A cross-tissue transcriptome-wide association study identifies novel susceptibility genes for lung cancer in Chinese populations

Although dozens of susceptibility loci have been identified for lung cancer in genome-wide association studies (GWASs), the susceptibility genes and underlying mechanisms remain unclear. In this study, we conducted a cross-tissue transcriptome-wide association study (TWAS) with UTMOST based on summary statistics from 13 327 lung cancer cases and 13 328 controls and the genetic-expression matrix over 44 human tissues in the Genotype-Tissue Expression (GTEx) project. After further evaluating the associations in each tissue, we revealed 6 susceptibility genes in known loci and identified 12 novel ones. Among those, five novel genes, including DCAF16 (Pcross-tissue = 2.57 x 10-5, P-Lung = 2.89 x 10(-5)), CBL (Pcross-tissue = 5.08 x 10(-7), P-Lung = 1.82 x 10(-4)), ATR (Pcross-tissue = 1.45 x 10(-5), P-Lung = 9.68 x 10(-5)), GYPE (Pcross-tissue = 1.45 x 10(-5), P-Lung = 2.17 x 10(-3)) and PARD3 (Pcross-tissue = 5.79 x 10(-6), P-Lung = 4.05 x 10(-3)), were significantly associated with the risk of lung cancer in both cross-tissue and lung tissue models. Further colocalization analysis indicated that rs7667864 (C > A) and rs2298650 (G > T) drove the GWAS association signals at 4p15.31-32 (OR = 1.09, 95%CI: 1.04-1.12, P-GWAS = 5.54 x 10(-5)) and 11q23.3 (OR = 1.08, 95%CI: 1.04-1.13, P-GWAS = 5.55 x 10(-5)), as well as the expression of DCAF16 (beta(GTEx) = 0.24, P-GTEx = 9.81 x 10(-15); beta(NJLCC) = 0.29, P-NJLCC = 3.84 x 10(-8)) and CBL (beta(GTEx) = -0.17, P-GTEx = 2.82 x 10(-8); beta(NJLCC) = -0.32, P-NJLCC = 2.61 x 10(-7)) in lung tissue. Functional annotations and phenotype assays supported the carcinogenic effect of these novel susceptibility genes in lung carcinogenesis.

Automated summary

This study conducted a cross-tissue transcriptome-wide association study using UTMOST, identifying 6 known susceptibility genes and 12 novel ones for lung cancer susceptibility. Among these, 5 novel genes were significantly associated with lung cancer risk in both cross-tissue and lung tissue models. Further analysis indicated that specific genetic variants drove the GWAS association signals and influenced the expression of certain genes in lung tissue, supporting their role in lung carcinogenesis.