Metabolism Open
Cross-trait genetic architecture of intracerebral hemorrhage highlights convergent vascular, hematologic and lipid pathways
By Zongheng Zhang, Aierpati Maimaiti, Jing Li and 3 others.
Abstract
Intracerebral hemorrhage (ICH) is a severe stroke subtype, but the extent to which its susceptibility architecture is shared with other complex traits remains unclear. Here we integrated FinnGen release 12 ICH summary statistics (5,112 cases and 450,016 controls) with 898 GWAS datasets curated through BIGA and the GWAS Catalog, and applied a layered analytical framework spanning linkage disequilibrium score regression, cross-trait association testing, locus annotation, Bayesian colocalization, multi-method gene prioritization, and two-sample Mendelian randomization. Among 46 nominal LDSC correlations (P < 0.05), only stroke and one atrial-fibrillation dataset remained significant after both Benjamini-Hochberg false-discovery-rate and Bonferroni correction; the hematologic and fatty-acid/lipid correlations were therefore treated as exploratory. Joint analyses yielded 34 MTAG-positive and 45 CPASSOC-positive trait pairs, whereas locus-level refinement highlighted recurrent regions at 1q22, 2q33, 5p15.33, 13q34, and 15q, with 15 regions showing evidence of colocalization (PP.H4 > 0.7). Integrative prioritization across SMR, FUSION, mBAT-combo, and PoPS nominated 117 candidate genes, including ten genes supported by all four approaches, with EHD4, ICA1L, NBEAL1, FAM117B, SH3PXD2A, CARF, SRR, HIC1, TLR6 , and FAM114A1 forming the highest-confidence set. Bidirectional Mendelian randomization identified seven FDR-supported trait-to-ICH associations involving atrial fibrillation, small-vessel stroke, erythrocyte indices, fatty-acid traits, and a clinical-treatment phenotype. No ICH-to-trait association survived multiple-testing correction, and the variable sensitivity support indicated that these findings should be interpreted as exploratory directional evidence rather than established causal effects. Together, these analyses position ICH within a broader cross-trait genetic network centered on cerebral small-vessel vulnerability, blood-cell homeostasis, and lipid composition, and provide a prioritized framework for downstream mechanistic and functional studies.



