Abstract 4730: Integrating pathway analysis and genetics of gene expression for genome-wide association study on basal cell carcinoma

Abstract

Abstract Genome-wide association studies (GWASs) have primarily focused on marginal effects for individual markers and have only incorporated external functional information after identifying robust statistical associations. A pathway-based approach using the classification of genes into functionally-related pathways may identify pathways enriched for associations with diseases. However, there are some limitations of assigning all the single nucleotide polymorphisms (SNPs) into the pathways by their physical locations, as most SNPs in a gene region will not represent functional variants of the gene and a gene may be regulated in trans by DNA variants that are far away from the structural gene. Thus, we integrated genetics of gene expression and the pathway analysis for the GWAS on basal cell carcinoma (BCC), and aimed to discover potential important biological pathways involved in the etiology of BCC. Specifically, we first conducted the GWAS of global gene expression in lymphoblastoid cell lines among 1355 individuals of British descent and identified 322,324 expression-associated SNPs (eSNPs), and then evaluated the association between these functionally annotated SNPs and BCC risk among 2,045 BCC cases and 6,013 controls in Caucasians. The associations of all the genes with BCC were represented by the eSNPs associated with their expression accordingly, and all the genes were assigned into 99 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways for the pathway enrichment analysis by using Kolmogorov-Smirnov statistic. We performed 1,000 permutations to access the significance of the pathway enrichment by shuffling the case-control status and repeating the association analysis. As a result, a total of 10 (10%) pathways out of the 99 KEGG pathways reached a nominal p-value ≤ 0.05, twice as high as the number expected by chance (0.05 × 99 = 4.95). After the correction for multiple test by the false discovery rate, three pathways were significantly associated with BCC risk, which were the autoimmune thyroid disease pathway (p-value < 0.001), and JAK-STAT signaling pathway (p-value = 0.02), and B cell receptor signaling pathway (p-value = 0.05). Of note, most of the genes enriched in the autoimmune thyroid disease pathway were the components of HLA-complex, which has been known to play an important role in tumor immunity. There is also some evidence of biological functions for the other two pathways on cancer immunity or oncogenesis from the reported studies. In conclusion, our approach which integrates the pathway information and gene expression-annotated SNP data may help identify biological pathways in the etiology of BCC, which may be missed in traditional GWAS approach focusing on marginal effects of individual markers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4730. doi:10.1158/1538-7445.AM2011-4730