An analysis of measures of effect size by age of onset in cancer genomewide association studies

Abstract

Many of the genetic variants identified via genome-wide association studies (GWAS) appear to have larger parameter estimates for younger onset cancers compared with adult onset cancers. We used data from the National Human Genome Research Institute (NHRGI) Catalog of Published GWAS to test the hypothesis that the magnitude of the parameter estimates is larger in younger onset compared to adult onset cancers. We found that the odds ratios in individuals categorized as childhood or young adult cancers were significantly higher than the odds ratios in individuals with adult cancers. The presence of larger effect sizes may mean that the variants associated with younger onset cancers explain a greater proportion of the population attributable risk estimates than the SNPs associated with adult onset cancers, which could improve early detection, treatment, and/or prevention.