Darbepoetin worsens 5-year survival in head and neck cancer

The erythropoiesis-stimulating agent darbepoetin alfa should not be used in combination with radiotherapy in patients with potentially curable head and neck cancer, late results of the halted Danish Head and Neck Cancer Study Group 10 trial indicate.
Although darbepoetin alfa (Aranesp), given at a weekly subcutaneous dose of 150 mcg, successfully raised low hemoglobin (Hb) levels, its use with accelerated radiotherapy was associated with significantly worse clinical outcomes, including locoregional tumor control and overall survival at 5 years, than was radiotherapy alone.

	Dr Jens Overgaard

This is in agreement with other similar trials, said study investigator Dr. Jens Overgaard, professor and head of the department of experimental clinical oncology at Aarhus (Denmark) University Hospital. “What is interesting is that all these trials began with the belief that at least we will do no harm – we might do good – and then it turned out that it is the opposite: You actually are doing harm,” Dr. Overgaard said in an interview, Sept. 16 at the biennial meeting of the European Society of Therapeutic Radiation and Oncology, where he presented the new data.
The DAHANCA 10 trial started in June 2002 to test the hypothesis that raising low Hb (defined as less than or equal to 14.5 g/dL) with darbepoetin during curative radiotherapy for squamous cell carcinoma of the head and neck would improve patient outcomes vs. radiotherapy alone. However, the study was stopped early in November 2006, when 522 of the originally anticipated 600 patients had been randomized, because of worse outcomes in the Aranesp-treated patients.
In all, data on 254 evaluable patients who were treated with darbepoetin plus radiotherapy (66-68 Gy given in 33-34 fractions, at 6 fractions per week) and 259 evaluable patients who were given radiotherapy alone were available for the 5-year follow-up analysis. Results showed that the addition of darbepoetin was associated with significantly worse rates of locoregional tumor control (50% vs. 63%; P = .003), death from cancer (53% vs. 65%; P = .02), disease-free survival (33% vs. 46%; P =.003), and overall survival (40% vs. 51%; P = .03).
There were no statistically significant differences in tumor characteristics between the two treatment arms that could perhaps help explain the findings, nor was there any difference in the rate of deaths that were not from cancer.
Hazard ratios for locoregional failure, overall death, and disease-specific death were 1.54, 1.29, and 1.41, respectively, for darbepoetin vs. no ESA use. Higher tumor grade, nodal involvement, male sex, and World Health Organization performance status were also significant predictors of worse outcome.
“The cause of the poorer outcome in the Aranesp group is not quite obvious, but of course it is tempting to guess that it is a potential tumor- or growth-stimulating effect of the agent,” which is currently being further explored, Dr. Overgaard said during his presentation.
He concluded, “The idea of using ESAs in this patient cohort is not something one should continue to do.”
Dr. Camilla Hoff, also of the department of experimental clinical oncology at Aarhus University Hospital, presented findings from the DAHANCA 5 trial. The 414-patient study demonstrated that raising Hb via blood transfusion around the time of radiotherapy also was unable to significantly improve patients’ outcomes vs. no transfusion.
“We know that with a lower hemoglobin level patients do worse, but the point is that you cannot just correct it,” Dr. Overgaard observed.
The Danish Cancer Society supported the DAHANCA 5 and DAHANCA 10 trials. Amgen provided additional support for the DAHANCA 10 trial. Dr. Overgaard and Dr. Hoff had no personal financial disclosures.