Background: We conducted a multicenter, randomized controlled phase III clinical trial (NCT02143388) to investigate the efficacy and toxicity of adjuvant capecitabine (AC) in addition to concurrent cisplatin and radiotherapy (CCRT) compared to CCRT alone in high-risk locoregionally advanced nasopharyngeal carcinoma (LANPC) patients. Methods: Eligibility criteria included AJCC/UICC 7^th ed TNM stage III-IVb and one of the following features: T3-4N2 or T1-4N3 or pre-treatment plasma EBV DNA concentration of >20,000 copy/ml or gross primary tumor volume (GTVnx) of >30 cm³ or a maximum standard uptake value (SUVmax) of >10.0 by ¹⁸FDG PET-CT within the primary tumor or multiple neck node metastases, with any larger than 4 cm. All patients were randomly assigned in a 1:1 ratio to receive CCRT (3-weekly cisplatin at 100 mg/m² for 2-3 cycles) followed by AC (1000 mg/m² bi-daily for 14 days every 21-day cycle for 8 cycles), or CCRT alone. The prescribed radiation doses were 68-72 Gy/30-32 fractions to the PTVnx, 60-68 Gy/30-32 fractions to PTVnd, 60-64Gy/30-32 fractions to PTV_high-risk, 54-58Gy/30-32 fractions to PTV_low-risk. Primary end point was failure-free survival (FFS). Results: Between Mar 2014 to Jul 2018, 180 patients were recruited (90 patients in CCRT+AC arm and 90 in CCRT alone arm). All patients completed RT and ≥2 cycles of concurrent cisplatin in both treatment arms (cumulative dose intensities for cisplatin were 200 mg/m² in both arms). 85 (94.4%) patients went on to receive AC, with 71 (78.9%) patients completing 8 cycles; 19 (22.4%) patients had dose reduction of AC. With a median follow-up of 44.8 mo, the 3-y FFS was significantly superior in the CCRT+AC arm than the CCRT arm for the intention-to-treat cohort (87.7% vs 73.3%; HR: 0.52 [95% CI: 0.29-0.77], P = 0.037). 3-year overall, distant metastasis-free and locoregional relapse-free survival were 92.6% vs 88.9% (HR [95% CI]: 0.66 [0.28-1.59]), 88.8% vs. 81.1% (HR: 0.67 [0.33-1.33]) and 91.5% vs 80.0% (HR: 0.50 [0.25-1.00]), respectively. Incidences of G3-4 acute toxicities were 57.8% (52 of 90) in CCRT+AC arm and 51.1% (46 of 90) in CCRT alone arm, with a higher incidence of hand foot syndrome (3.5% vs 0%), xerostomia (11.1% vs 3.3%), mucositis (23.3% vs 16.7%), and anemia (5.6% vs 2.2%) in the CCRT+AC arm. G3-4 late toxicities occurred in 13.3% (12 of 90) and 9.0% (8 of 89), respectively. Conclusions: The addition of capecitabine to CCRT conferred a superior disease control than CCRT alone in high-risk LANPC. Survivals in ITT and PP set. Clinical trial information: NCT02143388