This is a Phase II trial to study the effectiveness of two cycles versus three cycles of Concurrent Chemoradiotherapy in treating patients with Low Risk Locoregionally Advanced Nasopharyngeal Carcinoma based on pretreatment plasma EBV DNA.

Nasopharyngeal carcinoma (NPC) is endemic in Southern China and Southeast Asia. Several prospective randomized trials have demonstrated that concurrent chemoradiotherapy was superior to radiotherapy alone in the treatment of locoregionally advanced NPC. Concurrently, although cisplatin-based concurrent chemoradiotherapy was considered as the standard regimen for locoregionally advanced nasopharyngeal carcinoma, several prospective randomized trials have demonstrated that only 52-63% patients can finished three cycles cisplatin-based concurrent chemoradiotherapy( DDP, 100mg/m2,D1,D22 and D43), due to chemoradiotherapy induced toxicity. Combined analyses of NPC-9901 and NPC-9902 Trials indicated that the 5-year locoregional failure free survival was insignificant between the patients received between two and three cycles cisplatin-based concurrent chemoradiotherapy. Our previous studies have also demonstrated that the five years overall survival, distant metastasis free survival and 5-year locoregional failure free survival was not observed significant difference between the patients received between two and three cycles cisplatin-based concurrent chemotherapy. It indicated that one pressing questions remain to be resolved: can we define the optimal dose so that we can reduce toxicity by avoiding unnecessary overdose or an ineffective phase of treatment? Recently, the quantification of pretreatment plasma Epstein-Barr virus (EBV) DNA, which was considered the most potential biomarker to compliment TNM stage, was demonstrated a useful biomarker for the risk stratification, monitoring and prediction of the prognosis of NPC. The investigators previous study demonstrated that for local and regionally advanced NPC patients with EBV DNA <4,000copies/mL, the 3-year's PFS and DMFS was approximate to 90%. Pretreatment plasma EBV DNA levels might be applied to guide concurrent chemotherapy regimen for local and regionally advanced NPC patients. Therefore, the investigators make a hypothesis that the low risk locoregionally advanced NPC patients received two cycles of chemotherapy may gain similar long-term survival as those received three cycles of chemotherapy, leading to less chemotherapy induced toxicity. Therefore, this phase II non-inferiority, randomised controlled clinical trial was designed to assess efficacy for low-risk patients, identified with pretreatment plasma EBV DNA <4,000 copies/mL, received two cycle cisplatin-based chemotherapy compared with three cycle cisplatin-based chemotherapy

Inclusion Criteria:

• Patients with newly histologically confirmed non-keratinizing nasopharyngeal carcinoma, including WHO II or III
• Original clinical staged as T3-4N0-3 M0 or any T、N2-3M0（according to the 7th AJCC edition）
• No evidence of distant metastasis (M0)
• Pretreatment Plasm EB Virus DNA<4000copies/ml
• Male and no pregnant female
• Satisfactory performance status: ECOG (Eastern Cooperative OncologyGroup) scale 0-1
• WBC ≥ 4×109 /L and PLT ≥4×109 /L and HGB ≥90 g/L
• With normal liver function test (ALT、AST ≤ 2.5×ULN, TBIL≤ 2.0×ULN)
• With normal renal function test (Creatinine ≤ 1.5×ULN)

Exclusion Criteria:

• Patients have evidence of relapse or distant metastasis
• Histologically confirmed keratinizing squamous cell carcinoma (WHO I)
• Receiving radiotherapy or chemotherapy previously
• The presence of uncontrolled life-threatening illness
• Women of child-bearing potential who are pregnant or breastfeeding because of the potentially dangerous effects of the preparative chemotherapy on the fetus or infant.
• Receiving other ways of anti-cancer therapy.
• Suffered from other malignant tumors (except the cure of basal cell carcinoma or uterine cervical carcinoma in situ) previously.