Interaction analysis between germline susceptibility loci and somatic alterations in lung cancer

Emerging evidence indicates that germline variations may interact with somatic events in carcinogenesis. However, the germline-somatic interaction in lung cancer remains largely unknown. We investigated whether lung cancer driver genes (CDGs) were more likely to locate within cancer susceptibility regions. Pathway analysis was performed to identify common pathways underlying CDGs and cancer susceptibility genes (CSGs). Next, we analyzed the associations between lung cancer risk SNPs and somatic alterations, including mutations and copy number alterations, in the level of genes, pathways, and overall burden of alterations. Enrichment analysis showed that lung CDGs are more likely to locate within cancer susceptibility regions (p = 8.40 x 10(-3)). Both of lung CSGs and CDGs showed significant enrichment in pathways such as cell cycle and p53 signaling pathway. Gene-based analysis showed that rs36600 (22q12.2) was associated with somatic mutations within ARID1A (OR = 2.45, 95%CI: 1.47-4.08, p = 5.78 x 10(-4)). Pathway-based analysis of somatic truncation mutations identified rs2395185 and rs3817963 at 6p22.1 was associated with cell cycle pathway (OR = 1.56, p = 3.61 x 10(-4) for rs2395185; OR = 1.58, p = 4.15 x 10(-4) for rs3817963), and rs3817963 was also associated with MAPK signaling pathway (OR = 1.54, p = 8.58 x 10(-4)). Further analysis associated rs2395185 at 6p22.1 (HLA class II genes) with increased APOBEC3A expression (p = 9.50 x 10(-3)) and elevated APOBEC mutagenesis (p = 3.58 x 10(-3)). These results indicate germline-somatic interactions in lung tumorigenesis, and help to uncover the molecular mechanisms underlying lung cancer risk SNPs.