Genome-wide analysis of 3′ untranslated region alternative polyadenylation quantitative trait loci identified a potential novel susceptibility locus for lung cancer in cross-ancestry populations

Abstract

Genome wide association studies (GWASs) have revealed several lung cancer susceptibility loci; however, we still face key issues such as how to identify more ‘high-frequency and inefficient’ variants and decipher the causal variants. Alternative polyadenylation (APA) can shorten or prolong the 3′UTR of mRNA containing cis regulatory elements, which plays an important role in post transcriptional regulation. Specific variants can affect the 3′UTR APA, leading to differences in the risk of lung cancer among individuals carrying different alleles. In this study, a cross-ancestry two-stage case-control design was used to evaluate the associations of 3′UTR APA related variants (3′aQTL SNPs, 3′aSNPs) defined by GTEx in normal lung tissues with the risk of lung cancer based on the genome-wide meta-analysis (GWMA) from FinnGen, UK Biobank and VA Million Veteran Program (MVP), including 28,557 cases and 1,355,961 controls. The promising variants were validated in an independent Chinese population with 1169 lung cancer cases and 1046 controls. Functional annotations of the identified 3′aSNP and related genes were performed based on multiple public databases. Finally, we identified a potential 3′aSNP rs11583258 associated with the risk of lung cancer both in the GWMA [OR = 1.04 (1.02–1.06), P = 1.00 × 10–4] and in the validation stage [OR = 1.08 (1.01–1.16), P = 1.01 × 10–2] at 1p36.11. Function annotation integrating the results of multiple public datasets suggested the variants in this region might affect both the length of the 3′UTR of the UBXN11 transcripts and the expression of UBXN11 to affect the susceptibility of lung cancer.