Plasma EBV DNA and cell-free GAPDH DNA as prognostic factors of locally advanced nasopharyngeal carcinoma

Abstract

6053 Background: Plasma cell-free nucleic acids have been investigated to be potential prognostic and predictive markers of treatment response, residual disease, and survival of cancer patients. However, these nucleic acids will be confounded by normal cell death, which might be caused by cancer-related inflammation, infection, or chemotherapy. Nasopharyngeal carcinoma is endemic in Taiwan, highly related to EBV infection. Plasma EBV DNA level has been suggested to be predictive of disease status and outcomes. We try to find out whether plasma EBV and cell-free glyceraldehydes-3-phosphate dehydrogenase DNA before definite treatment of locally advanced NPC predict tumor outcomes better. Methods: 144 stage IV (AJCC version 6) NPC patients received induction chemotherapy followed by concurrent chemoradiotherapy in National Taiwan University Hospital from 1998 to 1999. Pre-treatment blood samples were colleted for real-time quantitative polymerase chain reactions of EBV and GAPDH DNA. The results will be analyzed by SPSS version 13 to see if correlated with tumor extent, locoregional/distant failure, and overall survival. Results: Our patients were mainly composed of T4(97%) and N2(56%) patients. 23 % were in N3 stage. Plasma EBV and cell-free GAPDH DNA levels were not correlated with T stage. Plasma EBV DNA level was significantly correlated with N3 status(Mann-Whitney test p=0.021). After Cox regression model, only age (p=0.024) was a significant predictor of recurrence-free survival; N3 status(p=0.007) and positive plasma EBV DNA (p=0.002) predicted shorter metastasis-free survival. N3 status (p=0.031) and positive plasma EBV DNA (p=0.002) were significantly related to poor survival by log-rank test; whereas T4 lesions (p=0.524) and plasma cell-free GAPDH DNA level (p=0.182) were not. After multivariate Cox regression, positive plasma EBV DNA (hazard ratio 4.936 p=0.006) was the only significant predictor of poor survival. Conclusions: To use circulating DNA as prognostic marker, a specific one, such as EBV DNA, might be better than a non-specific one, like GAPDH DNA. This is important in the development of tumor markers for other solid tumors. No significant financial relationships to disclose.