Genome-wide association studies (GWAS) have detected several susceptibility variants for urinary bladder cancer, but how gene regulation affects disease development remains unclear. To extend GWAS findings, we conducted a transcriptome-wide association study (TWAS) using PrediXcan to predict gene expression levels in whole blood using genome-wide genotype data for 6180 bladder cancer cases and 5699 controls included in the database of Genotypes and Phenotypes (dbGaP). Logistic regression was used to estimate adjusted gene-level odds ratios (OR) per 1-standard deviation higher expression with 95% confidence intervals (CI) for bladder cancer risk. We further assessed associations for individual single-nucleotide polymorphisms (SNPs) used to predict expression levels and proximal loci for genes identified in gene-level analyses with false-discovery rate (FDR) correction. TWAS identified four genes for which expression levels were associated with bladder cancer risk: SLC39A3 (OR = 0.91, CI = 0.87-0.95, FDR = 0.015), ZNF737 (OR = 0.91, CI = 0.88-0.95, FDR = 0.016), FAM53A (OR = 1.09, CI = 1.05-1.14, FDR = 0.022), and PPP1R2 (OR = 1.09, CI = 1.05-1.13, FDR = 0.049). Findings from this TWAS enhance our understanding of how genetically-regulated gene expression affects bladder cancer development and point to potential prevention and treatment targets.
Keywords: Bladder Cancer; Case-control; GWAS; TWAS.
© 2025. The Author(s).