Darbepoetin alfa (DA) is a FDA approved long acting erythropoietic protein. We hypothesized that correction of anemia in tumor-bearing mice by DA would secondarily increase the tumor pO2 and potentiate radiation-induced cell killing of tumor cells. To test this hypothesis, we used total body irradiation (TBI) to induce anemia in C3H mice. Murine squamous cell carcinoma tumor (SCC VII) and fibrosarcoma (RIF-1) models were used to study tumor responses to radiation in vivo. DA (30μg/kg) was administered i.p. either every two weeks or weekly. EPO-R RNA levels were measured in tumors from normal, anemic and DA treated mice in both tumor models. Tumors were locally irradiated with daily fractions of 250 cGy for 5 days. Following 500 cGy TBI, hemoglobin levels decreased and reached a nadir of 7.0 ± 0.9 gm/dL 14 days post TBI. Administration of DA reduced the depth and duration of anemia and improved the general health condition of anemic animals as evidenced by accelerated recovery of body weight following the TBI and maintenance of normal levels of activity compared to similarly irradiated animals not treated with DA. Mice treated with DA on the same day as the TBI had elevated hemoglobin levels with a nadir of 11.1 gm/dL on day 14 after TBI. Systemic administration of DA alone did not stimulate tumor growth in TBI-induced anemic mice. When combined with fractionated local tumor irradiation, administration of DA at any of the time points studied (18, 11, 4 and 0 days before initiation of local tumor irradiation) delayed tumor growth and increased the tumor growth delay time from 2.7 days for irradiation alone to 7.3 – 10.6 days for DA treated animals (p < 0.01). There was no statistically significant difference between tumor growth delay times for groups of mice treated with DA at various times before tumor irradiation. Although DA effectively corrected anemia in tumor-bearing mice and significantly decreased the number of hypoxic cells in the tumors as shown by EF5 staining, radiosensitization by DA was independent of the correction of anemia. EPOR RNA expression was barely detectible in tumors cultured in vitro. There were no differences in EPO-R RNA levels in tumors from anemic or DA treated mice (1–2 fold increase), although EPO-R transcription was upregulated in tumors grown in vivo compared to control tumors lines grown in vitro (40–80 fold increase). This may be due to hypoxic induction of EPO-R by tumors in vivo or expression of EPO-R by endothelial cells or infiltrating macrophages. Results from an experiment in non-anemic mice with RIF-1 tumors suggest that DA can sensitize tumor cells in non-anemic mice to radiation as well. These results support the idea that radiosensitization by DA is independent of hemoglobin and tumor pO2. It has long been assumed that anemia causes decreased tumor oxygenation and increased tumor radioresistance, and that correction of anemia would therefore increase tumor pO2, and result in enhanced radiosensitivity. However, the data presented here challenge this presumed relationship. These findings are promising and may have relevance to the treatment of patients with a variety of tumor types with radiation therapy.

Author notes

Corresponding author