Clustering of place of birth for women with breast cancer: differences by tumor characteristics

Abstract

There is increasing evidence that exposures in early life affect breast cancer risk, and that breast cancer etiology differs by tumor subtype. If environmental exposures in early life contribute to risk, it is expected that there would be clustering of women with breast cancer by their place of birth, and that clustering might differ by subtype. We examined spatial associations between place of birth and breast cancer by subtype, using hormone receptor status and molecular profiles of breast tumors. Data were drawn from the Western New York Exposures and Breast Cancer study, a population-based case-control study of incident, pathologically confirmed breast cancer (1996-2001) in Erie and Niagara Counties. Included were women born in the study area (579 cases and 931 controls). Clustering of breast cancer subgroups relative to controls was examined by the k-function method in groups stratified by estrogen receptor (ER), progesterone receptor (PR), and HER2 status, and by DNA methylation status and p53 mutation status, and the k-function difference was used to compare relative spatial aggregation and spatial range of the difference between case subgroups and controls. We found a tendency to cluster among ER positive, PR positive, and HER2 negative cases (i.e., luminal A subtype), especially among premenopausal women, but not among the other groups defined by hormonal receptor status, or by either methylation or p53 mutation status. While our findings cannot rule out clustering of cases by birth place because of shared behaviors related to residence location, they also suggest that early life environmental exposures may affect subsequent breast cancer risk, and that premenopausal breast tumors of the luminal A subtype may be more affected by these early life exposures than other subtypes.