. Page updated 17 August 2026. Methods.