Dynamic changes in plasma Epstein-Barr virus DNA load during treatment have prognostic value in nasopharyngeal carcinoma: a retrospective study

Abstract

Circulating plasma Epstein-Barr virus DNA (EBV DNA) is related to tumor recurrence and metastasis and has potential as a dynamic, sensitive, and specific marker in nasopharyngeal carcinoma (NPC). We investigated the clinical significance of assessing plasma EBV DNA load at various time points during treatment. Patients with NPC (n = 949) for whom plasma EBV DNA load was measured by real-time quantitative polymerase chain reaction (RT-qPCR) before treatment (pre-EBV) and at midtreatment (mid-EBV), end of treatment (end-EBV), and 3 months after completing treatment (3 m-EBV) were retrospectively assessed. Receiver operating characteristic (ROC) curve analysis was used to identify the optimal EBV DNA cutoff point for each time point. Overall survival (OS), distant metastasis-free survival (DMFS), and progression-free survival (PFS) were compared using Kaplan-Meier estimates. High pre-EBV, high mid-EBV, high end-EBV, and high 3 m-EBV were all associated with significantly poorer OS, DMFS, and PFS in the entire cohort. Detectable end-EBV and 3 m-EBV was associated with significantly poorer OS, DMFS, and PFS. Among patients with detectable end-EBV, adjuvant therapy significantly improved OS (HR 2.419; 95% CI 1.297-4.51, P = 0.03) and DMFS (HR 2.45; 95% CI 1.243-4.828, P = 0.04), but not PFS (P = 0.17). EBV DNA represents a dynamic biomarker for monitoring treatment and predicting survival in NPC. Assessing plasma EBV DNA before, during, and after chemoradiotherapy could be clinically valuable and enable selection of patients most likely to benefit from additional therapy and improve assessment of treatment response and disease surveillance. Further multicenter prospective investigations are warranted.

Keywords: Epstein-Barr virus DNA; nasopharyngeal carcinoma; survival; treatment.

Figure 1

Effect of AJCC classification and stage on overall survival. Kaplan–Meier overall survival curves for all 154 patients with NPC stratified by T category (A), N category (B), and overall stage (C).

Figure 2

Effect of plasma EBV DNA at different treatment time points on survival outcomes. Kaplan–Meier overall survival, distant metastasis‐free survival, and progression‐free survival curves for all 154 patients with NPC stratified by pre‐EBV (<2500 vs. ≥2500 copies/mL; A, B, C), mid‐EBV (<871 vs. ≥871 copies/mL; D, E, F), end‐EBV (<721 vs. ≥721 copies/mL; G, H, I), and 3 m‐EBV (<211 vs. ≥211 copies/mL; J, K, L).

Figure 3

Detectable end‐EBV and detectable 3 m‐EBV are associated with poor survival outcomes. Kaplan–Meier overall survival (A), distant metastasis‐free survival (B), and progression‐free survival (C) curves for all 154 patients with NPC stratified as Group 1 (undetectable end‐EBV and undetectable 3 m‐EBV; detectable end‐EBV and undetectable 3 m‐EBV; undetectable end‐EBV and detectable 3 m‐EBV) and Group 2 (detectable end‐EBV and detectable 3 m‐EBV).

Figure 4

Effect of adjuvant therapy on survival outcomes in patients with detectable end‐EBV. Kaplan–Meier overall survival (A) and distant metastasis‐free survival (B) curves for the 196 patients with detectable end‐EBV stratified by with or without adjuvant therapy.