To evaluate if correction of low hemoglobin (Hb) levels by means of darbepoetin alfa improves the outcomes of radiotherapy in patients with squamous cell carcinoma of the head and neck (HNSCC). Patients eligible for primary radiotherapy and who had Hb values below 14.0 g/dl were randomized to receive accelerated fractionated radiotherapy with or without darbepoetin alfa. Patients also received the hypoxic radiosensitizer nimorazole. Darbepoetin alfa was given weekly during radiotherapy or until the Hb value exceeded 15.5 g/dl. Following a planned interim analysis which showed inferiority of the experimental treatment the trial was stopped after inclusion of 522 patients (of a planned intake of 600). Of these, 513 were eligible for analysis (254 patients treated with darbepoetin alfa and 259 patients in the control group). Overall, the patients were distributed according to the stratification parameters (gender, T and N staging, tumor site). Treatment with darbepoetin alfa increased the Hb level to the planned value in 81% of the patients. The compliance was good without excess serious adverse events. The results showed a poorer outcome with a 5-year cumulative loco-regional failure rate of 47% vs. 34%, Hazard Ratio (HR): 1.53 1.16–2.02, for the darbepoetin alfa vs. control arm, respectively. This was also seen for the endpoints of event-free survival (HR: 1.36 1.09–1.69), disease-specific death (HR: 1.43 1.08–1.90), and overall survival (HR: 1.30 1.02–1.64). There was no enhanced risk of cardio-vascular events observed in the experimental arm or any significant differences in acute or late radiation related morbidity. All univariate analyses were confirmed in a multivariate setting. Correction of the Hb level with darbepoetin alfa during radiotherapy of patients with HNSCC resulted in a significantly poorer tumor control and survival.