A single combined assessment of low-density lipoprotein cholesterol (LDL-C), lipoprotein(a) (Lp[a]), and high-sensitivity C-reactive protein (hsCRP) detects individuals at increased long-term risk of atherosclerotic cardiovascular disease (ASCVD) via independent pathways, according to a study.
“Risk increased in a stepwise fashion across biomarker quintiles and was 2.4-fold higher among individuals with all three biomarkers elevated compared with those with none, an observation consistent across participants of African, European, and South Asian Surinamese descent,” the investigators said. [J Am Coll Cardiol 2026;88:1964-1977]
A total of 378 ASCVD events occurred over a median follow-up of 8.9 years. Compared with participants in the bottom quintile, those in the top quintile had elevated ASCVD risks (adjusted subdistribution hazard ratio [aHR] for LDL-C, 1.39, 95 percent confidence interval [CI], 1.09‒1.79; aHR for Lp[a], 1.86, 95 percent CI, 1.54‒2.25; aHR for hsCRP, 1.51, 95 percent CI, 1.34‒1.71).
No interaction occurred on either the multiplicative or additive scale based on formal interaction assessment. The risk of ASCVD increased with the number of elevated biomarkers, and participants with all three elevated biomarkers had an HR of 2.44 (95 percent CI, 1.46‒4.09).
Furthermore, dose-response patterns were consistent across ethnic groups (p=0.75 for interaction), but the absolute ASCVD risk was nearly threefold higher among South Asian Surinamese participants.
Notably, adding Lp(a) and hsCRP to a clinical model resulted in a modest improvement in discrimination (Δ area under the curve, 0.006; p=0.030) and reclassification (continuous net reclassification improvement, 0.165, 95 percent CI, 0.050‒0.289) at 10 years.
“These findings support a pathway-oriented approach to ASCVD risk assessment and highlight the potential value of combined biomarker evaluation, particularly for identifying individuals with high absolute risk across diverse populations,” the investigators said.
Risk algorithms
Current guideline-recommended risk algorithms, including PREVENT-ASCVD and SCORE2, combine both demographic and conventional risk factors (eg, atherogenic cholesterol burden) but do not include specific assessment of genetically determined Lp(a) and inflammatory markers, which potentially account for significant residual risk. [J Am Coll Cardiol 2014;63:25; Eur Heart J 2021;42:2439-2454]
In HELIUS, participants with increased Lp(a) and hsCRP, which neither appears in standard clinical risk models, had more than twofold higher risk of ASCVD than those without these elevations, according to the investigators.
“Apparent and optimism-corrected estimates were nearly identical, indicating limited overfitting, although both models were derived and evaluated within the same cohort,” the investigators said.
“These findings nonetheless suggest that Lp(a) and hsCRP capture prognostically relevant biology not reflected in conventional risk factors, supporting further investigation and external validation of their incorporation into clinical risk algorithms,” they added.
The current study involved 15,676 HELIUS participants with no previous myocardial infarction or ischaemic stroke (mean age 44.4 years, 56.0 percent women). Incident ASCVD was identified using nationwide registries.
The investigators examined associations across increasing biomarker quintiles and joint elevations using Fine-Gray competing-risk models and assessed additive interaction using the relative excess risk due to interaction, attributable proportion, and synergy index.
Finally, time-dependent area under the curve and continuous net reclassification improvement were used to assess the incremental predictive value of Lp(a) and hsCRP beyond a PREVENT-ASCVD‒based clinical model.