Nasopharyngeal carcinoma with cranial nerve palsy: the importance of MRI for radiotherapy

Purpose: To evaluate various prognostic factors and the impact of imaging modalities on tumor control in patients with nasopharyngeal cancer (NPC) with cranial nerve (CN) palsy. Material and Methods: Between September 1979 and December 2000, 330 NPC patients with CN palsy received radical radiotherapy (RT) by the conventional opposing technique at Chang Gung Memorial Hospital-Linkou. Imaging methods used varied over that period, and included conventional tomography (Tm) for 47 patients, computerized tomography (CT) for 195 patients, and magnetic resonance image (MRI) for 88 patients. Upper CN (II-VI) palsy was found in 268 patients, lower CN (IX-XII) in 13, and 49 patients had both. The most commonly involved CN were V or VI or both (23%, 12%, and 16%, respectively). All patients had good performance status (World Health Organization < 2). The median external RT dose was 70.2 Gy (range, 63-77.5 Gy). Brachytherapy was also given to 156 patients in addition to external RT, delivered by the remote after-loading, high-dose-rate technique. A total of 139 patients received cisplatin-based chemotherapy, in 115 received as neoadjuvant or adjuvant chemotherapy and in 24 concomitant with RT. Recovery from CN palsy occurred in 171 patients during or after radiotherapy. Patients who died without a specific cause identified were regarded as having died with persistent disease. Results: The 3-year, 5-year, and 10-year overall survival was 47.1%, 34.4%, and 22.2%. The 3-year, 5-year, and 10-year disease-specific survival (DSS) rates were 50.4%, 37.8%, and 25.9%. The 5-year DSS for patients staged with MRI, CT, and Tm were 46.9%, 36.7%, and 21.9%, respectively (p = 0.016). The difference between MRI and CT was significant (p = 0.015). The 3-year and 5-year local control rates were 62% and 53%, respectively. The 5-year local control was 68.2% if excluding patients who died without a specific cause. Patients who had an MRI had a significantly better tumor control rate than those evaluated with CT or Tm, with a 15-30% improvement in local tumor control and survival. Patients with extensive CN palsy had worse survival than those with only lower CN or upper CN involvement (5-year DSS 20.4% vs. 43.2% and 40.4%, respectively; p < 0.001). Patients who recovered from CN palsy had better survival than those who did not (47% vs. 26%, p < 0.001). Brachytherapy was associated with poorer local control, whereas a total external dose of more than 70 Gy improved local tumor control and marginally improved DSS. Subgroup analysis in CT and MRI patients group, either DSS or OS was significantly associated with imaging modality, N stage, or location of or remission of CN palsy. Conclusion: The use of MRI was associated with improved tumor control and survival of patients with NPC causing CN palsy. Patients recovering from CN palsy had better survival. A higher radiation dose delivered by external beam may achieve better tumor control than brachytherapy. (c) 2005 Elsevier Inc.