Unknown primary head and neck cancer treated with intensity-modulated radiation therapy: to what extent the volume should be irradiated.

To evaluate the efficacy and feasibility of irradiation with intensity-modulated radiotherapy (IMRT) technique in patients with head and neck carcinoma of unknown primary (HNCUP). Between February 2000 and November 2006, 22 consecutive patients diagnosed with HNCUP were treated with IMRT. Of these, four patients were excluded because two of them underwent IMRT only as a boost and the other two had distant metastasis at presentation and were treated palliatively. Of the remaining 18 patients eligible for analysis, 6 had definitive IMRT and 12 had postoperative IMRT (8 with neck dissection and 4 with excisional biopsy). One patient with abnormal lymph node found unexpectedly during carotid endarterectomy was treated to ipsilateral neck only. The target volumes for the other 17 patients include nasopharynx, oropharynx, retropharyngeal lymph nodes, and bilateral neck. In 1 patient who had extensive lymph node involvement, the hypopharynx and the larynx were defined as target. In all other patients the larynx including hypopharynx were defined as dose-limiting structures. The parotids, cervical esophagus, spinal cord, brain stem, optic nerves and the orbits are also outlined as dose-limiting structures. All patients completed their treatment without interruption and severe complications. With a median follow-up time of 25.5 months, 2 patients had persistent cervical lymphadenopathy and 2 had distant metastases. Another patient developed a lung cancer and a second pyriform sinus cancer 25 months after treatment. The 2-year overall survival, regional recurrence-free survival and distant metastases-free survival were 74.2%, 88.5%, and 88.2%, respectively. Six patients had prophylactic percutaneous endoscopic gastrostomy (PEG) tubes placed before radiation. Most had PEG tubes removed within 7 months after the completion of treatment. IMRT delivered to comprehensive bilateral neck and putative mucosal site (including nasopharynx, oropharynx, and retropharyngeal lymph nodes) appears to be effective for patients with HNCUP. Our preliminary results show that such approach does not compromise local-regional control. Further studies with more patients and longer follow-up are necessary to validate this approach.