TITLE: Breast Cancer Lymphatic Dissemination-Influence of Estrogen and Progesterone PRINCIPAL INVESTIGATOR:

s of Meetings Attended 2006 Gordon Conference: Molecular mechanisms in lymphatic function and disease J. Chuck Harrell, Wendy W. Dye, Carol A. Sartorius, Kathryn B. Horwitz. Department of Medicine and Program in Reproductive Sciences, University of Colorado Health Sciences Center, Aurora, CO, USA 80010 Characterizing a model of estrogen-dependent breast cancer lymphatic metastasis Background: Breast cancer metastasis kills over 400,000 women worldwide each year. The first location breast cancers spread to are usually the lymph nodes. Seventy-five percent of primary breast tumors are estrogen (ER) and/or progesterone receptor (PR) positive and nearly all lymph node metastases retain ER/PR. To date, no models have been employed to study the effects of hormones on cancer cell movement through lymphatic vessels into lymph nodes. Methods: ER+/PR+ human breast cancer cells (MCF-7, T47D) were retrovirally labeled to express the new fluorescent proteins ZsGreen or DsRed-Express. To establish tumors, the abdominal mammary glands of ovariectomized nude mice were injected with one-million fluorescent cancer cells in 100% matrigel. Mice were untreated or supplemented with estradiol for up to 12 weeks. Tumor growth and lymphatic dissemination were monitored weekly with fluorescent whole body imaging, and intravital fluorescence microscopy identified metastases at necropsy. These techniques allowed us to isolate matched trios of tumors, lymphatic vessel emboli, and lymph node metastases. Results: Tumor growth and lymph node spread of both cell lines was estrogen dependent. Seventy percent of estrogen treated tumors yielded metastases that could be monitored in living mice without the use of anesthetics. In the primary tumors, estrogen-induced tumor growth was found to induce intratumoral lymphangiogenesis. ER expression was drastically reduced with estrogen supplementation and this coincided with heterogeneous expression of PR. In estrogenized states the lymphatic emboli expressed tumor-like proportions of ER and PR, whereas the lymph node metastases had slightly increased ER levels and reduced PR expression. BrdU labeling showed proliferation rates that were equivalent in tumors and lymphatic emboli; lymph node metastases however, exhibited a 42% increase in proliferation rate compared to the primary tumor from the same mouse. Discussion: We have developed a model of ER+/PR+ breast cancer lymph node metastasis that allows for study of hormone regulation of this process within the tumor, the lymphatics and lymph nodes. Equivalent ER, PR, and proliferation rates in the primary tumor and lymphatic vessel emboli suggests that ER+ breast cancers need not lose estrogen responsiveness for lymphatic metastasis. These data also suggest that the lymph node microenvironment may contribute to the physiology of metastatic cancers. DOD grant to JCH BC050889. 2007 Keystone Symposia: Host cell interaction and response to the cancer cell Estrogen regulates different sets of genes in primary breast tumors and their lymph node metastases J. Chuck Harrell, Wendy W. Dye, Djuana M. E. Harvell, Carol A. Sartorius, Kathryn B. Horwitz. Department of Medicine and Program in Reproductive Sciences, University of Colorado Health Sciences Center, Aurora, CO, USA 80010 The majority of primary breast tumors and their metastases are estrogen receptor positive. Spread to lymph nodes (LN) indicates advanced disease, often foreshadowing further spread. To understand how hormones impact this process we established a fluorescent xenograft model of estrogen-dependent human breast cancer metastasis. Initial studies indicated increased proliferation rate and decreased estrogen-dependent progesterone receptor expression in lymphatic vessel (LV) and LN metastases compared to their matched primary tumors, suggesting that each microenvironment uniquely influences cancer aggressiveness. To determine genomic targets that control enhanced aggressiveness in LNs, matched pairs of primary tumors and their LN metastases from estradiol treated nude mice were removed and frozen. Pure tumor cells were isolated from serial sections by laser capture microdissection and processed for microarray analyses. Approximately 60% of transcripts were co-expressed in the two microenvironments. However, subsets of genes were uniquely expressed in tumors or their LN metastases. Among transcripts upregulated in the LNs were ones encoding CD44. This was confirmed by immunohistochemical analyses from matched tumors, LV emboli, and LN metastases, which showed that CD44 protein was upregulated in LV and LN metastases. We hypothesize that CD44 selects primary tumor cells for lymphatic transit. To determine if gene regulation by estrogen is altered in the LN microenvironment compared to the primary tumor, we also determined how the gene expression profiles in each compartment changed when estradiol was removed. Interestingly, different subsets of genes were estrogen regulated in the LN compared to the primary tumor. These results suggest that therapeutics targeting ER+ breast cancers differentially affect primary tumors and their metastases. DOD grant to JCH BC050889. Publications from the past 12 months Estrogen Receptor Positive Breast Cancer Metastasis: Altered Hormonal Sensitivity and Tumor Aggressiveness in Lymphatic Vessels and Lymph Nodes Joshua Chuck Harrell, 1,3 Wendy W. Dye, 1 D. Craig Allred, 4 Paul Jedlicka, 2 Nicole S. Spoelstra, 1 Carol A. Sartorius, 1 and Kathryn B. Horwitz 1,2,3 Departments of Medicine, Pathology, and Program in Reproductive Sciences, University of Colorado Health Sciences Center, Aurora, Colorado; and Breast Center, Baylor College of Medicine, Houston, Texas