Background: The value of Epstein-Barr virus (EBV) DNA assay during posttreatment follow-up of the patients with nasopharyngeal carcinoma (NPC) presenting with different pretreatment plasma EBV DNA levels remains unclear. In the present study, we aimed to evaluate the prognostic value of plasma EBV DNA assay during posttreatment followup in the patients with NPC who have undergone intensity-modulated radiotherapy. Methods: The medical records of 385 NPC patients treated with intensity-modulated radiotherapy between November 2009 and February 2012 were reviewed. All patients underwent plasma EBV DNA assays before treatment, within 3 months after treatment, and then every 3-12 months during posttreatment follow-up period. The recurrence rates for patients with different pretreatment and posttreatment follow-up plasma EBV DNA levels were analyzed. Results: Of the 385 patients, 267 (69.4%) had detectable pretreatment plasma EBV DNA (> 0 copy/mL) and 93 (24.2%) had detectable posttreatment EBV DNA during a median follow-up of 52.8 months (range 9.3-73.8 months). Detectable EBV DNA during posttreatment follow-up was found in 14.4% (17/118) and 28.5% (76/267) of patients with undetectable and detectable pretreatment EBV DNA, respectively, and was significantly associated with tumor recurrence in both patient groups. EBV DNA was detectable in 12.8% (40/313) of patients who remained disease-free, 56.4% (22/39) of patients with locoregional recurrence alone, and 93.9% (31/33) of patients with distant metastasis as the first recurrence event (P < 0.001); 6.5% (19/292) of patients with undetectable EBV DNA and 57.0% (53/93) of patient with detectable EBV DNA during posttreatment follow-up experienced tumor recurrence. Compared with other cut-off values, the cut-off value of 0 copy/mL for EBV DNA during posttreatment follow-up had the highest area under the ROC curve (AUC) value (0.804, 95% confidence interval 0.741-0.868) for predicting tumor recurrence (sensitivity, specificity, and accuracy: 73.6%, 87.2%, and 84.7%, respectively). Conclusion: Plasma EBV DNA level during posttreatment follow-up is a good marker for predicting distant metastasis but not locoregional recurrence in the patients with NPC irrespective of the pretreatment EBV DNA levels.