Skip to navigation Skip to main content Skip to footer

Approved research

Evaluating genome-wide association and gene-environment interaction in colorectal cancer

Principal Investigator: Dr Ulrike Peters
Approved Research ID: 8614
Approval date: November 1st 2016

Lay summary

1)Combine data from existing and ongoing genomewide association studies (GWAS) of colorectal cancer (CRC). We will combine data from UK Biobank with existing GWAS to identify rare and common genetic variation associated with risk and survival. 2)Investigate whether associations with genetic variants are modified by environmental predictors of CRC. These include demographic factors, obesity, exercise, non-steroidal anti-inflammatory drugs, postmenopausal hormones, alcohol, smoking, dietary factors, and exposures as measured by circulating biomarkers of nutrients, metabolism, inflammation, and sex hormones. Secondarily, we?ll conduct gene-gene, functional annotation, candidate gene, pathway, Mendelian randomization, mediation, and risk modeling analyses. Genetic and environmental factors contribute to the etiology of colorectal cancer (CRC). Genomewide association studies (GWAS) have identified novel loci contributing to CRC susceptibility and survival outcomes, and we expect that additional loci can be identified by combining data across large sample sets, as proposed here. To evaluate the impact of genetic variants on CRC risk and survival, it is important to explore the interplay between genetic and environmental factors in disease, and to translate these findings into public health applications, it is critical to understand the penetrance of new loci and develop risk models that include these risk factors. To comprehensively explore genetic and environmental contributions to CRC risk and survival, we propose to combine UK Biobank data with existing and ongoing CRC studies that are part of the Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO). These combined studies will allow for considerable statistical power to identify and characterize genetic variants associated with CRC and enable analyses of gene-environment interaction. In addition, this large sample size will facilitate exploratory analyses of gene-gene interaction, candidate genes, pathways, Mendelian randomization, mediation, as well as risk modeling of genetic variants identified by genome-wide association studies. To maximize statistical power to detect associations and interactions, data from the entire UK Biobank cohort, including approximately 1,163 incident and 2,271 prevalent cases of CRC (these are the current number of cases in the UK Biobank and we will include any newly diagnosed CRC cases as they occur), will be combined with that from GECCO, which includes over 35,000 CRC cases and 37,000 controls with directly sequenced/genotyped and imputed genome-wide genetic data.