Exploring Predictive Effects of Epstein-Barr Virus DNA Levels on Nasopharyngeal Cancer Staging and Relapse

Abstract

Nasopharyngeal carcinoma (NPC) is an Epstein-Barr virus (EBV)-associated malignancy with a distinct geographic distribution, particularly affecting populations in Southeast Asia. Plasma EBV DNA has emerged as a key biomarker for NPC, offering potential applications in disease burden assessment, treatment monitoring, and relapse prediction. However, the transition from EBNA-1 to BamHI-W targeted polymerase chain reaction (PCR) assays in clinical practice raises questions about assay comparability and prognostic significance. This retrospective cohort study examines the correlation between EBV DNA levels and clinical tumor-node-metastasis (TNM) staging and evaluates the predictive value of EBV DNA for relapse in a synthetic dataset, meaning anonymized, artificially constructed data modeled after real patient distributions, of 100 stage II NPC patients from the National Cancer Centre Singapore (NCCS). Patients were stratified based on EBV DNA assay type, EBNA-1 (pre-2016) or BamHI-W (post-2016), and analyzed using statistical methods including Spearman’s correlation, linear regression, Kaplan-Meier survival analysis, and Cox proportional hazards modeling. Results indicated no statistically significant correlation between EBV DNA levels and TNM staging for either assay, suggesting that EBV DNA may not directly reflect tumor burden. Similarly, no significant differences in relapse-free survival were observed between the two assay groups. However, male gender emerged as a significant predictor of relapse (HR = 11.885, p = 0.0369), aligning with prior research on sex-based differences in NPC progression. These findings contribute to the ongoing discussion on EBV DNA as a prognostic biomarker in NPC. While EBV DNA remains clinically valuable, its integration into patient risk stratification should consider demographic and molecular factors beyond assay selection. Future research should explore prospective validation in larger, multi-institutional cohorts and investigate complementary biomarkers to enhance NPC prognostication.