An integrative analysis of consortium-based multi-omics QTL and genome-wide association study data uncovers new biomarkers for lung cancer

The role of molecular traits (e.g., gene expression and protein abundance) in the occurrence, development, and prognosis of lung cancer has been extensively studied. However, biomarkers in other molecular layers and connections among various molecular traits that influence lung cancer risk remain largely underexplored. We conducted the first comprehensive assessment of the associations between molecular biomarkers (i.e., DNA methylation, gene expression, protein and metabolite) and lung cancer risk through epigenome-wide association study (EWAS), transcriptome-wide association study (TWAS), proteome-wide association study (PWAS) and metabolome-wide association study (MWAS), and then we synthesized all omics layers to reveal potential regulatory mechanisms across layers. Our analysis identified 61 CpG sites, 62 genes, 6 proteins, and 5 metabolites, yielding 123 novel biomarkers. These biomarkers highlighted 90 relevant genes for lung cancer, 83 among them were first established in our study. Multi-omics integrative analysis revealed 12 of these genes overlapped across omics layers, suggesting cross-omics interactions. Moreover, we identified 106 potential cross-layer regulatory pathways, indicating that cell proliferation, differentiation, immunity, and protein-catalyzed metabolite reaction interact to influence lung cancer risk. Further subgroup analyses revealed that biomarker distributions differ across patient subgroups. To share all signals in different omics layers with community, we released a free online platform, LungCancer-xWAS, which can be accessed at http://bigdata.njmu.edu.cn/LungCancer-xWAS/. Our findings underscore the importance of xWAS which integrating various types of molecular quantitative trait loci (xQTL) data with genome-wide association study (GWAS) data to deepen understanding of lung cancer pathophysiology, which may provide valuable insights into potential therapeutic targets for the disease.