Background: : The purpose of this study was to evaluate the prognostic value of magnetic resonance imaging (MRI)-detected cranial nerve (CN) involvement in nasopharyngeal carcinoma (NPC).
Methods: : Retrospective analysis was made of the magnetic resonance images and medical records of 924 consecutive patients with newly diagnosed NPC.
Results: : Of 924 patients, 82 (8.9%) initially presented with CN palsy. CN involvement was seen on MRI in 333 (36%) patients. In T3-4 disease, MRI-evidenced CN involvement was associated with poor 3-year overall survival (OS) (35.7% vs 89.2%, P = .001) and distant metastasis-free survival (DMFS) (77.1% vs 87.8%, P = .002) rates. The survival curves of OS and DMFS for T3 disease with MRI-detected CN involvement approximated those of T4 disease (P = .322 and P = .809, respectively). In patients with MRI-detected CN involvement, no significant differences were observed in 3-year OS (78.3% vs 72.9%, P = .120), local relapse-free survival (LRFS) (89.7% vs 84.1%, P = .154), or DMFS (79.6% vs 74.8%, P = .466) rates between those with and without intracranial or orbital CN involvement. Furthermore, in patients with clinical and/or MRI-detected CN involvement, there were no significant differences in the 3-year OS (74.2% vs 80.1%, P = .067), LRFS (86.7% vs 87.9%, P = .899), or DMFS (74.6% vs 84.6%, P = .094) rates between symptomatic and asymptomatic patients.
Conclusions: : The incidence of MRI-detected CN involvement was higher than CN palsy. MRI-detected CN involvement has a negative impact on the prognosis independent of lesion localization and symptoms. Cancer 2009. (c) 2009 American Cancer Society.