Genome-wide prioritization reveals novel gene signatures associated with cardiotoxic effects of tyrosine kinase inhibitors

Tyrosine kinase inhibitors (TKIs) are characterized as multi-targeted anticancer agents that lack specificity, leading to cardiovascular adverse effects. To date, there are no reliable means to predict the cardiotoxicity of TKIs under development. The present study assessed the usual variants of genes to determine the molecular targets of TKIs associated with heart failure (HF). Gene or gene products affected by TKIs were assessed using the Drug Gene Interaction Database. These genes were investigated in genome-wide association studies (GWAS) datasets associated with HF at a genome-wide significant level (P<1×10(−5)). Subsequently, single-nucleotide polymorphisms (SNPs) that reached the established GWAS threshold (P<5×10(−8)) were investigated for genome-wide significance. Based on a threshold score of 3, nine gene loci yielded associations according to their biological function using RegulomeDB. Finally, comprehensive functional analysis of SNPs was performed using bioinformatics databases to identify potential drug targets. Using rSNPBase, rs7115242, rs143160639 and rs870064 were found to interfere with proximal transcription regulation, while rs7115242, rs143160639 and rs117153772 were involved in distal regulation, and most SNPs participated in post-transcriptional RNA binding protein-mediated regulation. rs191188930 on platelet-derived growth factor receptor (PDGFR) α was associated with numerous TKI drugs, including sunitinib, pazopanib, sorafenib, dasatinib and nilotinib. Using RegulomeDB and HaploReg v4.1, rs191188930 was predicted to be located in enhancer histone markers. PhenoScanner GWAS analysis revealed that rs191188930 was associated with other diseases or phenotypes, in addition to HF. Genotype-Tissue Expression analysis indicated that the PDGFRα gene had the highest median expression in ‘Cells-Transformed fibroblasts’, and the Search Tool for the Retrieval of Interacting Genes/Proteins revealed the protein-protein interaction network of PDGFRα. The present findings demonstrated the overlap of TKI-induced genes and those mediating HF risk, suggesting molecular mechanisms potentially responsible for TKI-induced HF risk. Additionally, the present genetic study may be helpful to further investigate off-target drug effects.