Traditional GWAS focus primarily on the most significant genetic markers, often without sufficient power to detect relatively small effects conferred by most genetic variants. Moreover, the vast majority of variants identified by GWAS are common proxy SNPs, which have no direct biological relevance to disease. In addition, there is growing evidence that genes do not work in isolation. Instead, complex molecular networks and cellular pathways are often involved in disease susceptibility progression. Therefore, pathway analysis, which jointly considers multiple variants with moderate signals in related genes, may help evaluate the cumulative contributions of genes within particular biological pathways and identify pathways relevant to the etiology of disease. Pathway analysis approaches have been successfully applied to various complex diseases, including basal cell carcinoma (BCC) study conducted by our group. In addition, traditional pathway analyses simply assign SNPs to nearby genes based on their physical locations. However, most such SNPs do not represent functional variants of that gene, and multiple testing on these large numbers of SNPs may introduce many false positive findings. Moreover, some SNPs that are located in a structural gene but regulate the expression of another gene would be annotated inappropriately. Therefore, defining the expression quantitative trait loci (eQTLs) and assigning these expression-related SNPs (eSNPs) to the genes they regulate may help functionally annotate SNPs and define the cumulative contribution of particular genes in a better way. Our group integrated eQTL information into the BCC GWAS for pathway analysis and successfully identified several novel disease-related pathways. Here we propose to apply this method to squamous cell carcinoma (SCC) of the skin to systematically assess the associations of biological pathways with SCC risk. We will use gene expression data from the MuTHER project and the GTEx Portal to integrate the eQTL information on skin. Then we plan to take advantage of the previous SCC GWAS by our group in the discovery stage and utilize a SCC case-control study with GWAS data from the Kaiser Permanente Northern California health care system for replication. We propose the following specific aims: (1) Use two pathway-based approaches to evaluate the associations of biological pathways with SCC risk by estimating the effects of skin eSNPs within each pathway using our previous SCC GWAS data of 2,710 cases and 35,637 controls from the Nurses? Health Study and Health Professionals Follow-up Study. (2) Validate the associations of potential disease-related loci within the identified pathways in an external replication study with 6,891 SCC cases and 54,566 controls from the Kaiser Permanente Northern California health care system. This will be the first well-positioned post-GWAS study integrating GWAS on SCC and GWAS on gene expression in skin into a pathway analysis. The current study will utilize GWAS data on SNPs as well as gene expression to a greater extent and help uncover the potential relations between biological pathways and SCC development.
To our knowledge, this would be the first study to uniquely identify cSCC susceptibility pathways using large available GWAS data. This innovative work may identify genes/pathways that might be missed using the conventional GWAS approach, and provide new insights into the etiology of cSCC, which may lead to better strategies on cSCC prevention and treatment. Our pathway analysis approach can pinpoint clinical potential and actionable targets for treatment and identify the genetically high-susceptibility subgroups for intervention and early detection.
