GWAS and 3D chromatin mapping identifies multicancer risk genes associated with hormone-dependent cancers

Hormone-dependent cancers (HDCs) share several risk factors, suggesting a common aetiology. Using data from genome-wide association studies, we showed spatial clustering of risk variants across four HDCs (breast, endometrial, ovarian and prostate cancers), contrasting with genetically uncorrelated traits. We identified 44 multi-HDC risk regions across the genome, defined as overlapping risk regions for at least two HDCs: two regions contained risk variants for all four HDCs, 13 for three HDCs and 28 for two HDCs. Integrating GWAS data, epigenomic profiling and high-resolution promoter capture HiC maps from diverse cell line models, we annotated 53 candidate risk genes at 22 multi-HDC risk regions. These targets were enriched for established genes from the COSMIC Cancer Gene Census, but many had no previously reported pleiotropic roles. Additionally, we pinpointed lncRNAs as potential HDC targets and identified risk alleles in several regions that altered transcription factors motifs, suggesting regulatory mechanisms. Known drug targets were over-represented among the candidate multi-HDC risk genes, implying that some may serve as targets for therapeutic development or facilitate the repurposing of existing treatments for HDC. Our comprehensive approach provides a framework for identifying common target genes driving complex traits and enhances understanding of HDC susceptibility. AUTHOR SUMMARYWhile hormone-dependent cancers (HDCs) share several risk factors, our understanding of the complex genetic interactions contributing to their development is limited. In this study, we leveraged large-scale genetic studies of cancer risk, high-throughput sequencing methods and computational analyses to identify genes associated with four HDCs: breast, endometrial, ovarian and prostate cancers. We identified known cancer genes and discovered many that were not previously linked to cancer. These findings are significant because identifying genes associated with risk of multiple cancer types can enhance the gene mapping accuracy and highlight new therapeutic targets.