Abstract P5-04-05: Preclinical modeling of luminal breast cancer: Recapitulating progression to lethal and tamoxifen-resistant lung metastases in novel patient-derived xenotransplant models in prolactin-humanized mice

Abstract

Abstract Seventy to eighty percent of newly diagnosed breast cancer cases are estrogen receptor(ER)-positive and are classified as luminal. Despite hormone therapy, 25-30% of luminal breast cancers will recur within 15 years of surgical removal of the primary tumor, and many of these patients will die from currently incurable distant metastases. Lung and bone are the most common organ sites for distant breast cancer metastases, with either site affected in approximately 70% of patients based on autopsy studies. A major hurdle for therapeutic progress with luminal breast cancer is the historical difficulty to establish xenograft models of human luminal breast cancer in mice. In particular, there is a need for experimental ER-positive breast cancer models that recapitulate distant metastasis formation from orthotopic tumor implants in mammary glands. Such experimental models will allow surgical resection of the primary tumor followed by clinically relevant testing of targeted adjuvant agents against distant residual disease. We have engineered prolactin-humanized mice that have been backcrossed for ten generations into the Nod-SCOD-IL2Rgamma (NSG) immunodeficient background. These prolactin-humanized mice display improved take rates of patient-derived luminal breast cancer. Using prolactin-humanized mice we have established patient-derived serially transplantable luminal breast cancer models that metastasize to distant sites when grown as primary tumors in the mammary gland. One of the ER-positive lines, PDX2, effectively metastasizes to lungs in 100% of animals within 55 days of grafting into mammary glands. PDX2 lung metastases retain ER and progesterone receptor (PR) expression as well as expression of the luminal marker, GATA3, and display high degree of Ki67 positivity indicating rapidly proliferative lesions. Importantly, metastatic PDX2 lesions show only limited dormancy. In fact, when primary PDX2 tumors are surgically removed at Day 55, mice will die from lung metastases around Day 150. Estrogen supplementation is required for establishment of PDX2 tumors in mice. After primary PDX2 tumors are established in mammary glands in the presence of estradiol, primary tumors respond to tamoxifen with growth suppression but do not undergo tumor regression, and develop resistance to tamoxifen. Treatment of mice with adjuvant tamoxifen following surgical removal of primary PDX2 tumors led to extensive regression of existing lung metastases to barely detectable levels within 30 days of surgery. However, tamoxifen-refractory PDX2 lung metastases regrew during the next 30 day-period in the continued exposure to tamoxifen. Molecular phenotyping of the PDX2 model and other new luminal breast cancer models are ongoing, with the goal of characterizing tamoxifen-responsive and tamoxifen-refractory primary and metastatic lesions. Long-term, our intent is to use the PDX2 and other preclinical xenograft models in prolactin-humanized mice to systematically explore agents for synergy with anti-estrogens to establish curative combination treatments for metastatic luminal breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-04-05.