Radiation-induced cranial nerve palsy: a cross-sectional study of nasopharyngeal cancer patients after definitive radiotherapy.

PURPOSE To address the characteristics and the causative factors of radiation-induced cranial nerve palsy (CNP) in nasopharyngeal carcinoma (NPC) patients with an extensive period of followed-up. PATIENTS AND METHODS A total of 317 consecutive and nonselected patients treated with definitive external-beam radiotherapy between November 1962 and February 1995 participated in this study. The median doses to the nasopharynx and upper neck were 71 Gy (range, 55-86 Gy) and 61 Gy (range, 34-72 Gy), respectively. Conventional fractionation was used in 287 patients (90.5%). Forty-five patients (14.2%) received chemotherapy. RESULTS The median follow-up was 11.4 years (range, 5.1-38.0 years). Ninety-eight patients (30.9%) developed CNP, with a median latent period of 7.6 years (range, 0.3-34 years). Patients had a higher rate of CNP (81 cases, 25.5%) in lower-group cranial nerves compared with upper group (44 cases, 13.9%) (χ(2) = 34.444, p < 0.001). Fifty-nine cases experienced CNP in more than one cranial nerve. Twenty-two of 27 cases (68.8%) of intragroup CNP and 11 of 32 cases (40.7%) of intergroup CNP occurred synchronously (χ(2) = 4.661, p = 0.031). The cumulative incidences of CNP were 10.4%, 22.4%, 35.5%, and 44.5% at 5, 10, 15, and 20 years, respectively. Multivariate analyses revealed that CNP at diagnosis, chemotherapy, total radiation dose to the nasopharynx, and upper neck fibrosis were independent risk factors for developing radiation-induced CNP. CONCLUSION Radiation-induced fibrosis may play an important role in radiation-induced CNP. The incidence of CNP after definitive radiotherapy for NPC remains high after long-term follow-up and is dose and fractionation dependent.