3529 - Circulating Epstein-Barr Virus DNA Level Post Induction Chemotherapy Contributes to Prognostication in Nonmetastatic Nasopharyngeal Carcinoma Treated with Intensity Modulated Radiotherapy (IMRT) + Carbon-Ion Beam Radiotherapy (CIRT)
Presenter(s)
Y. Li1, and C. Hu2,3; 1Department of Radiation Oncology, Shanghai Proton and Heavy Ion Center, Fudan University Cancer Hospital, ShangHai, China, 2Department of Radiation Oncology, Shanghai Proton and Heavy Ion Center, Fudan University Cancer Hospital, Shanghai, China, 3Department of Oncology, Fudan University Shanghai Medical College, Shanghai, China
Purpose/Objective(s): Carbon-ion radiotherapy (CIRT) may further increase the therapeutic ratio for patients with newly diagnosed nasopharyngeal carcinoma (NPC). Plasma Epstein-Barr virus (EBV) DNA is an archetypal biomarker for endemic NPC employed for disease surveillance and early detection of recurrences. However, its accuracy is unknown. The purpose of the current study was to investigate the value of post-induction chemotherapy (IC) cell-free Epstein-Barr virus DNA (cfEBV DNApostIC) for prognostication in patients treated with intensity modulated radiotherapy (IMRT) (57.2 Gy/26F) plus CIRT boost (15 GyE/5F) in a single institution dataset with homogeneous surveillance procedures and routine EBV DNA testing using a harmonized assay.
Materials/Methods: 286 histologically-confirmed NPC patients with EBV DNA test quantified using the harmonized BamHI-W polymerase chain reaction-based assay pre-IC (cfEBV DNApreIC) and at IC completion. We analyzed the low (< 500 copies/mL), and high EBV DNA (= 500 copies/mL) for local recurrence-free survival (LRFS), regional recurrence-free survival (RRFS), distant metastasis-free survival (DMFS), disease-free survival (DFS) and overall survival (OS). Recursive partitioning analysis (RPA) with respect to the OS was applied to construct a risk stratification system incorporating cfEBV DNApostIC and critical risk factors.
Results: We observed that 247 (86.4%) and 39 (13.6%) patients had cfEBV DNApostIC undetectable and detectable respectively. CfEBV DNApostIC positive was associated with a significant inferior 3-year OS (97.2% versus 99.2%), DMFS (85.4% versus 95.2%) and DFS (80.6% versus 92.1%) than cfEBV DNApostIC negative (P < 0.05, respectively). Additionally, cfEBV DNApostIC was independently significant for OS (hazard ratio [HR] 8.43, 95% CI 1.09-64.93), and DFS (2.80, 1.16-6.78) in multivariate analyses (P < 0.05, respectively). RPA modelling base on cfEBV DNApostIC and 9th American Joint Committee on Cancer (AJCC) overall stage yielded three distinct risk groups: low-risk (stage I-II and undetectable cfEBV DNApostIC), median-risk (stage I-II and detectable cfEBV DNApostIC or stage III and undetectable cfEBV DNApostIC) and high-risk (stage III and detectable cfEBV DNApostIC), with 3-year OS of 99.3%, 99.2% and 94.4%, respectively (p < 0.001). The DFS rates in different RPA groups were also distinct (p = 0.022). The RPA model showed better risk discrimination than either the 9th AJCC overall stage or cfEBV DNApostIC alone.
Conclusion: CfEBV DNApostIC represents an effective indicator of prognostication in NPC patients treated with IMRT + CIRT. We developed a risk classification system that provides improved OS prediction over the current 9th AJCC staging system by combining cfEBV DNApostIC and the overall stage.