Abstract: Objective: To study the MR images of trigeminal nerve involvement in nasopharyngeal carcinoma (NPC)and its impact on prognosis. Methods: A total of 924 consecutive patients with newly diagnosed nonmetastatic NPC whohad undergone MRI scan of nasopharynx and neck were collected. The MR images and the medical records were reviewedretrospectively. Results: MRI-evidenced trigeminal nerve involvement was found in 315 (34.1%) patients, in which thepercentage of patients without trigeminal nerve palsy was 85.4%. In 88 branches of trigeminal nerve palsy, 79 branches(89.8%) were detected with corresponding intracranial or orbital trigeminal nerve involvement. In T 3 disease, MRI-evi-denced trigeminal nerve involvement was associated with poor 3-year overall survival (OS) rates, 80.0% vs. 88.1%(P=0.016). In patients with MRI-evidenced trigeminal nerve involvement, no significant differences were observed in the 3-year OS, distant metastasis-free survival(DMFS) and local relapse-free survival(LRFS)(P=0.672, 0.993 and 0.672, respec-tively) between patients with and without intracranial or orbital trigeminal nerve involvement. Furthermore, no significantdifferences were found in the 3-year OS, DMFS and LRFS (P=0.707, P=0.283 and P=0.542, respectively) between symp-tomatic and asymptomatic patients. Conclusion: The rate of MRI-evidenced trigeminal nerve involvement is high in na-sopharyngeal carcinoma. The main reason for trigeminal nerves palsy was intracranial or orbital trigeminal nerve involve-ment. MRI was precise in diagnosing the location of trigeminal nerve involvement and can be used to determine the prog-nosis of NPC.