Radiotherapy with or without mitomycin c in the treatment of locally advanced head and neck cancer: results of the IAEA multicentre randomised trial

Background and purpose: Single agent mitomycin c (MMC) has been shown to improve the outcome of radiotherapy in single institution trials. In order to confirm these findings in a broader worldwide setting, the International Atomic Energy Agency (IAEA) initiated a multicentre trial randomising between radiotherapy alone versus radiotherapy plus MMC. Material and methods: Patients with advanced head and neck cancer were treated with primary curative radiotherapy (66 Gy in 33 fractions with five fractions per week) +/- a single injection (15 mg/m(2)) of MMC at the end of the first week of radiotherapy. Stratification parameters were tumour localization, T-stage, N-stage, and institution. A total of 558 patients were recruited in the trial from February 1996 to December 1999. Insufficient accrual and reporting led to the exclusion of three centres. The final study population consisted of 478 patients from seven centres. Patients had stage III (n = 223) or stage IV (n = 255) squamous cell carcinoma of the oral cavity (n = 230), oropharynx (n = 140), hypopharynx (n = 65) or larynx (n = 43). Prognostic factors like age, gender, site, size, differentiation and stage were well balanced between the two arms. Results: The haematological side effects of MMC were very modest (< 5% grade 3 -4) and did not require any specific interventions. Furthermore, MMC did not enhance the incidence or severity of acute and late radiation side effects. Confluent mucositis and dry skin desquamation was common, occurring in 56% and 62% of patients, respectively. The overall 3-year primary locoregional tumour control, disease-specific and overall survival rates were 19, 36 and 30%, respectively. Gender, haemoglobin drop, tumour site, tumour and nodal stage were significant parameters for loco-regional tumour control. There was no significant effect of MMC on locoregional control or survival, except for the 161 NO patients, where MMC resulted in a better loco-regional control (3-year estimate 16% vs. 29%, P = 0.01). Conclusions: The study did not show any major influence of MMC on loco-regional tumour control, survival or morbidity after primary radiotherapy in stage III-IV head and neck cancer except in NO patients where loco-regional control was significantly improved. (C) 2003 Elsevier Science Ireland Ltd. All rights reserved.