The Expression of Erythropoietin and Its Receptor in Breast Cancer Is Associated with In Vivo Tumor Hypoxia.

Erythropoietin (EPO) is primarily produced in the adult kidney where hypoxia is the major stimulus for EPO expression under the control of an oxygen-sensing mechanism. A series of previous studies in our laboratories and others have demonstrated the expression of EPO and its receptor (EPOR) in breast cancer cells. In previous work, we found that the administration of antagonists of EPO-EPOR signaling was associated with delayed tumor growth in a rodent syngeneic breast cancer model, suggesting the presence of a functional EPO-EPOR system in cancer. The mechanisms of EPO and EPOR expression in tumor cells requires further study. In vitro experiments using monolayer cultures of breast cancer cells have suggested that EPO and EPOR expression in tumor cells may be hypoxia-regulated. The objective of our current studies was to determine the relationship between the expression of EPO, EPOR and tumor hypoxia in breast cancer to test the hypothesis that EPO or EPOR expression in malignant cells may be associated with the presence of in vivo tumor hypoxia, an important factor involved in resistance to radiation treatment, tumor aggressiveness and poor prognosis. Thirty-eight patients with primary breast cancer were enrolled in a tumor hypoxia study under a research protocol approved by the Institutional Review Board at the University of North Carolina Hospitals. All patients provided signed informed consent. The patients received an intravenous infusion of the hypoxia marker pimonidazole hydrochloride (Hypoxyprobe-1™) prior to tumor biopsy. Two or more biopsies were obtained from 33 tumors and 5 primary tumors had a single biopsy available. Contiguous sections from a total of 93 biopsies were analyzed by immunohistochemistry using monoclonal antibodies for the expression of EPO and EPOR and pimonidazole binding. We found EPO expression in tumor cells in 94% of the biopsies. Focal EPO expression pattern was observed in many tumors. EPOR expression was present in 93% of biopsies. The pattern of EPOR immunoreactivity was predominantly cytoplasmic but was found to be localized to the membrane in some sections. In many tumors, co-localization of EPO and EPOR expression in tumor cells was present when contiguous sections were examined. Tumor hypoxia was detected in 83% of the biopsies at variable levels and did not always co-localize with EPO or EPOR expression in tumor cells. Semi-quantitative analyses for EPO immunostaining and tumor hypoxia on a section-by-section basis revealed a significant positive correlation between levels of micro-regional EPO expression and pimonidazole binding (r = 0.6, P < 0.0001, n=93 by two-tailed Spearman’s rank correlation analysis). A similar significant positive correlation was found between levels of EPOR expression and pimonidazole binding (r = 0.63, P < 0.0001, n=93). In addition, there was a significant association between the semi-quantitative EPO score in tumor cells and the EPOR score (r = 0.6, P < 0.0001, n=93). We also determined the micro-vessel density (MVD), a marker of tumor angiogenesis, in 35 biopsies using factor VIII immunostaining. There was no correlation between EPO or EPOR expression and MVD in these samples. In conclusion, these data demonstrate for the first time that EPO and EPOR expression in breast cancer cells correlates with in vivo tumor hypoxia in clinical specimens of primary breast cancer.