Background and purpose. Intensity modulated radiation therapy (IMRT) offers an opportunity to generate dose distributions highly conformal to the target volume. Head and neck cancer patients, referred for radiotherapy, may be good candidates to benefit from IMRT. This paper discusses the clinical implementation of IMRT for oropharyngeal and oral cavity tumors, and reports the clinical results of the 14 patients treated with this technique at Ghent University Hospital (GUH). Patients and Methods. Between May 1999 and May 2001, 14 patients were treated with IMRT at GUH for oropharyngeal or oral cavity tumors. Two groups of patients can be distinguished. The first group consists of eight patients re-irradiated with IMRT for a locoregional relapse. The second group of six patients were treated with IMRT for a primary tumor. For the first group, IMRT was used to treat the relapse by generating a concave dose distribution, i.e. to combine a homogeneous target re-irradiation with a dose to the spinal cord as low as possible. For the second group, IMRT was applied in order to achieve a more homogeneous dose distribution inside the PTV and to preserve parotid gland function. Results. The majority of the patients of group 1 (6/8) relapsed in field within four months after the end of the re-irradiation, with a median overall survival of 7 months. For group 22, two patients died shortly after the end of the IMRT treatment, the other four patients are free of tumor relapse with a median follow-up of 5 months (1-13 months), The acute toxicity due to radiation was acceptable for both patient groups. Dysphagia and pain was more present in group 1. Regarding late complications for the group of re-irradiations (group 1), no myelitis, carotid rupture or cranial nerve palsy was observed. One patient of group 1 developed osteoradionecrosis of the mandible and feeding tube dependency was present for another patient. No fatal late complications were observed in this group. For the first two patients of group 2, sparing of the parotid function was not a treatment objective. For the other patients of group 2, the mean dose to the contralateral parotid gland ranged from 17 to 25 Gy, which resulted in a decrease of subjective symptoms of xerostomia compared to patients treated with conventional radiotherapy. Conclusions. The implementation of IMRT for oropharyngeal and oral cavity tumors results in a homogeneous target irradiation and allows to re-irradiate locoregional relapses with acceptable adverse effects. Sparing of the parotid gland by IMRT is feasible, although this may be significantly influenced by the delineation method of the elective lymph node regions. (C) 2002 Elsevier Science Ltd. All rights reserved.