Multivariate GWAS reveals shared genetic etiology and pleiotropic loci across carcinomas

Carcinomas, which arise from epithelial tissues and account for more than 90% of cancers, share molecular programs while exhibiting site-specific biology. However, the genetic partitioning of common versus distinct components remains unclear. We harmonized genome-wide association study (GWAS) summary statistics for 429,158 European-ancestry cases across nine common carcinoma types, and triangulated evidence at the genome-wide, regional, and locus levels to delineate shared and cancer-specific risk. We show that cross-carcinoma overlap is likely systematically underestimated, because loci within the same genomic regions can have discordant effects. To address this, we constructed a cross-carcinoma hierarchical latent-factor model, performed follow-up multivariate GWAS to identify novel pleiotropic loci, and subsequently integrated multi-omics data to prioritize effector genes. This framework partitions general and cancer-specific genetic liability, revealing pleiotropy obscured by conventional analyses. Subsequent multi-omics gene prioritization implicated convergent epithelial growth and differentiation programs, nominating tractable targets for biomarker development, prevention, and mechanism-informed therapies.