Identification of breast cancer susceptibility genes via trans-ethnic Mendelian randomization and colocalization analyses

Breast cancer remains a major global health burden, with persistent challenges including recurrence, metastasis, and drug resistance limiting the efficacy of current treatments. The identification of novel therapeutic targets is essential for advancing precision oncology. We employed a 2-sample Mendelian randomization (MR) framework utilizing large-scale eQTL and genome-wide association study (GWAS) datasets from European cohorts to identify genetic targets for breast cancer. Colocalization analysis, phenome-wide association studies (PheWAS), protein-protein interaction networks, and functional enrichment analyses (GO/KEGG) were conducted. Single-cell gene expression was also analyzed. Candidate drugs were predicted using pharmacogenomic databases and subsequently validated through molecular docking simulations. MR analysis identified 480 candidate targets, among which 51 were successfully validated. Colocalization analysis highlighted 7 genes - DNPH1, SYT11, RCCD1, LAMB2, SLC22A5, CBX6, and FAAH - with strong evidence of causal association. Functional annotation and protein-protein interaction (PPI) network analysis revealed their involvement in key cancer-related pathways. Their expression patterns were also analyzed at the single-cell level. Molecular docking confirmed stable binding affinities between the identified target proteins and their predicted drug candidates. This integrative genomics and bioinformatics approach identified 7 promising drug targets for breast cancer. These findings offer novel avenues for the development of targeted therapies and underscore the importance of genetic epidemiology in guiding drug discovery.