Identification of Lung eQTLs in GWAS-nominated Lung Cancer Susceptibility Loci

Lavoie-Charland E¹, Postma DS², Sin DD³, Hao K⁴, Nickle DC⁵, Timens W², Paré PD³, Laviolette M¹, and Bossé Y^1,6

1. Institut universitaire de cardiologie et de pneumologie de Québec, Quebec City, Canada; 2. University of Groningen, University Medical Center Groningen, GRIAC Research Institute, Groningen, The Netherlands; 3. The University of British Columbia Centre for Heart Lung Innovation, St Paul’s Hospital, Vancouver, Canada; 4. Icahn School of Medicine at Mount Sinai, New York, NY, United States; 5. Merck & Co, MRL, Seattle, Washington, United States; 6. Department of Molecular Medicine, Laval University, Quebec City, Canada

Lung cancer is the leading cause of cancer-related deaths worldwide in both men and women. Genome wide association studies (GWAS) have been successful to identify inherited genetic variants associated with lung cancer. The goal of this study is to use a complementary genomic approach to identify causal genes/variants underpinning GWAS-nominated loci.

GWAS on lung cancer published between 2008 and 2014 were reviewed and SNPs associated with lung cancer were tabulated. Lung cancer susceptibility loci were defined as ± 50 Kb from the disease-associated SNPs and/or suspected causal genes. Lung cancer associated-SNPs and their proxies (r² > 0.7) were selected for expression quantitative trait loci (eQTL) analysis in a discovery cohort of 409 individuals with genome-wide genotyping (Illumina Human1M-Duo) and gene expression profiling (Affymetrix) in the lung. Results were replicated in two validation sets (n = 363 and 339).

Twenty lung cancer susceptibility loci were abstracted from the literature. Significant lung eQTLs were identified and replicated for five loci: 13q31, 3q28, 12p13, 6q22, 5q31. On 13q31, the risk allele for rs2352028 was associated with lowered gene expression levels of GPC5 (p = 8.21x10^-12), confirming previous results. On 3q28 and 12p13, lung eQTL results support the contribution of suspected candidate genes TP63 and RAD52 implying low expression from the risk alleles for rs10849605 (p = 6.95x10^-23) and rs10937405 (p = 1.84x10^-8), respectively. Finally, the lung cancer-associated SNPs rs9387478 on 6q22 and rs247008 on 5q31 were associated with the mRNA expression levels of DCBLD1 (p = 7.99x10^-14) and SLC22A5 (p = 2.12x10^-4), respectively.

This study identified the most likely causal genes in a number of lung cancer susceptibility loci (DCBLD1 on 6q22 and SLC22A5 on 5q31), confirmed the contribution of suspected genes (TP63 on 3q28 and RAD52 on 12p13), and validated preliminary evidence (GPC5 on 13q31) suggesting that lung cancer risk is mediated through gene regulation in the lung.