Purpose: Plasma Epstein-Barr virus DNA (pEBV DNA) is an important prognostic marker in nasopharyngeal carcinoma (NPC). This study tested the hypotheses that pEBV DNA reflects tumor burden and metabolic activity by evaluating its relationship with tumor volume and 18F-fluorodeoxyglucose (18F-FDG) uptake in NPC.
Methods and materials: Pre-treatment pEBV DNA analysis, 18F-FDG positron emission tomography-computed tomography scan (PET-CT) and magnetic resonance imaging (MRI) of the head and neck were performed in 57 patients. Net volume (cm3) of the primary tumor (T(vol)) and regional nodes (N(vol)) were quantified on MRI. 18F-FDG uptake was expressed as the maximum standardized uptake value (SUV(max)) at the primary tumor (T(suv)) and regional nodes (N(suv)). Lesions with SUV(max) > or = 2.5 were considered malignant. Relationship between SUV(max), natural logarithm (log) of pEBV DNA, and square root (sq) of MRI volumes was analyzed using the Wilcoxon test. A linear regression model was constructed to test for any interaction between variables and disease stage.
Results: Log-pEBV DNA showed significant correlation with sq-T(vol) (r = 0.393), sq-N(vol) (r = 0.452), total tumor volume (sq-Total(vol) = T(vol) + N(vol), r = 0.554), T(suv) (r = 0.276), N(suv) (r = 0.434), and total SUV(max) (Total(suv) = T(suv) + N(suv), r = 0.457). Likewise, sq-T(vol) was correlated to T(suv) (r = 0.426), and sq-N(vol) with N(suv) (r = 0.651). Regression analysis showed that only log-pEBV DNA was significantly associated with sq-Total(vol) (p < 0.001; parameter estimate = 8.844; 95% confidence interval = 3.986-13.703), whereas Sq-T(vol) was significantly associated with T(suv) (p = 0.002; parameter estimate = 3.923; 95% confidence interval = 1.498-6.348).
Conclusion: This study supports the hypothesis that cell-free plasma EBV DNA is a marker of tumor burden in EBV-related NPC.