Accurate prediction of risk and personalized approaches to prevention of breast cancer or its recurrence are needed to reduce both cost of treatment and morbidity.
We have developed gene expression signatures that distinguish premalignant breast lesions in women providing biomarkers to improve diagnosis and identify therapeutic targets. Primary explant cultures of human breast tissues retain estrogen and progesterone receptors and are used to determine the spectrum and variation in responses to estrogenic compounds and drugs. Conditionally immortalized primary breast cell cultures provide tools for screening responses to therapies and genetic engineering of cells. Our work in primary human tissues is paralleled by transplantable models of premalignancy and tumors in immunocompetent mice. These are suitable for preclinical studies of cancer immunotherapies.
Accurate diagnostic tools for premalignancy would allow early interventions preventing as many as 10,000 cases of breast cancer annually in the US. The tests would also reduce the “overdiagnosis” of breast cancer that has led to the unnecessary treatment of early lesions that would not have progressed to cancer as well as more targeted therapies for the subset of women who are at greatest risk.
Detection of early lesions provides the greatest opportunity to successfully treat cancer. However, diagnostic tests to accurately identify the lesions that will progress from those that remain latent and selecting effective therapies are needed. The challenge is exemplified by our demonstration of p53 mutations predisposing to mammary tumors in one mouse strain while another strain is nearly completely resistant. This underscores the fact that even a powerful genetic lesion such as p53 mutations can have very different consequences in different patients.
We have sought to utilize clinical specimens to define patterns of gene expression that distinguish early breast lesions (atypical hyperplasias) that subsequently progress to invasive cancer from those that remain indolent. To accomplish this, we have optimized methods for microdissection of lesions from formalin-fixed clinical specimens and amplification of RNA to yield robust gene expression profiles. This project builds on our prior experience in developing gene expression signatures defining the effects of hormonal exposures in breast tissues from mice and women.
We have also collected primary breast epithelial cells from patients to create a panel representing the genetic diversity of the population. The cells can be modified using gene editing to test the effects of oncogenes in different patients. We also utilize explant cultures of breast tissues that maintain the architecture and responses to estrogen and progesterone. These results emphasize the wide range of responsiveness among women and the different effects of ligands binding to the subtypes of estrogen receptors (alpha and beta). Studies in mice have shown that ligands selectively binding to estrogen receptor beta enhance genome surveillance via the p53 pathway. Hormonal regulation of p53 provides opportunities for chemoprevention.
Photos: Parallel studies in mice and human tissues are used to identify conserved pathways that can confer risk and can be harnessed to promoter resistance to breast cancer.