|Grant Number:||5R01CA104021-05 Interpret this number|
|Primary Investigator:||Hall, Per|
|Project Title:||Genetic Determinants of Postmenopausal Breast Cancer|
DESCRIPTION (provided by applicant): A central concern regarding the widespread use of menopausal hormones is the increased breast cancer risk as recently described in several large studies. Since many women nevertheless have considerable benefit from this therapy, mainly alleviation of menopausal symptoms, it would be of outmost value if we were able to identify those women in whom the use of menopausal hormones should be discouraged due to an unacceptably increased breast cancer risk. Since this treatment could be seen as a representation of any hormonal exposures, it can serve as a model for how other hormonal factors, such as age at menopause and BMI, interact with breast cancer susceptibility genes. We will study genetic variation in pathways that regulate estrogen exposure (estrogen metabolism pathway) and regulate response to estrogen exposure (receptors and associated proteins, transcription cofactors and responsive genes) influence the risk of breast cancer. Further, we want to investigate how the influence of this genetic variation is affected by intake of exogenous hormones. The influence of estrogen in the breast tissue is mediated via one of its receptors, estrogen receptor a (ER) and transcription induced by the ER is dependent on co-regulators. We have experimentally, using T47D cells and expression arrays, identified estrogen responsive genes and genes that are associated with ER status in primary breast cancers. We have also constructed a database of gene expression data that integrates all six publicly available and some unpublished breast cancer microarray datasets. The purpose of the database is to facilitate the mining of genes with robust associations, with clinicopathological variables in human breast cancer. We propose a systematic approach for gene association study that first employs bioinformatics and genomic approaches to expand the number of likely candidate genes from several to hundreds. In addition, by using haplotype-tagging SNPs (ht-SNPs), we reduce the genotyping load by approximately 75%, while retaining thorough characterization of each gene under evaluation.