Ethnic differences in lipoprotein(a)-associated ASCVD risk and the impact of standard modifiable cardiovascular risk factors: insights from the multi-ethnic HELIUS cohort

Background Plasma lipoprotein(a) [Lp(a)] levels are genetically determined, vary by ethnicity and independently contribute to atherosclerotic cardiovascular disease (ASCVD). The relationship between Lp(a) and ASCVD risk across diverse ethnicities and its interaction with standard modifiable cardiovascular risk factors (SMuRFs) in relation to ASCVD remain unclear. Purpose To explore ethnic differences in Lp(a) levels, their association with ASCVD, and the impact of SMuRFs on Lp(a)-related risk in a multi-ethnic, population-based cohort. Methods A total of 15,758 participants from the HEalthy Life In an Urban Setting (HELIUS) cohort of Dutch, African Surinamese, South-Asian Surinamese, Ghanaian, Turkish and Moroccan descent without overt ASCVD with available Lp(a) measurements were included. SMuRFs were defined as hypertension (BP ≥140/90 mmHg), dyslipidaemia (LDL-C ≥95th percentile), diabetes (fasting glucose ≥7.0 mmol/L or glucose-lowering medication), and current smoking. Cox proportional hazards model were used to evaluate associations between Lp(a) and SMuRFs with ASCVD (composite of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke or coronary revascularization). Continuous net reclassification improvement (NRI) analyses were performed to assess the added value of high Lp(a) over SCORE2 alone in risk classification. Results Median Lp(a) levels varied by ethnicity (p < 0.05), being highest in Ghanaians (85 nmol/L) and African Surinamese (69 nmol/L), then Moroccan (38 nmol/L), South-Asian Surinamese (35 nmol/L), Turkish (35 nmol/L), and lowest in Dutch (16 nmol/L). Over an 8.9-year median follow-up, 387 ASCVD events occurred. High Lp(a) was associated with an increased ASCVD risk. ASCVD incidence rose 2.2-fold from 1.74 (95% CI, 1.08–2.46) to 3.85 per 1,000 person-years (95% CI, 2.85–4.92) comparing the lowest to highest deciles. After adjusting for age, sex, and ethnicity, both Lp(a) above the cohort-wide 80th percentile (HR 1.34; 95% CI 1.06–1.69) and the ethnicity-specific 80th percentile (HR 1.28; 95% CI 1.01–1.60) were associated with ASCVD. Only the ethnicity-specific threshold significantly improved risk reclassification over SCORE2 (NRI 0.126; 95% CI 0.003–0.238). In those with Lp(a) above the ethnicity-specific 80th percentile, predicted ASCVD risk was highest in the presence of all SMuRFs (14.3%) and declined stepwise to 8.0%, 4.9%, 2.3%, and 0.9% with three, two, one, and zero SMuRFs, respectively. Conclusion Marked ethnic differences in Lp(a) levels were observed, and high Lp(a) was consistently associated with future ASCVD. Ethnicity-specific Lp(a) thresholds significantly improved reclassification in contrast to a universal threshold in primary prevention. Moreover, in individuals with high Lp(a), ASCVD risk declined stepwise with fewer concomitant SMuRFs. These findings underscore the importance of using ethnicity-specific Lp(a) thresholds to tailor global risk factor management in primary prevention.Higher Lp(a) is associated with ASCVD Lp(a), SMuRFs and ASCVD risk