Genome-wide Association Studies Meet Chemoprevention

Abstract

The T variant at rs6983267 on 8q24 is a weakly protective allele for colorectal cancer discovered in the conduct of GWASs (2–4). This locus resides in an apparent enhancer domain for MYC (5), and deletion of the region in mice has been associated with resistance to APCmin-induced intestinal tumorigenesis (6) by decreasing TCF7L2 interaction with CTNNB1 (TCF4/β-catenin)–mediated transactivation of the MYC oncogene (5). Separately, the protective effect of aspirin for colorectal cancer is, in part, mediated by the ability of aspirin to block prostaglandin E2/EP2 receptor–mediated stabilization of CTNNB1(7,8). Based on the consistent effects of aspirin alone and the rs6983267 variant alone, Nan et al. (1) tested the joint effect of the rs6983267 genotype and aspirin use on the risk of colorectal cancer in a nested case–control study, pooling subjects from the Nurses’ Health Study (n = 472 case patients; n = 1013 control subjects) and the Health Professionals Follow-up Study (n = 368 case patients; n = 1686 control subjects). Carriage of the low-risk T variant among regular users of aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs) confers half the risk of colorectal cancer when compared with nonaspirin users of the same genotype. Although increased dose and duration of aspirin use appear to enhance the protective effect of NSAIDs in TT/TG carriers, the protective effect of aspirin in these subjects is present even at modest doses. However, there is no overall protective effect of regular aspirin use on colorectal cancer among GG carriers, and longer duration of aspirin use was associated with a non-statistically significant decrease in risk, suggesting no benefit of aspirin in the non-T genetic background. Important to this work is the functional demonstration by the authors that TCF7L2 demonstrates lower binding to the T allele 6983267 relative to the G allele in cell line studies in the presence of aspirin and does so in a dose-dependent manner. The findings from the binding studies strengthen the observed differential effect of aspirin by genotype in the human studies, favoring the hypothesis that the benefit of aspirin, and other NSAIDs, is specific to a subset of the population. This study adds to the emerging evidence that genetic variation at 8q24 is functionally relevant as a risk variant. This study also strengthens the emerging hypothesis that the antitumorigenic effect of aspirin and other NSAIDs is partly mediated by a disrupting effect on the transactivation of MYC. This is shown in this study as the benefit of aspirin limited to the T carriers (responsive genotype), for which the lower risk of colorectal cancer in this background is largely limited to tumors driven by nuclear CTTNB1 that overexpress the MYC oncogene. Collectively, the work of Nan et al. (1) provides compelling evidence that aspirin interacts with the genetic background at rs6983267 to influence the risk of colorectal cancer and that the effect is mechanistically coupled to the ability of the region to bind the CTTNB1/TCF7L2 complex and transactivate the MYC oncogene. In the GG background, regular aspirin appears less efficacious, supporting evidence that those with the GG genotype have a more constitutively active enhancer function for MYC expression. Importantly, the non-statistically significant reduction in risk with long-duration aspirin use in this background suggests that higher dosing or long use is necessary to achieve benefit in the GG background. These questions can and should be addressed quickly in randomized controlled trials of NSAIDs.