Common Genetic Variation in Cell Cycle Regulatory Genes and Etiology of Intrinsic Breast Cancer Subtype: A Candidate Gene Approach

Taylor, Nicholas Jay

Download PDF

Request Version for Screen Reader

Last Modified

March 22, 2019

Creator

Taylor, Nicholas Jay
- Affiliation: Gillings School of Global Public Health, Department of Epidemiology

Abstract

A large proportion of unexplained risk for breast cancer remains to be accounted for. Contributing factors may be environmental, genetic, or a combination of both and there is considerable debate about which of these factors is most important. However, the scope and magnitude of the genetic contribution to the causation of breast cancer remains unclear. Genetic risk factors for breast cancer remain to be discovered, and with heterogeneity of breast cancer being characterized into intrinsic molecular subtypes, the difficulty in identifying these risk factors is diminishing. This dissertation took a candidate gene approach based on factors involved in cell cycle regulation (AURKA, BRCA1, BARD1, BRIP1, and ZNF350) to identify single nucleotide polymorphisms (SNPs) associated with overall rate of breast cancer and intrinsic breast cancer subtype in the Carolina Breast Cancer Study (CBCS). A total of 65 SNPs on five genes were genotyped in 1,946 cases and 1,747 controls in African American and Caucasian participants of the CBCS. In addition, 144 ancestry informative markers were also genotyped in these individuals to estimate individual ancestry and adjust logistic models for potential population stratification. Race-stratified odds ratios were calculated, as estimates of rate ratios, along with 95% confidence intervals for the associations between SNP genotypes and breast cancer using logistic regression and adjusting for age and ancestry. These associations were also estimated by intrinsic subtype of breast cancer in a similarly adjusted combined race group. The intronic SNP rs6092309 on AURKA showed an inverse association with overall rate of breast cancer among African Americans (OR=0.69, 95%CI=0.53-0.90), with inverse associations also noted across all strata of intrinsic subtype. Exploratory race-stratified, subtype-specific analyses for some AURKA SNPs suggested race-specific effects. Three SNPs in high LD with one another on BRCA1 (rs16941, rs16942, and rs1799966) had positive associations with overall rate of breast cancer among Caucasians. One SNP on BARD1 (rs28997576: OR=1.42, 95%CI: 1.00-2.03) also showed a positive association with overall rate of breast cancer among Caucasians. These results suggest that associations between genetic exposures and rate of breast cancer may differ by intrinsic subtype and possibly by race within subtype. Replication of these findings in larger populations of African American and Caucasian women will be required to make more accurate interpretations.

Date of publication