Association of Plasma Epstein-Barr Virus DNA With Outcomes for Patients With Recurrent or Metastatic Nasopharyngeal Carcinoma Receiving Anti–Programmed Cell Death 1 Immunotherapy

IMPORTANCE: Anti–programmed cell death 1 (anti–PD-1) immunotherapy features a durable response and improved survival in a small subset of patients with recurrent or metastatic nasopharyngeal carcinoma (RM-NPC). The association between plasma Epstein-Barr virus (EBV) DNA titer dynamics and efficacy of anti–PD-1 monotherapy has been reported, while its value in predicting long-term outcomes and monitoring disease progression is unclear for patients with RM-NPC who are receiving anti–PD-1 monotherapy. OBJECTIVE: To evaluate the role of plasma EBV DNA titers in prognosis prediction and surveillance of disease progression for patients with RM-NPC who are receiving anti–PD-1 monotherapy. DESIGN, SETTING, AND PARTICIPANTS: Patients with RM-NPC from the POLARIS-02 prospective clinical trial, the largest cohort to receive anti–PD-1 monotherapy, were included in this study. From December 22, 2016, to February 19, 2019, 17 participating centers in China screened 279 patients with RM-NPC; 190 patients were enrolled and followed up until February 19, 2020. Plasma EBV DNA was detected before treatment and every 4 weeks until disease progression. MAIN OUTCOMES AND MEASURES: Plasma EBV DNA as a predictor for progression-free survival (PFS), overall survival (OS), durable clinical benefit (defined as PFS of ≥6 months), and disease progression. RESULTS: Of 179 patients with RM-NPC receiving anti–PD-1 therapy, 148 (82.7%) were men, and the median age was 46 years (range, 22-71 years). A higher baseline EBV DNA titer was associated with shorter median OS (hazard ratio, 1.88; 95% CI, 1.22-2.89; P = .004). Patients with a ratio of the EBV DNA titer at week 4 to that at baseline (W4 to baseline ratio) greater than 0.5 had shorter median OS (hazard ratio, 2.18; 95% CI, 1.30-3.65; P < .001) than those with a W4 to baseline ratio of 0.5 or less. Patients with higher baseline EBV DNA titers had a lower durable clinical benefit rate than those with lower baseline EBV DNA titers (19 of 97 [19.6%] vs 27 of 71 [38.0%]; P = .01). Similarly, patients with a W4 to baseline ratio greater than 0.5 had a lower durable clinical benefit rate than those with a W4 to baseline ratio of 0.5 or less (9 of 86 [10.5%] vs 32 of 54 [59.3%]; P < .001). In addition, a significant EBV DNA titer increase was present at a median of 2.6 months (IQR, 0.9-4.5 months) prior to radiographic progression. CONCLUSIONS AND RELEVANCE: This study of plasma EBV DNA in patients with RM-NPC who are receiving anti-PD-1 monotherapy suggests that plasma EBV DNA could be a useful biomarker for outcomes and monitoring disease progression.