Germline variants affect lung-to-tumor transcription dynamics and NSCLC outcomes

Abstract

The influence of germline genetics on the molecular transition from lung to tumor and different transcriptional patterns of tumors remains underexplored. Additionally, little is known about the genetic contributions to lung cancer relapse and mortality following surgery. The transcriptional transition from non-affected lung tissue to tumor was investigated using an extended eQTL approach. Associations between variants and expression differences in tumors relative to matched surrounding lung tissue were examined in 515 surgical cases of lung adenocarcinoma. We also used Cox regression in a genome-wide association study to identify variants associated with survival ( n = 3472) and relapse ( n = 3369) in patients with NSCLC. The number of significant eQTL was approximately threefold lower in tumor than in lung tissue. Variants that showed differential cis-regulatory control on expression between lung and tumor tissues were more frequently located in bivalent or distal regulatory elements. Those variants altered the transcription factor binding landscape and the expression of genes that are differentially expressed in tumors. Co-expression modules derived from WGCNA identified module-QTLs and putative regulatory signaling genes (CDK15, CD79B, CCDC80, and SCARA5) involved in immunity and mesenchymal cell fate. Differential eQTLs for three genes (NRG1, SECISBP2L, and JAML) colocalized with GWAS-nominated lung cancer risk loci. Prognostic lung cancer loci identified 6p22.3-MBOAT1 and 16p13.3-RBFOX1. The lung-to-tumor transcriptomic transition is characterized by an attenuated contribution of germline variants. Differential eQTLs inform differences in gene expression between lung and tumor tissues, implicate putative cancer-related genes and pathways, and identify target genes underlying GWAS loci.