Pleiotropy and eQTL analysis identify shared genetic architecture of lipid traits in African-ancestry populations.

Dyslipidemia is a leading modifiable risk factor for cardiovascular disease (CVD), yet the shared genetic architecture of lipid traits remains poorly characterized, particularly in African-ancestry populations who bear a disproportionate CVD burden but are underrepresented in genomic research. We conducted multi-trait meta-analyses of genome-wide association summary statistics for HDL cholesterol, LDL cholesterol, triglycerides, and total cholesterol in 34 039 individuals of African ancestry, comprising 27 289 discovery participants from three continental African cohorts (ACCME, AWI-Gen, and UGR) and 6750 replication participants from the UK Biobank. Multi-trait analysis and eQTL colocalization across metabolic and cardiovascular tissues were then performed to identify shared genetic signals and characterize their tissue-specific regulatory mechanisms. Multi-trait analysis identified 90 genome-wide significant variants. Of these, 43 were novel variants with no prior exact-variant lipid association records. Eight variants across six genomic regions reached genome-wide significance in the multi-trait analysis and 70 of the 90 primary signals remained significant after total cholesterol was excluded. The strongest associations localized to established lipid loci including APOE/NECTIN2/TOMM40 (P = 8.09 × 10-133), CETP (P = 9.44 × 10-108), LIPC (P = 9.42 × 10-51), and LPL (P = 5.79 × 10-22), with 92% of signals replicating in the UK Biobank cohort. eQTL colocalization linked seven lead variants to tissue-specific gene expression and prioritized candidate regulatory genes. Together, these findings characterize shared lipid-associated signals with heterogeneous, trait-dependent effects in continental African populations and provide population-relevant evidence for candidate regulatory mechanisms.